Il y a des phrases qui restent longtemps en tête. Elles s’installent discrètement et finissent par faire partie du décor de notre pensée. Pour moi, c’est une citation d’André Gide qui tient en huit mots : « Le sage est celui qui s’étonne de tout ». Dans ma vie professionnelle, ou ici sur Cachem, c’est la curiosité et le partage qui m’animent le plus. Ce plaisir simple de chercher, de comprendre et de transmettre. Mais si je suis totalement honnête avec moi-même et avec vous, il existe un troisième moteur plus discret… peut-être moins avouable et mal compris : le doute.

Douter, vraiment ?

Dans beaucoup de milieux (surtout professionnels), le doute est perçu comme une faiblesse. Quelqu’un qui doute manquerait de conviction, d’assurance ou de leadership. On préfère souvent ceux qui tranchent vite, affichent des certitudes et avancent sans jamais se retourner. Je ne suis pas d’accord…

Pour moi, douter n’est pas une faille : c’est une force. C’est pour moi un moteur de réflexion, ne pas s’endormir sur ses acquis, remettre en question ce que l’on croit savoir et laisser de la place à ce que l’on n’a pas encore compris.

Ce que Gide appelle l’étonnement, c’est exactement ça. Non pas la naïveté de celui qui ignore tout, mais la lucidité de celui qui continue à regarder le monde avec curiosité, même après avoir beaucoup appris. Comment peut-on croire qu’on a déjà fait le tour d’un sujet ? Douter ne signifie pas rester immobile : on peut questionner ses certitudes, prendre du recul puis décider avec davantage de lucidité et de discernement.

Cachem.fr : pourquoi je continue (et pourquoi parfois je doute)

Le doute ouvre la réflexion

Dans mon quotidien, le doute joue un rôle concret. Il me pousse à vérifier, à croiser les sources, à envisager d’autres points de vue. Là où la certitude ferme les portes, le doute les laisse entrouvertes. Oui, cela peut ralentir parfois, mais il approfondit presque toujours.

C’est aussi, je crois, ce qui rend le partage plus sincère. Quand j’écris un article ou que j’échange autour d’un sujet, je ne prétends pas détenir une vérité absolue. Je partage ce que j’ai compris, ce que j’ai trouvé pertinent… tout en laissant une place aux questions et à la remise en perspective. C’est notamment pour cela que les commentaires sont toujours ouverts sur Cachem.

Les commentaires font vivre un blog… et comment en tirer le meilleur

Cette honnêteté intellectuelle, je la dois en grande partie au doute.

Prendre du recul n’est pas reculer

La phrase de Gide m’a aussi appris quelque chose de plus difficile à accepter : prendre du recul n’est pas une forme de capitulation. Ce n’est pas admettre qu’on a tort. C’est se donner une chance supplémentaire de mieux comprendre.

Le monde actuel valorise l’instantanéité, la réaction immédiate, l’opinion formulée en quelques secondes, la certitude affichée en permanence. Savoir s’arrêter pour observer, réfléchir et douter relève presque d’un acte de résistance. Bon OK, je vais peut-être un peu loin…

Rester humble face à la complexité des choses, c’est la posture que j’ai choisi…

Alors, si vous me lisez depuis quelque temps sur le blog, vous avez probablement remarqué une chose : je ne prétends pas tout savoir. Il m’arrive de changer d’avis. Je remets souvent en question mes propres choix. Non, ce n’est pas de l’inconstance ou de la fantaisie… mais le doute et la curiosité continue de faire leur travail.

Finalement, le doute guide souvent bien mieux que les certitudes. Et vous, est-ce que vous doutez ? N’hésitez pas laisser un commentaire…

Pendant des années, le NAS était un boitier discret rangé dans un placard pour de la sauvegarde et du partage de fichiers. En 2026, certains modèles embarquent des processeurs surpuissants, un emplacement PCIe pour GPU et peuvent faire tourner des LLM en local. Mais est-ce vraiment une bonne idée ?

Quand le NAS a arrêté d’être simplement du stockage

Il y a 10 ans, un NAS était un appareil relativement humble : quelques disques durs en RAID, une interface Web basique et la satisfaction de savoir que vos photos de famille ne disparaîtraient pas du jour au lendemain. C’était son rôle et il le tenait bien…

Bien sûr, il était possible de lui ajouter quelques fonctionnalités supplémentaires : antivirus, serveur d’impression, station de téléchargement, serveur Web, etc. Puis Docker est arrivé.

Synology, QNAP, Asustor, TerraMaster ont progressivement intégré la gestion de conteneurs dans leurs interfaces. Et là, tout a basculé… Les forums ont explosé en tutoriel pour faire tourner AdGuard Home (bloqueur de pub), Jellyfin (votre Netflix personnel), Immich (votre Google Photos à vous), Bitwarden (vos mots de passe en local) ou encore Home Assistant (votre domotique sous contrôle total). Soudain, le NAS n’était plus un périphérique de stockage, c’était un vrai serveur.

Docker sur NAS : la révolution applicative

Comprendre pourquoi Docker a tout changé, c’est comprendre ce qu’il apporte : l’isolation. Chaque application fonctionne dans son propre conteneur, avec ses dépendances, sans polluer le système hôte. Pour un NAS qui doit avant tout rester stable, c’est idéal… enfin, en théorie.

En pratique, chaque application consomme de la mémoire vive (RAM), du temps processeur (CPU) et de l’espace disque. Là où les NAS d’entrée de gamme (ou les plus anciens) embarquaient des processeurs ARM modestes avec 512 Mo de RAM, les usages actuels exigent bien davantage.

Le processeur du NAS : de parent pauvre à pièce maîtresse

Historiquement, le dimensionnement du CPU d’un NAS était une préoccupation secondaire à la conception… quelque chose d’assez puissant pour gérer les entrées/sorties disques et le chiffrement, mais pas davantage. Cette époque est révolue !

Intel N100 et ses cousins

La génération de NAS sortie entre 2023 et 2025 s’est largement standardisée autour de processeurs comme l’Intel Celeron J4125 et N95. Ces puces offrent un excellent équilibre : faible consommation, transcodage matériel et performances suffisantes pour faire tourner confortablement plusieurs applications simultanément.

Un NAS équipé d’un N305 avec 16 Go de RAM représente aujourd’hui la configuration idéale pour l’utilisateur qui souhaite un serveur applicatif polyvalent sans alourdir sa facture d’électricité. C’est d’ailleurs le consensus des communautés homelab : pour 80 % des usages domestiques, ce profil CPU est amplement suffisant.

2026 : la montée en puissance s’accélère

Les constructeurs, eux, ne semblent pas vouloir s’arrêter là. L’année 2026 marque un tournant sur le marché des NAS haut de gamme, avec des annonces qui auraient semblé absurdes il y a 3 ans.

Plusieurs modèles intègrent désormais des processeurs haute performance avec des NPU (Neural Processing Unit) intégrés. Plus fort encore, certains proposent d’un emplacement PCIe pour des cartes GPU Nvidia (RTX 4060 ou carte d’inférence A2).

Attention, ce type de configuration consomme entre 35 et 65W en charge (contre 8-15W pour un N100). Sur une année de fonctionnement continu, la différence représente plusieurs dizaines d’euros sur votre facture d’électricité. La montée en puissance doit donc se justifier par des besoins réels.

L’IA locale sur NAS : révolution ou effet d’annonce ?

L’intelligence artificielle locale (faire tourner des LLM comme Llama, Mistral ou Phi-4 sur son propre matériel) est devenue le nouveau Graal du homelab. Des outils comme Ollama ou LM Studio permettent désormais à n’importe qui d’héberger son propre assistant IA privé, sans envoyer la moindre donnée dans le Cloud.

Les NAS de 2026 mettent cet argument en avant comme un atout commercial majeur. Le Ryzen AI 9 HX Pro 370, avec ses 50 TOPS de puissance NPU, peut faire tourner des modèles 7B (7 milliards de paramètres) à une vitesse tout à fait acceptable. Avec une carte GPU Nvidia en PCIe, on passe à un niveau différent : des modèles 13B ou 30B deviennent envisageables.

Mais soyons honnête, un NAS est-il vraiment la bonne machine pour faire de l’IA ?

Arguments pour l’IA sur NAS :

• Machine déjà allumée 24h/24, pas besoin de PC supplémentaire ;
• Intégration directe avec les données stockées localement ;
• Un seul équipement à administrer ;
• Les nouveaux modèles compacts (Phi-4, Gemma 3) tournent efficacement sur CPU/NPU ;
• Aucun envoi de données vers le cloud, la confidentialité préservée

Limites à considérer :

• Un GPU dédié (même RTX 3060) reste 5-10× plus rapide pour l’inférence ;
• Coût important : NAS + GPU > PC dédié ;
• Thermique : un NAS est conçu pour les disques et SSD, pas pour un GPU chaud ;
• Risque de concurrence pour les ressources avec les conteneurs Docker ;
• Maintenance plus complexe en cas de panne du GPU.

L’IA locale sur un NAS est une option crédible pour des usages légers comme un chatbot personnel interrogeant vos documents, de la transcription audio locale ou de l’analyse d’images simples. Pour de l’inférence intensive ou de la génération d’images (Stable Diffusion), un PC dédié avec GPU reste de loin la solution la plus efficiente.

Faut-il tout mettre dans son NAS ?

Voilà la vraie question de fond, celle que tout passionné de homelab finit par se poser. Et honnêtement, il n’y a pas de réponse universelle… mais 2 logiques s’affrontent clairement.

NAS tout-en-un

Un seul appareil à gérer, une seule prise électrique, une seule interface d’administration. Pour l’utilisateur qui débute en homelab ou qui veut une solution simple et économique, un NAS bien dimensionné (N305/16Go + quelques disques) fait très bien le travail : stockage, partage, sauvegardes, Jellyfin, Immich, Bitwarden, Home Assistant… tout cela tourne parfaitement sur ce profil matériel, sans dépenser 500€ supplémentaires en serveur séparé.

Dissociation

À mesure que les besoins s’étoffent (plus d’applications, des machines virtuelles, de la virtualisation réseau, des charges IA…), la logique évolue. Un NAS reste fondamentalement un appareil de stockage : son système de refroidissement, sa durée de vie, sa conception sont optimisés pour des disques durs en fonctionnement permanent, pas pour un CPU sollicité à 95 % de charge pendant 8 heures.

La séparation entre stockage et applicatifs présente un avantage opérationnel réel : si votre serveur Docker tombe, vos données sur le NAS restent intactes et accessibles. Si votre NAS rencontre un problème de disque, vos services continuent de fonctionner. Cette résilience a de la valeur, surtout si vous hébergez des services critiques (ex. : Bitwarden).

Mon humble avis

Un NAS équipé d’un processeur de type N150/N305 gère parfaitement les applicatifs courants du homelab, sans sourciller. C’est un équilibre excellent. En revanche, l’IA locale intensive mérite une machine dédiée : un PC avec GPU sera toujours plus performant à budget et consommation comparables.

Les NAS ultra-puissants de 2026 sont impressionnants sur le papier, mais leur proposition de valeur doit se confronter à la réalité des usages réels (et à votre facture d’électricité).

L’IA sur un NAS peut avoir un intérêt, c’est un indéniable… mais pour des usages ciblés. Les processeurs continueront de progresser tout en maîtrisant leur consommation énergétique. Ce qui n’est pas encore possible aujourd’hui le sera certainement demain.

Et l’avenir ?

L’arrivée de l’IA dans les NAS grand public est réelle et irréversible. Les cas d’usage vont se multiplier : reconnaissance d’objets dans vos photos (Immich le fait déjà), transcription automatique de réunions, assistants contextuels connaissant vos fichiers… Ces fonctions légères, intégrées nativement par les éditeurs, s’accommoderont très bien d’un CPU / NPU, même modeste.

La question n’est donc pas de savoir si l’IA a sa place sur un NAS (elle y est déjà), mais à quelle profondeur vous voulez l’intégrer… et si le jeu en vaut financièrement la chandelle. La mode des NAS « IA » de 2026 ressemble un peu à celle des NAS « 4K transcoding » de 2018 : une vraie capacité, que 10 % des utilisateurs exploiteront réellement à plein régime.

En synthèse

Le NAS a profondément évolué. De simple disque réseau, il est devenu un vrai serveur domestique, capable de faire tourner une constellation d’applications via Docker. Un processeur Intel N305 avec 16 Go de RAM est aujourd’hui la configuration raisonnable pour un homelab polyvalent et économe en énergie.

Les NAS de 2026 avec leurs Ryzen AI, leurs processeurs mobiles Core i7 mobiles et GPU sont de vraies prouesses techniques. Ils ouvrent des possibilités inédites (IA locale, inférence LLM, transcodage massif…). Mais cette puissance a un coût : financier, énergétique et en complexité de maintenance.

Pour la majorité des utilisateurs, un NAS bien dimensionné couvre largement les besoins applicatifs courants. L’IA locale intensive mérite une machine dédiée. Et pour les besoins les plus exigeants, l’architecture dissociée NAS + serveur applicatif reste la solution la plus robuste et la plus évolutive.

L’avenir du NAS comme serveur domestique universel est probable. Mais en 2026, nous n’en sommes qu’aux premiers chapitres… et les usages réels restent encore loin des promesses marketing.

What is the Gl.iNet Comet Q KVM?

The Gl.iNet Comet Q is a compact KVM-over-IP device built around a different kind of deployment than most existing entries in this category. Instead of focusing on HDMI-connected desktops, servers, or rack hardware, the Comet Q is designed around a direct USB-C connection, allowing it to interface with smartphones, tablets, laptops, and other compatible host devices through a single pre-attached cable. Alongside local access, it also integrates WiFi-based networking, remote internet control, a built-in touchscreen, and USB-C pass-through for power delivery to the connected device. Based on the early demonstration shown during a visit to Gl.iNet in Shenzhen, the Comet Q appears to be aimed at portable remote access, field support, and off-site troubleshooting, while also expanding the wider Comet KVM range into a more mobile and lower-power form factor.

Interested in Gl.iNet KVM Devices? Here are some great options available NOW: Gl.iNet Comet Pro $179 – HERE on Amazon Gl.iNet Comet 5G $299 – HERE on Amazon Gl.iNet Comet PoE $129 – HERE on Amazon

Gl.iNet Comet Q Hardware Specifications

The Gl.iNet Comet Q is built around a notably smaller hardware platform than the rest of the Comet KVM family. According to the specification sheet provided, it uses a dual-core ARM Cortex-A53 processor, paired with 512MB of LPDDR4 memory and 64GB of onboard storage. This places it below the Comet, Comet PoE, Comet Pro, and Comet 5G in raw system resources, but that appears consistent with its intended role as a highly compact USB-C based access device rather than a more traditional full-size KVM endpoint.

In terms of connectivity, the Comet Q differs significantly from the rest of the range. Rather than relying on HDMI input, it uses a USB-C connection with DisplayPort Alt Mode support for video input. This is the key functional distinction in the lineup, as it allows the device to connect directly to supported modern phones, tablets, and laptops without requiring a separate HDMI capture path. The copied specifications also indicate USB 2.0 Type-A and Type-C connectivity, alongside 1 x RJ45 Gigabit Ethernet port.

Wireless support is also listed as part of the Comet Q feature set. The specification sheet references 802.11a/b/g/n/ac/ax support, with 2.4GHz and 5GHz operation included across the lineup. Although the pasted table is clearly the result of OCR and contains some formatting inconsistencies, the Comet Q is positioned as a wireless-enabled KVM device rather than a purely wired one, which aligns with the functionality shown in the demonstration. This is important because the device is intended to support both local network access and wider remote access scenarios.

Power and physical design are clearly central to the Comet Q hardware profile. It is rated for Type-C power input at 5V/2A, with listed power consumption of less than 2.5W, making it the lowest-power device in the copied Comet family specifications. It also includes a 1.8-inch touchscreen, which is smaller than the displays used on some of the larger Comet models, but appropriate for quick status checks, local configuration, and access control on a device intended for portable use.

Environmental and physical figures place the Comet Q firmly in the compact end of the lineup. The operating temperature is listed as 0°C to 40°C, consistent with the rest of the family. The OCR copy of the table does not clearly preserve the final dimensions and weight entry for the Comet Q in the same way as the other models, but the wider specification set still makes clear that this is intended to be a lighter, lower-power, more travel-friendly device than the HDMI-based Comet units already in the range.

Specification	Gl.iNet Comet Q
Model	GL-RMQ1
CPU	Dual-core ARM Cortex-A53
Memory	512MB LPDDR4
Storage	64GB
Wireless Protocol	802.11a/b/g/n/ac/ax
Wireless Bands	2.4GHz, 5GHz
Ethernet Port	1 x RJ45 10/100/1000Mbps
USB Ports	USB 2.0 Type-A, USB 2.0 Type-C
Power Input	Type-C (5V/2A)
Power Consumption	<2.5W
Screen	1.8-inch touchscreen
Video Input	1 x USB-C (DP Alt supported)
Operating Temperature	0°C to 40°C
Notes	USB-C based KVM design intended for compatible mobile and computing devices

 

What is the Comet Q KVM bringing to the market that is new?

The main distinction of the Gl.iNet Comet Q is its physical design and target use case. Most KVM-over-IP devices are built around HDMI capture and are designed for desktops, servers, mini PCs, or rack-mounted hardware. The Comet Q instead shifts the concept toward a much smaller USB-C based form factor, with a pre-attached cable and integrated display in a body that is intended to be carried and deployed quickly. That makes it structurally different from the more static, cabling-heavy approach seen in much of the current KVM market.

Portability is another clear differentiator. The Comet Q is designed to operate from USB-C power at under 2.5W, which creates a very different deployment model from larger KVM appliances that often assume fixed placement, dedicated power, and a more permanent network setup. In practical terms, this makes the device easier to use in travel scenarios, temporary support jobs, meeting environments, mobile workstations, and short-term remote access situations where carrying a larger HDMI-based KVM would be less practical.

Its support for USB-C connected client devices also broadens the type of hardware that can be managed. The Comet Q is positioned not only for laptops and compact computers, but also for phones and tablets that support the necessary USB-C display and data standards. That gives it a role that is uncommon in the KVM-over-IP space, where Android phones, tablets, and similarly compact devices are not usually the primary focus. In that respect, the Comet Q is not just reducing size, but also changing the class of device a KVM can be attached to.

The single-cable approach is also important. Based on the demonstration and the listed hardware details, the Comet Q is intended to combine connection, control, and power handling through USB-C, while also supporting network access over LAN, WiFi, and remote internet connectivity. That creates a simpler deployment path than a conventional KVM setup that may require separate video, USB, power, and networking connections. The result is a product that appears to reduce setup complexity while extending KVM access to devices and environments that are not well served by existing HDMI-first designs.

How Does the Comet Q Compare with the Rest of the Gl.iNet KVM Lineup?

Within the wider Gl.iNet Comet series, the Comet Q sits as the most specialised and least traditional model in the range. The RM1 Comet, RM1PE Comet PoE, RM10 Comet Pro, and RM10RC Comet 5G are all built around a more conventional KVM design, using HDMI input and, in some cases, HDMI output for passthrough or expanded deployment. The Comet Q moves away from that approach by replacing HDMI capture with USB-C video input via DP Alt Mode, which changes both the kind of device it can connect to and the environments where it is likely to be used.

In hardware terms, the Comet Q is also the most lightweight system in the lineup. Its dual-core ARM Cortex-A53 processor and 512MB of LPDDR4 memory place it below the other Comet devices, which generally use quad-core ARM processors and 1GB of DDR3L memory. Its sub-2.5W power draw is also the lowest figure listed across the range. That lower hardware ceiling makes sense in context, as the Comet Q appears to prioritise mobility, compact deployment, and low power operation over the broader feature scope of the higher-end HDMI-based models.

The other Comet devices are more clearly structured for fixed installations or more complex remote management roles. The Comet PoE adds Power over Ethernet support for simpler networked deployment, the Comet Pro adds both HDMI input and output, and the Comet 5G extends this further with cellular connectivity through 4G LTE and 5G RedCap support. Compared with those, the Comet Q is not trying to be the most feature-rich model. Instead, it fills a separate position by targeting USB-C connected client hardware and a more portable usage model than the rest of the lineup.

This makes the Comet Q less of a direct replacement for the other Comet units and more of a complementary product. The HDMI-based models remain better suited to desktops, servers, fixed workstations, and network infrastructure where traditional video capture and broader connectivity options matter more. The Comet Q, by contrast, is better understood as a compact access tool for modern mobile and USB-C centric devices, where physical size, single-cable deployment, and lower power use are more important than maximum processing resources or infrastructure-oriented connectivity.

Interested in Gl.iNet KVM Devices? Here are some great options available NOW: Gl.iNet Comet Pro $179 – HERE on Amazon Gl.iNet Comet 5G $299 – HERE on Amazon Gl.iNet Comet PoE $129 – HERE on Amazon

Gl.iNet Comet Q KVM Price & Release Date?

At the time of filming, Gl.iNet had not confirmed a final release schedule for the Comet Q, and availability was still being discussed internally. The device shown in Shenzhen appeared to be relatively close to completion from a hardware and interface perspective, but it was still clearly in a pre-release state, with software behaviour, feature scope, and final implementation details still being adjusted. Gl.iNet also indicated that the launch route under consideration could involve Kickstarter, which suggests the company is still assessing demand and market positioning for this particular model.

Pricing was also not final at the time of the demonstration. The only estimate provided was a broad target range of around $100 to $200, with the expectation that the final retail position would likely sit closer to the lower end of that range than the upper end. Until Gl.iNet confirms official launch pricing, regional availability, and a release timetable, the Comet Q remains a revealed but not yet fully commercialised addition to the wider Comet KVM lineup.

SUBSCRIBE TO OUR NEWSLETTER
[contact-form-7]
Join Inner Circle

This description contains links to Amazon. These links will take you to some of the products mentioned in today's content. As an Amazon Associate, I earn from qualifying purchases. Visit the NASCompares Deal Finder to find the best place to buy this device in your region, based on Service, Support and Reputation - Just Search for your NAS Drive in the Box Below

Need Advice on Data Storage from an Expert?

Finally, for free advice about your setup, just leave a message in the comments below here at NASCompares.com and we will get back to you. Need Help? Where possible (and where appropriate) please provide as much information about your requirements, as then I can arrange the best answer and solution to your needs. Do not worry about your e-mail address being required, it will NOT be used in a mailing list and will NOT be used in any way other than to respond to your enquiry. [contact-form-7] TRY CHAT Terms and Conditions   Alternatively, why not ask me on the ASK NASCompares forum, by clicking the button below. This is a community hub that serves as a place that I can answer your question, chew the fat, share new release information and even get corrections posted. I will always get around to answering ALL queries, but as a one-man operation, I cannot promise speed! So by sharing your query in the ASK NASCompares section below, you can get a better range of solutions and suggestions, alongside my own.

WE LOVE COFFEE

 

 

Minisforum Expand Their NAS Range with 2 New Flash NAS – The S5 and S7 SSD Systems

Minisforum has revealed 2 new all-flash NAS systems, the Minisforum S5 and Minisforum S7, as part of a joint event with Intel in Xiamen, China. The announcement, covered in the official Minisforum press release and by VideoCardz, places both devices under the company’s “Agent NAS” branding, with a focus on SSD-only storage, local AI services, and compact edge computing rather than conventional HDD-based NAS design. The S5 is presented as the smaller and quieter model for environments where noise output is a priority, while the S7 is positioned as a larger all-flash system derived from Minisforum’s workstation platform, intended for users with higher storage density and network connectivity requirements.

Minisforum S5 NAS Hardware

The Minisforum S5 is the smaller of the 2 new all-flash NAS systems and is built around a compact, fanless design. Its storage configuration is based on 5 M.2 2280 PCIe 4.0 x1 SSD slots, with no support for conventional 3.5-inch or 2.5-inch hard drives. This places the S5 closer to a silent flash storage appliance than a traditional desktop NAS, with the design clearly prioritising low noise, reduced mechanical vibration, and a smaller physical footprint.

The system uses an Intel Core Series 3 platform, identified in wider reporting as Wildcat Lake, alongside up to 16GB of LPDDR5X-7200 memory. The RAM appears to be soldered rather than user-upgradeable, which fits with the compact and fanless design. Connectivity is stronger than many small NAS systems, with 10GbE, 2.5GbE, USB4, HDMI, USB-A, and Wi-Fi 7 all listed. The use of PCIe 4.0 x1 per SSD slot means the device is not designed to deliver full-speed PCIe 4.0 x4 performance from each individual NVMe drive, but instead uses multiple SSDs for an all-flash pool in a low-noise enclosure.

Minisforum S7 NAS Hardware

The Minisforum S7 is the larger of the 2 systems and is based on the Minisforum MS-03 workstation platform. Unlike the S5, which is focused on passive cooling and compact silent operation, the S7 is aimed at higher-capacity all-flash storage and more advanced networking. It supports up to 7 NVMe SSDs, making it the more suitable model for users who need a denser flash-based storage pool in a small workstation-style chassis.

The S7 also has a more network-heavy configuration than the S5, with dual 10G SFP+ ports, 10GbE RJ45, 2.5GbE RJ45, and 2x USB4 40Gbps ports. Minisforum has also listed an LED status display for checking system and storage activity directly from the device. Full details on PCIe lane allocation for the 7 SSD slots have not yet been confirmed, so it is not currently clear whether all slots operate at the same speed or whether any of them receive more bandwidth than others.

Minisforum S5 NAS Price and Release Date

Minisforum has not confirmed pricing or a release date for the S5 NAS at the time of the announcement. The official material presents the system as part of the company’s new Agent NAS lineup, but does not include retail availability, regional rollout details, or launch pricing. Given the use of 5x NVMe SSD slots, 10GbE, USB4, Wi-Fi 7, soldered LPDDR5X memory, and a fanless chassis, the final price will be important in determining how it compares with both conventional HDD-based NAS systems and other compact all-flash NAS devices.

SUBSCRIBE TO OUR NEWSLETTER
[contact-form-7]
Join Inner Circle

This description contains links to Amazon. These links will take you to some of the products mentioned in today's content. As an Amazon Associate, I earn from qualifying purchases. Visit the NASCompares Deal Finder to find the best place to buy this device in your region, based on Service, Support and Reputation - Just Search for your NAS Drive in the Box Below

Need Advice on Data Storage from an Expert?

Finally, for free advice about your setup, just leave a message in the comments below here at NASCompares.com and we will get back to you. Need Help? Where possible (and where appropriate) please provide as much information about your requirements, as then I can arrange the best answer and solution to your needs. Do not worry about your e-mail address being required, it will NOT be used in a mailing list and will NOT be used in any way other than to respond to your enquiry. [contact-form-7] TRY CHAT Terms and Conditions   Alternatively, why not ask me on the ASK NASCompares forum, by clicking the button below. This is a community hub that serves as a place that I can answer your question, chew the fat, share new release information and even get corrections posted. I will always get around to answering ALL queries, but as a one-man operation, I cannot promise speed! So by sharing your query in the ASK NASCompares section below, you can get a better range of solutions and suggestions, alongside my own.

WE LOVE COFFEE

 

Dans l’univers de la signature électronique, un acteur s’est largement imposé : DocuSign. Mais ce dernier coûte très cher… Entre abonnements mensuels, limitations d’enveloppes et frais annexes (SMS, vérification d’identité, support), la facture peut grimper très vite. Face à ce modèle, DocuSeal propose une approche radicalement différente : open source, gratuit et auto-hébergeable.

DocuSeal : le DocuSign 100% gratuit

Lancée en 2023, DocuSeal est née d’un besoin simple : signer des documents sans abonnement. En quelques semaines, le projet devient une solution complète et aujourd’hui largement adoptée sur GitHub.

Fonctionnellement, DocuSeal couvre l’essentiel (et même plus) :

• Transformation de PDF en formulaires interactifs ;
• Éditeur drag-and-drop avec 13 types de champs ;
• Gestion multi-signataires avec ordre personnalisé ;
• Notifications et rappels automatisés ;
• Signature mobile fluide ;
• Audit complet et vérification des signatures ;
• Modèles et envois en masse ;
• API complète pour intégration SI…

On est loin d’un simple clone, DocuSeal vise clairement les usages professionnels.

Docker pour faciliter l’auto-hébergement

L’un des points forts de DocuSeal, c’est son déploiement. Une simple commande Docker suffit pour lancer une instance complète. Pas de dépendance complexe, pas de SaaS opaque.

Ce choix technique change tout :

• Vos documents restent sur votre infrastructure ;
• Aucun tiers n’accède à vos contrats ;
• Conformité facilitée (RGPD, confidentialité interne) ;
• Scalabilité maîtrisée selon vos besoins.

Dans un contexte où la souveraineté des données devient critique, c’est là aussi un sacré avantage.

Zéro coût, zéro limite

Là où DocuSign facture chaque fonctionnalité, DocuSeal adopte un modèle sans friction :

• Documents illimités ;
• Signataires illimités ;
• Stockage illimité ;
• Aucun coût caché.

Une petite société peut économiser plusieurs milliers d’euros par an… quelque soit la taille de l’équipe. Tout au plus, il faudra passer par un VPS pour quelques euros par mois. A noter que Docuseal propose sur ses serveurs avec une offre gratuite et une payante. Et contrairement à ce que l’on pourrait penser d’un projet open source récent, DocuSeal affiche déjà un niveau de maturité sérieux avec des certifications ISO 27001, SOC 2, HIPAA, ainsi que la conformité au RGPD.

Docker

La façon la plus simple pour profiter de DocuSeal, c’est très certainement l’utilisation avec Docker… par exemple sur un NAS. L’éditeur fournit un docker-compose.yml complet et facile à adapter.

services:
  app:
    depends_on:
      postgres:
        condition: service_healthy
    image: docuseal/docuseal:latest
    ports:
      - 3000:3000
    volumes:
      - ./docuseal:/data/docuseal
    environment:
      - FORCE_SSL=${HOST}
      - DATABASE_URL=postgresql://postgres:postgres@postgres:5432/docuseal

  postgres:
    image: postgres:18
    volumes:
      - './pg_data:/var/lib/postgresql/18/docker'
    environment:
      POSTGRES_USER: postgres
      POSTGRES_PASSWORD: postgres
      POSTGRES_DB: docuseal
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U postgres"]
      interval: 5s
      timeout: 5s
      retries: 5

  caddy:
    image: caddy:latest
    command: caddy reverse-proxy --from $HOST --to app:3000
    ports:
      - 80:80
      - 443:443
      - 443:443/udp
    volumes:
      - ./caddy:/data/caddy
    environment:
      - HOST=${HOST}

L’application fonctionne avec la base de données PostgreSQL et Caddy pour le reverse proxy.

En synthèse

DocuSeal n’est pas juste une alternative économique. C’est une remise en question du modèle SaaS appliqué à la signature électronique. Pour les équipes techniques, les startups ou les entreprises sensibles à la confidentialité, c’est une option crédible, robuste et alignée avec les pratiques modernes d’auto-hébergement (lien vers GitHub).

UGREEN NAS and OpenClaw – How to Install it, Setup Your AI and Understanding The Risks!

OpenClaw has now moved from a manual self-hosted setup into the UGREEN UGOS Pro App Center, making it possible to install the assistant gateway directly on supported UGREEN NASync systems rather than building it manually through a VM, terminal commands, or a separate always-on PC. In practical terms, OpenClaw is not the AI model itself. It is the local assistant layer that connects your NAS, files, tools, skills, and messaging channels to an LLM such as OpenAI, Gemini, DeepSeek, MiniMax, OpenRouter, or a local model where supported. This matters because a NAS is where many users already keep their long-term data, backups, media, documents, and project files, but it also means OpenClaw needs to be treated as a privileged automation tool rather than a simple chatbot. The App Center version lowers the installation barrier, but the real value and the real risk both come from what you allow it to access, what model you connect it to, and which skills or messaging channels you enable.

Security Considerations Before Giving OpenClaw NAS Access

OpenClaw should be approached as a privileged automation layer, not as a normal chat assistant. On UGREEN NAS, it runs through Docker and is designed to read, write, delete, move files, publish messages, and execute system-level operations depending on what folders, skills, and channels you enable. UGREEN’s own notes state that the application runs inside a Docker container with root privileges, which is necessary for broad automation but also increases the potential damage from incorrect commands, poor configuration, compromised plugins, or prompt-injection style attacks. The main rule before installation is therefore straightforward: do not give OpenClaw access to anything you are not prepared for it to modify, and make sure you have a working backup before testing it on real data.

The risks increase further if you connect OpenClaw to a remote LLM provider or public messaging platform. When using OpenAI, Gemini, MiniMax, DeepSeek, OpenRouter, or similar services, your prompts, file context, directory details, task instructions, logs, or extracted content may be sent outside the NAS depending on the action being performed. OpenClaw’s own GitHub description warns that real messaging surfaces should be treated as untrusted inputs, and recent reporting has also highlighted malicious third-party OpenClaw skills that attempted to steal credentials, wallet data, and browser information. For NAS use, the safest starting point is to use a test folder, avoid private or business-critical data, do not expose the service directly to the public internet, install only trusted skills, and treat WhatsApp, Telegram, Discord, Slack, or similar integrations as external access points into your NAS assistant.

Preparing Folders and Installing OpenClaw from UGOS Pro

Before installing OpenClaw, create the folders it will use and decide how much of the NAS it should be allowed to see. In Files, either create a dedicated shared folder or a personal folder for OpenClaw’s workspace, for example openclaw-data or openclaw-workspace. This should ideally be an empty folder, as it will be used for temporary files, generated content, working data, and task execution. Separately, create or identify the folder that OpenClaw will be allowed to access for real NAS file operations. For first-time testing, this should be a limited test directory rather than a folder containing live backups, sensitive documents, business files, or irreplaceable media. The workspace path and file access path should not overlap, and the access path should not sit inside the workspace folder. UGREEN also notes that Docker should be installed and updated first, as OpenClaw relies heavily on the Docker container environment.

Once the folders are ready, open App Center in UGOS Pro, find OpenClaw under the app list, and select Install. During the installation wizard, set the Workspace path to the empty folder created for OpenClaw’s internal working area, then set the File access path to the NAS folder or folders that the assistant is permitted to read or modify. Multiple access paths can be added, but this should be done deliberately, as these paths define the practical scope of what OpenClaw can act on. Next, create a Gateway token, which will be required when signing in to the OpenClaw web interface. After reviewing the risk notice, tick the confirmation box and start the installation. The package is installed through the App Center, but it still deploys and runs through Docker in the background, so installation time will depend on NAS performance, internet connection speed, and the state of the Docker environment. OpenClaw’s own Docker documentation also describes the gateway token and container-based control UI as central parts of the deployment model.

Point-by-point setup:

1. Open Files in UGOS Pro.
2. Create an empty folder for the OpenClaw workspace, such as openclaw-workspace.
3. Create a separate test folder for OpenClaw file access, such as openclaw-test-data.
4. Avoid selecting folders that contain private, business-critical, backup, password, financial, or personal archive data.
5. Open App Center.
6. Confirm Docker is installed and updated.
7. Search for OpenClaw under the app list.
8. Click Install.
9. Set the Workspace path to the empty OpenClaw working folder.
10. Set the File access path to the limited folder OpenClaw can manage.
11. Add additional access paths only if they are required.
12. Create and record a Gateway token.
13. Read the installation risk notice.
14. Tick the confirmation box.
15. Click Install and allow the deployment to complete.

Linking OpenClaw to an AI Model Provider

OpenClaw needs an AI model before it can act as an assistant. The UGREEN App Center installation can collect model details during setup, but these can also be managed later from the OpenClaw console under Model providers. The information required is usually the same across providers: a base URL or request endpoint, a model name, and an API key.

OpenAI’s current API reference lists https://api.openai.com/v1 as the standard API base, with chat completions available under /chat/completions, while Google documents Gemini’s OpenAI-compatible endpoint as https://generativelanguage.googleapis.com/v1beta/openai/. These details matter because a wrong endpoint, wrong model name, or invalid key will usually result in provider errors inside OpenClaw rather than a NAS-side installation problem.

For a UGREEN NAS setup, most users will start with a remote model provider such as OpenAI, Google Gemini, DeepSeek, MiniMax, Groq, OpenRouter, or another OpenAI-compatible API. iDX models with local AI model support may also allow local model use, but that depends on the local model service exposing a usable API endpoint and key.

A remote model is easier to configure, but it can send task instructions, file context, extracted text, and other prompts outside the NAS. A local model reduces this dependency, but it may require more RAM, more setup, and a compatible local inference service. OpenClaw supports model provider configuration and key rotation through its own provider system, so the NAS app should be treated as the deployment layer rather than the only place where model behaviour can be managed.

Point-by-point setup:

1. Open the OpenClaw shortcut from the UGREEN NAS desktop.
2. Sign in using the Gateway token created during installation.
3. Go to Model providers in the left-side menu.
4. Click Add provider.
5. Select the provider you want to use, such as OpenAI, Google Gemini, DeepSeek, MiniMax, or another supported provider.
6. Enter the provider’s Base URL or full API endpoint.
7. Enter the required API key from the provider’s developer console.
8. Enter or select the Model name that matches the provider’s supported model ID.
9. Save the provider configuration.
10. Go to the Default model area.
11. Select the model OpenClaw should use by default.
12. Click Save to apply the default model.
13. Open Chat and send a basic test prompt, such as What model are you running on?
14. If OpenClaw returns a provider error, check the API key, model name, endpoint format, account billing status, and provider rate limits.
15. If using a local model on an iDX system, use the local service IP address and port as the base URL rather than a public cloud endpoint.

Common examples:

Opening the OpenClaw Console and Testing the Assistant

Once installation and model configuration are complete, OpenClaw can be launched from the UGREEN NAS desktop. Clicking the OpenClaw shortcut opens the web console in a browser, where the first prompt will ask for the Gateway token created during installation. After signing in, the Overview page shows whether the gateway is active, along with container runtime details such as uptime, CPU usage, memory usage, gateway port, process information, and overall service status. If the service is running correctly, the status area should show an active or healthy state, and the Open OpenClaw button can be used to launch the native OpenClaw interface in a new browser tab.

The first test should be simple. Open the Chat page, send a basic message, and confirm that the configured model responds. After that, test only against the limited folder path selected during setup. For example, ask it to list files in the permitted test directory, create a new folder inside it, or summarize a non-sensitive test document. This confirms that OpenClaw, the model provider, and the NAS file permissions are all working together. If the model does not respond, check the model provider settings first. If file actions fail, check that the command references the correct mounted path shown in the OpenClaw app configuration rather than a folder name as displayed in the normal UGREEN Files interface.

Point-by-point setup:

1. Open the OpenClaw shortcut from the UGREEN NAS desktop.
2. Enter the Gateway token.
3. Click Sign in or Connect.
4. Open Overview.
5. Confirm the service status is active.
6. Check the runtime snapshot for CPU usage, memory usage, and uptime.
7. Click Open OpenClaw if you want to use the native OpenClaw interface.
8. Open Chat from the left-side menu.
9. Send a basic test message, such as Hello.
10. Ask which model is active, for example What model are you using?
11. Test file access only inside the approved test folder.
12. Use a low-risk command, such as Create a folder called OpenClaw Test in [your mounted test path].
13. Open Files in UGOS Pro and confirm the folder was created.
14. If the command fails, check the actual accessible path under Control Panel > About > Apps > OpenClaw.
15. Review Operation Logs if OpenClaw responds incorrectly, fails to access files, or reports a gateway or provider error.

Choosing Skills and Plug-ins for NAS-Based Use

OpenClaw skills and plug-ins extend what the assistant can do beyond basic chat. In a NAS environment, these additions should be chosen more cautiously than they might be on a laptop or test VM, because a skill may request access to files, shell commands, browser sessions, messaging platforms, or external services. OpenClaw’s public site describes it as an assistant that can act through channels such as WhatsApp, Telegram, and other chat apps, while community skill indexes now list thousands of available skills. That breadth is useful, but it also means the skill ecosystem should not be treated as automatically safe or suitable for storage systems. (openclaw.ai, clawskills.sh)

For a UGREEN NAS, the sensible starting point is to enable only the skills that match a specific NAS task. File management, system monitoring, web browsing, document parsing, OCR, and basic notification workflows are the most relevant categories. Avoid installing skills that request broad shell access, browser credential access, crypto wallet access, password manager access, or unclear third-party scripts unless they have been reviewed carefully. This is not theoretical. Reports in early 2026 documented malicious OpenClaw skills that attempted to steal browser data, SSH credentials, wallet information, and other sensitive data, which is particularly relevant when the assistant is being installed on a machine that stores personal or business files.

Point-by-point setup:

1. Open the OpenClaw console from the UGREEN NAS desktop.
2. Sign in with the Gateway token.
3. Open the native OpenClaw interface using Open OpenClaw, if required.
4. Go to the Skills, Plug-ins, or Skills Store area.
5. Search for skills by function rather than installing large bundles.
6. Start with NAS-relevant categories only.
7. Check the skill description, source, permissions, and install method.
8. Avoid any skill that asks you to run unclear terminal commands.
9. Install 1 skill at a time.
10. Test it only against the approved OpenClaw test folder.
11. Check Operation Logs after each test.
12. Remove any skill that behaves unexpectedly or asks for broader permissions than needed.
13. Avoid using “always allow” approvals until the workflow has been tested repeatedly.
14. Keep a note of which skills are installed and what each one can access.
15. Review installed skills periodically, especially after OpenClaw or UGOS Pro updates.

Recommended NAS-related starting points:

Connecting OpenClaw to WhatsApp, Discord, and Telegram

OpenClaw can be used through external messaging platforms so that commands can be sent to the NAS assistant without opening the UGREEN web interface each time. The supported channel list includes WhatsApp, Telegram, Discord, Slack, Google Chat, Signal, Microsoft Teams, Matrix, Feishu, LINE, Mattermost, Nextcloud Talk, and others, but WhatsApp, Telegram, and Discord are likely to be the most relevant for home and small-office users. OpenClaw’s own channel notes state that multiple channels can run at the same time, but they also warn that inbound messages should be treated as untrusted input and that DM pairing or allowlists are used for access control. (github.com, docs.openclaw.ai)

For NAS use, Telegram is usually the simplest starting point because it relies on a bot token from BotFather. WhatsApp normally uses QR pairing and stores more session state on disk, which means it may need more care during backups, container resets, or reinstallation. Discord is more useful when OpenClaw needs to operate inside a server, channel, or team context, but it should be restricted to private channels and trusted roles rather than broad server-wide access. The UGREEN console provides a channel management area where plugins can be enabled, configured, and monitored, but more advanced channel setup may still require working inside the OpenClaw interface or Docker container depending on the platform and plugin. (docs.openclaw.ai, openclaw-openclaw.mintlify.app)

Point-by-point setup:

1. Open the OpenClaw console from the UGREEN NAS desktop.
2. Sign in using the Gateway token.
3. Go to Channels.
4. Select the channel you want to enable, such as Telegram, WhatsApp, or Discord.
5. Click Enable for the required channel plugin.
6. Wait until the plugin status changes to Ready.
7. Click Add channel.
8. Enter the required account, bot, or pairing details.
9. Configure DM access rules, pairing mode, or allowlist behaviour where available.
10. Bind the channel to the correct OpenClaw agent or default assistant.
11. Send a low-risk test message from the external app.
12. Confirm that OpenClaw replies through the same channel.
13. Test with a harmless NAS action inside the approved test directory only.
14. Check Operation Logs if messages are received but not answered.
15. Disable the channel if unexpected users, groups, or servers can trigger the assistant.

Telegram setup:

1. Open Telegram.
2. Search for @BotFather.
3. Create a new bot using /newbot.
4. Copy the bot token provided by BotFather.
5. Return to OpenClaw > Channels.
6. Enable the Telegram plugin.
7. Add a Telegram channel.
8. Paste the bot token.
9. Configure whether the bot can respond in DMs, groups, or both.
10. Send a test message to the bot.

WhatsApp setup:

1. Open OpenClaw > Channels.
2. Enable the WhatsApp plugin.
3. Add a WhatsApp channel.
4. Enter the phone number or pairing account details requested by the setup window.
5. Generate the QR pairing code.
6. Open WhatsApp on your phone.
7. Go to linked devices.
8. Scan the QR code.
9. Wait for the WhatsApp channel status to become active.
10. Send a test message to confirm OpenClaw responds.

Discord setup:

1. Create or use a Discord server where you control permissions.
2. Create a dedicated private channel for OpenClaw commands.
3. Create a Discord bot in the Discord Developer Portal.
4. Copy the bot token.
5. Return to OpenClaw > Channels.
6. Enable the Discord plugin.
7. Add a Discord channel.
8. Paste the bot token.
9. Restrict the bot to trusted channels and roles.
10. Send a test command in the private OpenClaw channel.

UGREEN x OpenClaw: Useful, but Only with Controlled Access

OpenClaw on UGREEN NAS is a notable step towards making AI-assisted NAS management more accessible, mainly because the App Center version removes much of the older manual deployment work. Instead of installing Ubuntu in a VM, configuring Node.js, running installation scripts, and manually binding the gateway, supported UGREEN NASync users can now install OpenClaw through UGOS Pro and complete the main path, token, and model setup through a guided interface. That makes the initial process easier, but it does not make OpenClaw a basic consumer NAS feature. It is still an automation agent with access to files, tools, model providers, messaging channels, and potentially system commands.

The value depends on how tightly it is configured. Used against a limited folder, with a known model provider, a small number of trusted skills, and private messaging channels, OpenClaw can help with file organisation, document handling, system checks, reminders, and assistant-style NAS interaction. Given broad storage access, untested skills, exposed web access, or remote AI services without understanding the data flow, it becomes a much higher-risk deployment. For most users, the best approach is to begin with a test directory, avoid sensitive data, keep backups current, and expand access only after confirming exactly how OpenClaw behaves in day-to-day use.

Want to Support the work me and Eddie do at NASCompares? If you found this article helpful and are going to buy a UGREEN NAS from the brand’s official site or from Amazon, use the links below. Using these links will result in a small commission coming to us (which costs you nothing extra) and it allows us to keep doing what we do! Thank you for keeping the internet a fair and sustainable place!

Buy the UGREEN NAS from your local Amazon Store	Buy the UGREEN NAS solutions Directly from UGREEN.COM

 

---

 

SUBSCRIBE TO OUR NEWSLETTER
[contact-form-7]
Join Inner Circle

This description contains links to Amazon. These links will take you to some of the products mentioned in today's content. As an Amazon Associate, I earn from qualifying purchases. Visit the NASCompares Deal Finder to find the best place to buy this device in your region, based on Service, Support and Reputation - Just Search for your NAS Drive in the Box Below

Need Advice on Data Storage from an Expert?

Finally, for free advice about your setup, just leave a message in the comments below here at NASCompares.com and we will get back to you. Need Help? Where possible (and where appropriate) please provide as much information about your requirements, as then I can arrange the best answer and solution to your needs. Do not worry about your e-mail address being required, it will NOT be used in a mailing list and will NOT be used in any way other than to respond to your enquiry. [contact-form-7] TRY CHAT Terms and Conditions   Alternatively, why not ask me on the ASK NASCompares forum, by clicking the button below. This is a community hub that serves as a place that I can answer your question, chew the fat, share new release information and even get corrections posted. I will always get around to answering ALL queries, but as a one-man operation, I cannot promise speed! So by sharing your query in the ASK NASCompares section below, you can get a better range of solutions and suggestions, alongside my own.

WE LOVE COFFEE

 

UniFi UNVR Gen 2 and UNVR Gen 2 Pro: What Has Actually Changed?

UniFi’s UNVR range has always occupied a fairly clear role in the Protect ecosystem: a dedicated rackmount recorder for users who have outgrown smaller gateway-based recording, or who want their surveillance storage separated from the rest of their network hardware. With the new UNVR Gen 2 and UNVR Gen 2 Pro, Ubiquiti is shifting that role further. These are still network video recorders first, but the hardware and software changes point toward a more active surveillance appliance, with higher camera support, local AI processing, HDMI output for live viewing, and closer integration with the newer direction of UniFi Protect. The issue is that this also comes with a much higher price than the previous UNVR and UNVR Pro, so the question is not simply whether the Gen 2 models are better, but whether the added hardware and features are relevant enough to justify the increase for different types of deployments.

UNVR G.2 and UNVR G.2 Pro – Specifications

The UNVR Gen 2 is the 1U model in the new range and keeps the same general rackmount class as the original UNVR, with 4 2.5/3.5″ HDD or SSD bays. Its camera support is rated at up to 50 HD cameras, 35 2K cameras, or 25 4K cameras, with support for 150+ Access Hubs. Networking is handled by 1 10G SFP+ port and 1 2.5GbE RJ45 port, while the chassis measures 442.4 x 43.7 x 325 mm.

Internally, it moves to a Qualcomm Kryo CPU built on Arm Cortex technology using a 4 nm process, with 1 Prime core at 3.2 GHz, 4 Performance cores at 2.8 GHz, and 3 Efficiency cores at 2.0 GHz. Memory is increased to 8 GB, drive power budget remains 75W, and maximum system power consumption is listed at 100W.

The UNVR Gen 2 Pro is the larger 2U model and increases the drive count to 8 2.5/3.5″ HDD or SSD bays, compared with 7 bays on the previous UNVR Pro. Its camera support is rated at up to 100 HD cameras, 70 2K cameras, or 50 4K cameras, again with support for 150+ Access Hubs. The Pro model uses the same Qualcomm Kryo CPU arrangement as the smaller Gen 2 model, but increases memory to 16 GB. Networking consists of 1 10G SFP+ port and 1 2.5GbE RJ45 port, with a listed chassis size of 442.4 x 87.4 x 325 mm. The drive power budget rises to 155W, while maximum power consumption is listed at 200W.

Both Gen 2 models also include HDMI output for the built-in ViewPort function, which allows a Protect multi-view to be assigned directly to a connected display rather than requiring a separate ViewPort device.

Specification	UniFi UNVR Gen 2	UniFi UNVR Gen 2 Pro
Price	$699	$999
Form factor	Rackmount 1U	Rackmount 2U
Dimensions	442.4 x 43.7 x 325 mm	442.4 x 87.4 x 325 mm
Drive bays	4 x 2.5/3.5″ HDD/SSD	8 x 2.5/3.5″ HDD/SSD
Managed cameras	50 HD / 35 2K / 25 4K	100 HD / 70 2K / 50 4K
Managed Access Hubs	150+	150+
Networking	1 x 10G SFP+ / 1 x 2.5GbE RJ45	1 x 10G SFP+ / 1 x 2.5GbE RJ45
HDMI output	Yes, built-in ViewPort	Yes, built-in ViewPort
ViewPort stream limit	Up to 16 streams	Up to 16 streams
Processor	Qualcomm Kryo CPU built on Arm Cortex technology, 4 nm	Qualcomm Kryo CPU built on Arm Cortex technology, 4 nm
CPU configuration	1 Prime core at 3.2 GHz, 4 Performance cores at 2.8 GHz, 3 Efficiency cores at 2.0 GHz	1 Prime core at 3.2 GHz, 4 Performance cores at 2.8 GHz, 3 Efficiency cores at 2.0 GHz
Memory	8 GB	16 GB
Integrated Edge AI	Yes	Yes
AI detections	Up to 1,000 per hour	Up to 1,000 per hour
Edge AI features	Natural Language Search, Object Indexing in Find Anything, Person ReID, Search by Image	Natural Language Search, Object Indexing in Find Anything, Person ReID, Search by Image
Max. drive power budget	75W	155W
Max. power consumption	100W	200W
Power method	Universal AC input, 100 to 240V AC, 50/60 Hz	Universal AC input, 100 to 240V AC, 50/60 Hz
Power supply	Internal PSU, 100W	Internal PSU, 200W
Minimum NVR version	Not specified in supplied notes	5.1.10
Minimum Protect version	7.1.46	7.1.46

UNVR Gen 2 vs Original UNVR: Where the Price Increase Comes From

The clearest difference between the original UNVR and the UNVR Gen 2 is the change in hardware platform. The older UNVR uses a quad-core ARM Cortex-A57 processor at 1.7 GHz with 4 GB of memory, while the UNVR Gen 2 moves to the newer Qualcomm Kryo ARM-based CPU platform and 8 GB of memory. The network layout has also changed, with the older model using 1 GbE RJ45 alongside 10G SFP+, while the Gen 2 model upgrades the RJ45 connection to 2.5GbE. Storage bay count remains the same at 4 bays, but camera capacity changes from 60 HD, 30 2K, or 18 4K cameras on the original UNVR to 50 HD, 35 2K, or 25 4K cameras on the Gen 2 model. That means the newer system is not a straight increase across every camera category, but it does raise support for higher-resolution 2K and 4K deployments.

The pricing difference is more substantial than the specification changes would suggest if this were only a conventional recorder update. The original UNVR is positioned at $299, while the UNVR Gen 2 is positioned at $699. The explanation appears to be that UniFi is treating the Gen 2 model as a more complete Protect appliance rather than just a higher-performance version of the old 4-bay recorder.

The HDMI output effectively integrates ViewPort-style live display support, while the built-in Edge AI features shift part of the workload that would otherwise require additional hardware such as an AI Key. This does not make the older UNVR obsolete for simpler recording tasks, but it does change the buying decision. The Gen 2 model is aimed more clearly at deployments that need local AI search, image-based search, person re-identification, and direct live monitoring from the recorder itself.

Specification	Older UNVR / UNVR Pro	New UNVR Gen 2 / UNVR Gen 2 Pro
Models compared	UNVR / UNVR Pro	UNVR Gen 2 / UNVR Gen 2 Pro
Price	$299 / $499	$699 / $999
Form factor	1U / 2U	1U / 2U
Dimensions	442 x 325 x 44 mm / 442 x 325 x 87 mm	442.4 x 43.7 x 325 mm / 442.4 x 87.4 x 325 mm
Drive bays	4 x 2.5/3.5″ HDD/SSD / 7 x 2.5/3.5″ HDD/SSD	4 x 2.5/3.5″ HDD/SSD / 8 x 2.5/3.5″ HDD/SSD
Managed HD cameras	60 / 70	50 / 100
Managed 2K cameras	30 / 35	35 / 70
Managed 4K cameras	18 / 24	25 / 50
Managed Access Hubs	150 / 150	150+ / 150+
Networking	1 x 10G SFP+ and 1 x GbE RJ45	1 x 10G SFP+ and 1 x 2.5GbE RJ45
HDMI output	No integrated ViewPort	Yes, integrated ViewPort via HDMI
ViewPort stream limit	Requires separate ViewPort device	Up to 16 streams
Processor	Quad ARM Cortex-A57 cores at 1.7 GHz	Qualcomm Kryo CPU built on Arm Cortex technology, 4 nm
CPU configuration	4 cores	1 Prime core at 3.2 GHz, 4 Performance cores at 2.8 GHz, 3 Efficiency cores at 2.0 GHz
Memory	4 GB / 8 GB	8 GB / 16 GB
Integrated Edge AI	No	Yes
AI features	Requires additional UniFi AI hardware for expanded AI functionality	Natural Language Search, Object Indexing in Find Anything, Person ReID, Search by Image
AI detections	Not specified	Up to 1,000 per hour
Max. drive power budget	75W / 135W	75W / 155W
Max. power consumption	100W / 160W	100W / 200W
Power supply	Internal AC/DC, 120W / 200W	Internal PSU, 100W / 200W
Power redundancy	USP-RPS DC input supported	USP-RPS DC input supported
Main practical difference	Dedicated UniFi Protect recording and storage appliances	Higher-resolution camera scaling, integrated display output, and local AI search features

Protect 7.1 and the Shift Toward Local AI Surveillance

UniFi Protect 7.1 is an important part of the UNVR Gen 2 release, because several of the headline hardware features depend on the newer Protect software stack. The Gen 2 recorders include built-in Edge AI functionality, with support for Natural Language Search, Object Indexing in Find Anything, Person ReID, and Search by Image. In practical terms, this changes how recorded footage can be searched. Instead of relying only on a timeline, motion events, or predefined smart detections, the system is designed to help users locate more specific events across stored footage using more descriptive search methods. The built-in AI functionality is local and license-free, but for larger or busier deployments, UniFi still recommends adding 1 or more AI Keys to expand processing capacity, reduce Edge AI latency, and lower the chance of missed events.

Protect 7.1 also expands the broader surveillance feature set beyond the Gen 2 recorders themselves. Custom Video Walls are now available in Site Manager, dashboard widgets have deeper customization, and live camera views can be configured with webhook shortcuts for triggering automations from the camera interface. Smart detections have been retrained for improved accuracy across UniFi cameras, PTZ tracking has been expanded to include vehicles, and 360 cameras now support native immersive downloads. ONVIF support is also more developed, with audio and motion detection support for third-party cameras, which is significant for sites migrating gradually from existing surveillance hardware into UniFi Protect. The update also introduces U.S.-only Noonlight dispatch services for sensor and video monitoring at $199 per year, DC-09 support for third-party monitoring integrations, and SuperLink Remote Control support for customizable site control. Below is a full breakdown of the feaures of UniFi Protect, and which require AI assistance (either edge based on the camera/AI-port, or local using an AI assisted server or AI-Key):

Bottom Line: A More Capable NVR, but Not a Like-for-Like Replacement

The UNVR Gen 2 and UNVR Gen 2 Pro make more sense when viewed as expanded Protect appliances rather than direct replacements for the older UNVR and UNVR Pro. The new models add faster processing, more memory, 2.5GbE RJ45 networking, HDMI output for built-in ViewPort use, higher 2K and 4K camera ceilings, and local Edge AI features that change how recorded footage can be searched and reviewed. The Pro model also gains an 8th drive bay, which makes it a cleaner fit for larger retention requirements than the older 7-bay UNVR Pro. For sites already planning to use AI search, Person ReID, image-based search, or a direct HDMI monitoring display, the higher price is easier to explain because those functions would otherwise involve additional hardware or a less integrated setup.

That does not mean the price increase will make sense for every Protect installation. The original UNVR and UNVR Pro remain better aligned with users who mainly need reliable recording, centralized Protect storage, and conventional camera management without paying for a broader AI-enabled appliance. The Gen 2 models are therefore best judged by deployment requirements rather than by bay count alone. For new or expanding surveillance environments with higher-resolution cameras, active monitoring, ONVIF migration plans, and a need to search footage more intelligently, the UNVR Gen 2 range has a clearer role. For simpler sites where AI features and HDMI ViewPort output are not a priority, the older UNVR models still have a practical argument, provided UniFi continues to keep them available.

SUBSCRIBE TO OUR NEWSLETTER
[contact-form-7]
Join Inner Circle

This description contains links to Amazon. These links will take you to some of the products mentioned in today's content. As an Amazon Associate, I earn from qualifying purchases. Visit the NASCompares Deal Finder to find the best place to buy this device in your region, based on Service, Support and Reputation - Just Search for your NAS Drive in the Box Below

Need Advice on Data Storage from an Expert?

Finally, for free advice about your setup, just leave a message in the comments below here at NASCompares.com and we will get back to you. Need Help? Where possible (and where appropriate) please provide as much information about your requirements, as then I can arrange the best answer and solution to your needs. Do not worry about your e-mail address being required, it will NOT be used in a mailing list and will NOT be used in any way other than to respond to your enquiry. [contact-form-7] TRY CHAT Terms and Conditions   Alternatively, why not ask me on the ASK NASCompares forum, by clicking the button below. This is a community hub that serves as a place that I can answer your question, chew the fat, share new release information and even get corrections posted. I will always get around to answering ALL queries, but as a one-man operation, I cannot promise speed! So by sharing your query in the ASK NASCompares section below, you can get a better range of solutions and suggestions, alongside my own.

WE LOVE COFFEE

 

Lime Technology vient de publier Unraid 7.3.0, une mise à jour majeure de son système d’exploitation destiné aux NAS. Au programme : un nouveau processus d’installation, une refonte du mécanisme de licence, une mise à jour majeure de Docker et de nombreuses corrections importants (notamment Copy Fail et Dirty Frag). Voici ce qu’il faut réellement retenir de cette version…

Unraid 7.3.0

L’une des nouveautés les plus visibles de cette version concerne l’expérience d’accueil des nouveaux utilisateurs. Un assistant d’intégration (Onboarding Wizard) prend désormais en charge la configuration initiale : langue, fuseau horaire, thème visuel, paramètres de licence et choix de la méthode de démarrage . Accessible également depuis Outils → Assistant d’intégration, il permet aux utilisateurs existants de revoir leur configuration ou de migrer vers le démarrage interne.

Ce démarrage interne (internal boot) constitue d’ailleurs l’évolution structurelle majeure de cette version. Unraid peut désormais démarrer depuis un pool ZFS dédié, indépendamment de la traditionnelle clé USB. Cette approche réduit surtout la dépendance aux clés USB, jugées moins fiables à long terme. Un point important tout de même : le support de démarrage doit être accessible via des pilotes Linux natifs au moment du démarrage. Les appareils nécessitant des pilotes tiers ne sont pas compatibles.

La licence migre vers le TPM

Unraid introduit une nouvelle méthode de licence liée au TPM, appelée à coexister avec l’activation traditionnelle par clé USB. Toutes les nouvelles clés et les clés de remplacement utilisent désormais ce mécanisme par défaut, jugé plus robuste. Les utilisateurs existants peuvent effectuer cette migration manuellement à l’aide de la documentation officielle. C’est un changement discret, mais important pour la pérennité des installations : la perte ou le remplacement d’une clé USB ne devrait plus rimer avec perte de licence.

Docker 29 : attention aux adresses MAC

Docker passe de la branche 27 à la branche 29.4.3 et ce saut de version entraîne un changement de comportement : les adresses MAC des conteneurs sont désormais générées de façon aléatoire à chaque démarrage. Pour les déploiements reposant sur des réservations DHCP, des règles de pare-feu ou des ACL de switch, cela peut être problématique.

Unraid répond à ce besoin en introduisant un champ optionnel d’adresse MAC fixe directement dans les templates Docker. Les valeurs héritées présentes dans les paramètres supplémentaires (–mac-address=) sont migrées automatiquement lorsque c’est possible. Par ailleurs, des « conteneurs fantômes » (phantom containers) devenus visibles après la migration Docker 27→29 sont désormais filtrés de l’interface, sans altérer l’état interne de Docker.

Stockage : ZFS gagne en visibilité et en contrôle

Plusieurs améliorations touchent le stockage. Les fichiers corrompus dans un pool ZFS sont maintenant affichés dans l’interface, ce qui facilite le diagnostic. La taille maximale de l’ARC ZFS est désormais configurable directement depuis Réglages → Paramètres disque, sans avoir à passer par un paramètre de pilote personnalisé.

Des régressions importantes sont également corrigées : les disques 4Kn et certaines configurations LSI HBA rencontraient des problèmes de compatibilité de taille de secteur avec XFS, c’est résolu. Des correctifs concernent aussi le réveil intempestif des pools ZFS toutes les 24 heures, la détection de périphériques avec des noms longs (sdp, sdap…), etc.

Virtualisation, interface et réseau

QEMU monte en version 10.2.2, libvirt en 12.2.0, et le firmware OVMF est rafraîchi. Un bug de blocage avec virtiofs sur certains systèmes Linux invités est corrigé. Du côté réseau, Unraid enrichit son support matériel AMD avec les modules XDNA, ACP et NPU, ainsi que des firmwares Bluetooth et Wi-Fi Intel mis à jour. Une page dédiée à Tailscale fait son apparition dans les réglages, facilitant la découverte du plugin.

L’interface web bénéficie de nombreuses corrections : gestion des fins de ligne Windows dans les fichiers de configuration GRUB et Syslinux, affichage de la RAM, isolation des cœurs CPU, redémarrage automatique du daemon SSH après une reprise réseau, et formatage des notifications Discord.

En synthèse

Unraid 7.3 n’est pas une mise à jour cosmétique. Le démarrage interne, la migration TPM, la gestion des MAC Docker et les corrections ZFS constituent des changements structurels qui améliorent la fiabilité à long terme des installations. Le noyau Linux passe en version 6.18.23, et l’ensemble de la distribution de base est mis à jour avec des versions récentes de curl, OpenSSL, PHP 8.4, rclone, et bien d’autres composants. Une mise à jour à planifier sérieusement pour tout utilisateur soucieux de la stabilité de son infrastructure.

source

Copy Fail, Dirty Frag et Fragnesia sont 3 failles de sécurité différentes touchant quasiment toutes les distributions Linux. Elles permettent à un utilisateur malveillant d’obtenir un accès root (super-utilisateur) et par conséquent, disposer de tous les droits sur le système. La très grande majorité des NAS fonctionnant ave un noyau Linux, il y a de fortes chances que votre appareil soit concerné…

Synology

Commençons par le leader du secteur : Synology. Le fabricant a rapidement communiqué sur Copy Fail et Dirty Frag. Ses systèmes (DSM, SRM, BeeStation…) ne sont pas impactés par ces deux failles. Concernant Fragnesia (qui vient tout juste d’être divulguée), Synology n’a pas communiqué dessus. Cette faille exploite les mêmes modules du noyau Linux que Dirty Frag (esp4, esp6 et rxrpc), il serait donc possible que les NAS Synology soient immunisés.

Le fabricant est souvent pointé du doigt pour l’utilisation d’un noyau Linux ancien, certes robuste… mais dépourvu de certaines fonctionnalités récentes. Force est de constater que cela joue ici en sa faveur.

QNAP

QNAP publie régulièrement des mises à jour, y compris du noyau. Pour autant, aucun correctif n’est encore disponible. Il investigue toujours… le fabricant est toujours en cours d’analyse et ne propose pour l’heure qu’un ensemble de recommandations (ex. : ici et là)..

QNAP souligne que l’exploitation de ces failles nécessite un accès SSH ou Telnet, des fonctionnalités que QTS et QuTS hero réservent exclusivement au groupe Administrateur. Le risque demeure néanmoins réel.

Asustor

Asustor a réagi rapidement en publiant un premier correctif pour Copy Fail dès le 12/05 (voir l’annonce ici). Pour Dirty Frag, il faudra attendre encore quelques jours pour en savoir plus.

Pour ce qui est de Fragnesia, le fabricant n’a pas encore communiqué. La faille étant très récente, il faudra patienter encore quelques jours avant d’en savoir davantage.

TerraMaster

Terminons avec TerraMaster. Le fabricant chinois a indiqué que TOS 6 (et les versions précédentes) ne sont pas impactés par Copy Fail et Dirty Frag (comme Synology). En revanche, TOS 7, actuellement en version bêta, est bien concerné. Un correctif est déjà annoncé pour arriver très prochainement. Aucune information n’a été communiquée au sujet de Fragnesia.

En synthèse

Seuls les systèmes actuels de Synology et TerraMaster semblent épargnés par Copy Fail et Dirty Frag. Asustor a été le premier à publier un correctif (et le seul pour le moment), un autre ne devrait pas tarder. QNAP semble encore en phase d’analyse… Nous n’avons pas trouvé d’information concernant UGREEN.

Concernant Fragnesia, aucun fabricant n’a communiqué pour le moment.

Minisforum est en grande forme et annonce l’arrivée de 2 nouveaux NAS « All-Flash » : les S5 et S7. Ces machines reposent sur les processeurs Intel Core Ultra Série 3 et Intel Core Série 3. Elles illustrent surtout la volonté du constructeur de pousser l’IA générative directement dans les foyers et les petites entreprises… sans dépendre du cloud. Regardons de plus près ces nouveaux NAS.

 

S5 : le NAS qui veut se faire oublier

Le Minisforum S5 mise sur la discrétion et la performance. Ce NAS fonctionne exclusivement avec des SSD et n’a aucun ventilateur. Le boîtier dispose de 5 emplacements M.2 2280 PCIe 4.0 destinés aux SSD NVMe. Comme le montrent les visuels, le design est plutôt compact, sobre et moderne.

Le S5 est animé par un processeur Intel Core Series 3, sans plus de précision supplémentaire pour le moment. Il pourrait s’agir d’un Intel Core 7 350. Ce que l’on sait en revanche, c’est qu’il appartient à la famille Wildcat Lake et qu’il intègre un NPU capable d’atteindre 17 TOPS. Le NAS serait livré avec 16 Go de mémoire vive.

Côté connectique, le Minisforum S5 propose une configuration complète pour un format réduit :

• 2 ports USB4 ;
• 2 ports USB 3.2 Gen 2×2 ;
• 1 port réseau RJ45 10 Gb/s ;
• 1 port réseau RJ45 2,5 Gb/s ;
• 1 sortie HDMI 2.1

Toutes les interfaces sont positionnées à l’arrière du boîtier. Pour une machine aussi compacte, c’est assez impressionnant. On a vu des NAS bien plus volumineux et bien moins équipées.

S7 : une machine pensée pour le homelab

Le Minisforum S7 vise un public différent. Ce modèle reprend la plateforme MS-03 du constructeur et l’adapte dans une configuration NAS full-flash équipée de 7 emplacements NVMe.

Ici, le design est plus imposant et un écran LED est présent façade pour afficher l’état du système, l’activité réseau ou encore certaines informations de monitoring.

Le S7 est construit autour d’un processeur Intel Core Ultra 7 356H (16 cœurs) capable d’atteindre 4,7 GHz, épaulé par un iGPU Intel Xe3. Ce dernier obtient un score de 34 066 points selon Passmark. L’architecture intègre également un NPU pouvant atteindre 50 TOPS.

La connectivité réseau confirme clairement les ambitions du produit :

• 2 ports 10 Gb/s SFP+ ;
• 1 port RJ45 10 Gb/s ;
• 1 port RJ45 2,5 Gb/s ;
• 2 ports USB4.

Sur le papier, le S7 coche pratiquement toutes les cases du NAS orienté virtualisation, IA locale et homelab haut de gamme.

MinisCloud OS et MinisOpenClaw : l’IA locale

Comme le modèle N5 Max, Minisforum met en avant son système MinisCloud OS et MinisOpenClaw (son agent IA maison dérivée d’OpenClaw). Reste toutefois une question essentielle : un agent IA ne sert à rien sans modèle derrière lui. Toute la problématique sera donc de savoir si ces NAS disposeront réellement de suffisamment de puissance pour faire tourner des modèles d’IA localement dans de bonnes conditions. A noter que le système MinisCloud OS serait installé sur une clé USB.

Si Minisforum parvient à proposer une expérience fluide et des cas d’usage pertinents, ces machines pourraient être intéressantes pour les utilisateurs souhaitant conserver leurs données en local tout en profitant des outils IA modernes. Sur le papier, la proposition est séduisante…. mais il faudra vérifier ce que cela donne en conditions réelles.

En synthèse

Minisforum poursuit son offensive sur le marché des NAS avec 2 modèles clairement orientés IA locale et hautes performances réseau. Les S5 et S7 misent sur des SSD NVMe, une connectivité 10 Gb/s et des processeurs Intel intégrant des NPU dédiés à l’accélération IA. Une approche cohérente avec l’évolution du marché vers des solutions capables d’exécuter des modèles localement, sans dépendance au Cloud.

Les prix et dates de disponibilité n’ont pas encore été dévoilés, mais ces NAS sont à surveiller…

source

Terramaster D1 Buyers Guide – D1 SSD vs D1 SSD PLUS vs D1 SSD Pro

TerraMaster is a storage hardware brand best known for NAS, DAS, and direct-attached expansion products aimed at home users, content creators, small offices, and prosumer workloads. While NAS devices remain the better choice for shared network storage, remote access, multi-user collaboration, and always-on services, DAS products are often the more practical option when the priority is direct speed, portability, simplicity, and lower setup complexity. An M.2 NVMe SSD enclosure such as the TerraMaster D1 range is designed to connect directly to a host system over USB or Thunderbolt, making it more suitable for fast local transfers, field backup, editing work, boot drives, and quick storage expansion without the overhead of network configuration. The D1 range is currently split across 3 tiers: the D1 SSD, D1 SSD Plus, and D1 SSD Pro. All 3 are diskless single-slot M.2 2280 NVMe SSD enclosures supporting up to 8TB drives, but they differ significantly in interface bandwidth, chassis design, target workload, and price. The standard D1 SSD is the rugged USB 10Gbps option, the D1 SSD Plus moves to USB4 40Gbps for higher transfer speeds, and the D1 SSD Pro is the Thunderbolt 5 / USB 80Gbps model aimed at the fastest single-drive workflows.

TerraMaster D1 SSD – The 10Gb/s Budget One!

The TerraMaster D1 SSD is the entry-level model in the D1 range, using a single M.2 2280 NVMe SSD slot and a USB 10Gbps interface. This places it in a different category from the Plus and Pro models, as it is not intended to chase the highest possible external SSD speeds. Instead, it is aimed more at users who need a compact, portable NVMe enclosure for everyday transfers, local backups, field storage, and general-purpose expansion.

TerraMaster lists maximum sequential performance at up to 1020MB/s read and 1010MB/s write when tested with a Samsung 990 Pro 4TB SSD, which is close to the expected practical limit of USB 3.2 Gen 2 storage.

The main point of difference with this model is the enclosure design. The D1 SSD uses a unibody aerospace-grade aluminum alloy shell and is rated at IP67 for dust and water resistance. TerraMaster also states that the enclosure can withstand up to 1.2 tons of pressure, which gives it a more rugged profile than a typical low-cost NVMe USB enclosure.

That makes it better suited to users who regularly move drives between locations, carry storage in a work bag, or need something for outdoor shoots, site work, travel, or emergency backups. It is still a single-drive enclosure, so it should not be treated as a complete backup strategy on its own, but the physical protection is clearly one of its main selling points.

Cooling is handled passively, with the metal body functioning as the heatsink rather than relying on a small internal fan. TerraMaster describes the design as offering 2.5x the usual heat dissipation area, and the lack of a fan also means there is no operating noise from the enclosure itself. The D1 SSD supports PCIe 3.0, 4.0, and 5.0 M.2 2280 NVMe SSDs up to 8TB, although the external USB 10Gbps connection will cap performance long before a modern high-end NVMe SSD reaches its own internal limits. File system support includes NTFS, APFS, Mac OS, FAT32, EXT4, and exFAT, while official operating system support is listed for Windows, macOS, and Linux.

---

TerraMaster D1 SSD Plus – The Perfect Middle!

The TerraMaster D1 SSD Plus is the middle option in the D1 range, moving from the USB 10Gbps connection of the standard D1 SSD to a USB4 40Gbps interface. This makes it a more suitable choice for users who want a compact single-drive NVMe enclosure but need substantially higher transfer speeds for larger files. TerraMaster positions this model around portable creative workflows, Mac mini storage expansion, macOS boot drive use, Windows workstation cache storage, and general high-speed external storage.

The company lists tested speeds of up to 3853MB/s read and 3707MB/s write, though the formal specification table gives 3641MB/s read and 3498MB/s write with a Samsung 990 Pro 4TB SSD.

The D1 SSD Plus keeps the same basic single-drive structure as the rest of the range, supporting 1 M.2 2280 NVMe SSD with PCIe 3.0, 4.0, or 5.0 compatibility and a maximum internal raw capacity of 8TB. The higher-speed USB4 connection makes a more obvious difference when working with large video files, project folders, image libraries, and other data sets where USB 10Gbps can become a bottleneck.

It is also compatible with Thunderbolt 5, Thunderbolt 4, Thunderbolt 3, USB4, USB 3.2, USB 3.1, and USB 3.0, although real-world performance will depend heavily on the host port, cable, SSD, file system, and operating system. TerraMaster notes that some Windows computers with Thunderbolt 3 may have incomplete USB4 support, which can lead to recognition problems or reduced speeds.

The enclosure uses a fanless aluminum alloy chassis, with TerraMaster claiming 3x the typical heat dissipation area for passive cooling. This means the drive remains silent in use, while the metal body handles heat transfer from the installed NVMe SSD. At 246g, it is heavier than the standard D1 SSD, but still portable enough for laptop bags, travel kits, and desk-to-desk workflows. Unlike the rugged standard D1 SSD, the Plus model is not presented around IP67 protection or crush resistance, with the emphasis instead placed on speed, broad protocol compatibility, quiet cooling, and electrical protection against short circuits, voltage surges, and electrostatic discharge.

---

TerraMaster D1 SSD Pro – The INSANE one

The TerraMaster D1 SSD Pro is the highest-end model in the D1 range, moving the enclosure platform up to USB 80Gbps and Thunderbolt 5 class connectivity. This places it well above the D1 SSD Plus in available interface bandwidth and makes it the model most clearly aimed at professional creative workloads rather than general portable storage.

TerraMaster positions the D1 SSD Pro for 8K video editing, large project transfers, direct-edit external workflows, and users who want the fastest option in the range for a single M.2 NVMe drive. The listed performance reaches up to 7061MB/s read and 6816MB/s write when tested with a Samsung 990 Pro 4TB SSD on a Mac M4 Pro system using AJA System Test.

Although the D1 SSD Pro supports PCIe 3.0, 4.0, and 5.0 M.2 2280 NVMe SSDs up to 8TB, the specification notes a single PCIe 4.0 x4 M.2 slot. This means the enclosure can physically and logically accept a wide range of modern NVMe drives, including examples such as the Samsung 990 Pro and WD SN850X, but performance will still depend on the SSD installed, the host system, and the connection used.

It is compatible with Thunderbolt 5, Thunderbolt 4, Thunderbolt 3, USB4, and USB 3.2/3.1/3.0, so it can step down for broader system support. TerraMaster also includes a color-coded connection indicator, using white for Thunderbolt or USB4 high-speed connections and orange for USB 3.2 or lower, which helps identify when the enclosure is not running at its intended performance level.

The chassis is built around an all-aluminum dual-fin passive cooling design rather than active cooling. TerraMaster describes the cooling system as using a built-in thermal pad, 38 fins, and a full aluminum-alloy shell to help control temperatures during sustained transfers or editing workloads. The enclosure remains fanless and silent, which may matter in audio-sensitive editing rooms or recording environments.

It also includes hardware protection against short circuits, voltage surges, and electrostatic discharge, along with a supplied protective case for travel. At 300g and 121 x 58 x 37mm, it is the largest and heaviest of the 3 models, but that size is tied to the higher-speed interface and more substantial passive heatsink design.

---

Backupper Software for PC/Mac/Mobile

TerraMaster’s D1 SSD range is also tied into the company’s backup software ecosystem, although the specific tools vary by platform. On Windows, the main utility is TPC Backupper, a free backup and synchronization tool for personal use. It can be used to back up a full operating system, entire disks, selected partitions, individual folders, application data, settings, and Microsoft Outlook emails. It supports full, incremental, and differential backup methods, along with file synchronization and disk or partition cloning. Backup destinations can include local disks, external USB storage, network shares, NAS systems, and cloud storage, which allows it to fit into a wider 3-2-1 backup routine rather than only working with TerraMaster hardware.

For mobile use, TerraMaster refers to the TDAS App for iOS and Android, allowing users to back up photos, videos, and other phone files to the connected private drive instead of relying only on cloud storage. The D1 SSD is listed as compatible with iOS 15.6 or later and Android 11 or later for this function. The D1 SSD Plus and D1 SSD Pro are also supplied with software-led backup features in their product positioning, though the clearest desktop backup support is on Windows through TPC Backupper. TerraMaster does not present the D1 range as a replacement for a NAS-based backup system, but the combination of portable NVMe storage and simple backup utilities gives the enclosures a role in local copies, travel backups, temporary project protection, and fast restores when connected directly to a host device.

Terramaster D1 SSD vs D1 SSD PLUS vs D1 SSD PRO – Which Is Best For You?

The TerraMaster D1 SSD is the most suitable option for users who want a low-cost, compact, and physically protected NVMe enclosure rather than the fastest possible external SSD. Its USB 10Gbps interface is enough for general file transfers, local backups, document libraries, photo collections, and moving data between systems, while the IP67-rated casing and crush-resistant design give it a stronger focus on travel, outdoor use, site work, and less controlled environments. For users who mainly need durable portable storage and do not work with very large media files every day, this is the most practical model in the range.

The TerraMaster D1 SSD Plus is the better fit for users who want a noticeable performance step up without moving to the cost of the Pro model. Its USB4 40Gbps interface makes more sense for content creators, photographers, video editors, Mac mini users, and laptop owners who regularly move large files or want fast external project storage. It lacks the rugged IP67 focus of the standard D1 SSD and the 80Gbps ceiling of the Pro, but it sits in the middle as the more balanced option for users who need high-speed storage in a portable, fanless enclosure.

The TerraMaster D1 SSD Pro is the model aimed at users with the most demanding workflows and the right host hardware to support it. Its USB 80Gbps and Thunderbolt 5 class positioning makes it the obvious choice for 8K video work, very large project files, high-speed scratch storage, and users who want the fastest enclosure in the D1 family. It is also the most expensive, largest, and heaviest of the 3, so it makes the most sense when the available bandwidth can actually be used. For everyday backup or standard portable storage, the lower models are easier to justify, but for performance-led creative work, the D1 SSD Pro is the clear top-tier option.

SUBSCRIBE TO OUR NEWSLETTER
[contact-form-7]
Join Inner Circle

This description contains links to Amazon. These links will take you to some of the products mentioned in today's content. As an Amazon Associate, I earn from qualifying purchases. Visit the NASCompares Deal Finder to find the best place to buy this device in your region, based on Service, Support and Reputation - Just Search for your NAS Drive in the Box Below

Need Advice on Data Storage from an Expert?

Finally, for free advice about your setup, just leave a message in the comments below here at NASCompares.com and we will get back to you. Need Help? Where possible (and where appropriate) please provide as much information about your requirements, as then I can arrange the best answer and solution to your needs. Do not worry about your e-mail address being required, it will NOT be used in a mailing list and will NOT be used in any way other than to respond to your enquiry. [contact-form-7] TRY CHAT Terms and Conditions   Alternatively, why not ask me on the ASK NASCompares forum, by clicking the button below. This is a community hub that serves as a place that I can answer your question, chew the fat, share new release information and even get corrections posted. I will always get around to answering ALL queries, but as a one-man operation, I cannot promise speed! So by sharing your query in the ASK NASCompares section below, you can get a better range of solutions and suggestions, alongside my own.

WE LOVE COFFEE

 

Le Minisforum N5 Max avait été présenté lors du CES 2026, mais on en sait désormais davantage à son sujet. Le constructeur a officialisé son prix ainsi que sa date de lancement. Avec son processeur AMD Ryzen AI Max+ 395, ce NAS à 5 baies s’annonce comme l’un des modèles les plus puissants jamais commercialisés sur le marché du stockage réseau. Voici ce qu’il faut retenir…

Minisforum N5 Max

Minisforum s’est imposé depuis 2018 comme un spécialiste reconnu des Mini PC performants. En 2025, la marque avait déjà fait une première percée sur le marché des NAS avec le N5 Pro, un modèle aux caractéristiques particulièrement ambitieuses.

Avec le N5 Max, le constructeur conserve le même châssis compact que les autres modèles de la gamme N5. L’appareil mesure 199 × 202,4 × 252,3 mm pour un poids d’environ 5 kg.

Ce nouveau boîtier embarque 5 baies SATA et 5 emplacements M.2 NVMe  (1 slot PCIe 4.0 x4 et 4 slots PCIe 4.0 x1). Un SSD de 128 Go préinstallé accueille le système d’exploitation et occupe l’un des emplacements NVMe.

Le Minisforum N5 Max est construit autour d’un processeur AMD Ryzen AI Max+ 395 (capable d’atteindre 5,1 GHz), épaulé par un iGPU Radeon 8060S, ainsi qu’un NPU affichant une puissance de 126 TOPS (tera-opérations par seconde). Le NAS serait livré avec 64 Go de mémoire LPDDR5x (non extensible).

Vous l’aurez compris, Minisforum positionne clairement ce modèle comme un véritable serveur local dédié à l’intelligence artificielle. L’objectif est de permettre l’exécution locale de modèles IA, sans dépendre d’infrastructures cloud externes. Selon le site PassMark, ce processeur obtient 55 141 points… un niveau de performances bien éloigné des configurations habituellement. À noter que le fabricant annonce un TDP de 55W.

Connectique

Le N5 Max ne fait aucun compromis côté connectivité. Le boîtier intègre notamment :

• 2 ports RJ45 10 Gb/s ;
• 2 ports USB4 à 80 Gb/s ;
• 1 port USB4 à 40 Gb/s ;
• 1 sortie audio-vidéo HDMI 2.1.

Une fiche technique qui vise clairement les usages professionnels avancés, le traitement de données massif ou encore les workflows vidéo lourds.

MinisCloud OS et MinisOpenClaw : l’IA locale au centre du projet

Le système maison MinisCloud OS intègre la plateforme MinisOpenClaw (OpenClaw préinstallé). Celle-ci propose un assistant IA local (images et vidéos déjà pris en charge, support de tous les fichiers prévu d’ici la fin du troisième trimestre 2026), une recherche sémantique, la gestion de snapshots ZFS, des machines virtuelles, l’isolation multicomptes et un contrôle de permissions en un clic (cette dernière devrait arriver un peu plus tard, d’ici fin 2026). L’objectif affiché est clair : offrir de l’IA générative et du traitement de données localement, sans dépendance au cloud.

Prix et disponibilité

Ce nouveau NAS vise les professionnels qui travaillent sur des fichiers lourds, les entreprises qui souhaitent exécuter des modèles d’IA locaux sans passer par un fournisseur cloud… ou encore les particuliers les plus exigeants.

Le N5 Max serait vendu 2 899 $ (prix en euro non dévoilé pour le moment) dans une configuration incluant 64 Go de RAM et le SSD système de 128 Go. Pour comparaison, le N5 Pro (Ryzen AI 9 HX PRO 370) démarre à 959€ et le N5 Air (Ryzen 7 255) débute à 559€.

Le N5 Max occupe donc clairement le sommet de la gamme, avec un écart tarifaire significatif justifié par une puissance de calcul sans équivalent dans ce segment.

Synology élargit son offre avec 2 nouvelles caméras IP : BC510 et TC510. Selon le fabricant, ces dernières sont conçues pour répondre aux besoins des entreprises et organisations à la recherche d’une solution flexible, intelligente et compatible avec différents environnements réseau. Mais une information, discrètement glissée dans leur fiche produit, mérite qu’on s’y arrête…

Synology BC510 & TC510

Ces nouvelles caméras viennent remplacer les BC500 et TC500 lancées en 2023. L’objectif ici est de proposer des produits combinant analyses IA, stockage et gestion dans le Cloud au sein d’une même architecture. Les BC510 et TC510 sont pensées pour fonctionner aussi bien avec l’environnement natif Synology qu’avec des systèmes tiers via le protocole standard ONVIF.

Sur le plan technique, les 2 modèles partagent les mêmes caractéristiques : capteur offrant une résolution de 2880 x 1620 px à 30 images par seconde, angle de vision horizontal de 110° et vision nocturne portant jusqu’à 30 mètres. Elles sont certifiées IP66 et IP67, garantissant une résistance élevée à la poussière, à la pluie et aux conditions extérieures difficiles. Des spécifications globalement similaires à celles des modèles précédents.

L’un des principaux arguments reste l’intégration de fonctions IA directement en périphérie, au niveau de la caméra elle-même. Cette approche permet de traiter les analyses localement plutôt que sur le serveur, avec pour objectif une meilleure réactivité et une réduction de la charge côté NAS.

Parmi les fonctionnalités annoncées : comptage de personnes et de véhicules, reconnaissance des plaques d’immatriculation, détection d’intrusion ainsi qu’Instant Search (système pour accélérer la recherche dans les archives vidéo). Ces outils s’adressent aux entreprises, commerces, sites industriels ou collectivités souhaitant automatiser une partie de leur surveillance sans multiplier les ressources matérielles.

Enfin, Synology anticipe déjà l’avenir en annonçant la compatibilité de ces modèles avec sa future plateforme de surveillance cloud VSaaS. Une façon de positionner les BC510 et TC510 comme des solutions évolutives, capables d’accompagner les besoins des organisations sur le long terme.

Le retour des licences

Contrairement aux précédentes générations de caméras du fabricant, les nouvelles Synology BC510 et TC510 nécessitent une licence Surveillance Station. Oui, vous avez bien lu… Jusqu’à présent, les caméras Synology intégraient directement leur licence, permettant une utilisation sans coût supplémentaire dans son écosystème de vidéosurveillance. C’était d’ailleurs un avantage concurrentiel clairement différenciateur.

Pour rappel, chaque NAS Synology inclut par défaut 2 licences caméra pour Surveillance Station. Au-delà, l’ajout de nouvelles caméras impose l’achat de licences supplémentaires. Concrètement, une installation comprenant 4 caméras nécessite l’achat de 2 licences additionnelles, les 2 premières étant fournies avec le NAS.

Jusqu’ici, les caméras Synology échappaient à cette règle grâce à une licence embarquée directement dans le matériel. Avec les BC510 et TC510 (et également la BC800Z), ce fonctionnement semble définitivement abandonné. Ces nouveaux modèles nécessitent désormais une licence dédiée pour être exploitée dans Surveillance Station, au même titre que des caméras tierces Axis, Reolink, Sony, Foscam, etc.

Pour l’heure, Synology n’a pas détaillé les raisons de ce changement. Difficile toutefois de ne pas y voir une évolution stratégique et commerciale autour de son activité de vidéosurveillance. Nous avons contacté le fabricant afin d’obtenir des éclaircissements. Cet article sera mis à jour dès que nous aurons de plus amples informations.

Prix et disponibilité

Les 2 caméras sont déjà disponibles à la commande. Il faudra débourser 299,95€ pour chacune des caméras (oui, elles sont au même prix). Un prix identique au lancement de la génération précédente.

Synology Cameras Now Need a License for Surveillance Station

Synology has changed the licensing position for selected cameras in its 2026 surveillance camera range, with the newly listed BC510, TC510, and BC800Z now requiring a Surveillance Device License when used with Synology Surveillance Station. This marks a notable shift from the previous value proposition of Synology-branded cameras, which had been positioned as tightly integrated first-party devices that did not require an additional camera license. The new BC510 and TC510 have been introduced as AI-enabled bullet and turret cameras with 5MP resolution, 30 FPS recording, 110° horizontal field of view, IP66/IP67 protection, 30 m night vision, edge AI analytics, ONVIF support, and compatibility with Synology’s wider surveillance ecosystem, including its upcoming cloud-based surveillance platform. But why has Synology changed it’s stance on camera license requirements with this new series?

What Changed in Synology Cameras and the License Requirements?

Synology’s camera licensing policy has changed for part of its 2026 camera generation. The BC510, TC510, and BC800Z are now listed by Synology as requiring a Surveillance Device License, with Synology’s license documentation stating that these models require 1 license per camera. This means these Synology-branded cameras are now treated in the same basic licensing structure as regular IP cameras, where each camera consumes 1 available Surveillance Station camera license.

This is a significant change because Synology’s own cameras previously had a clear licensing advantage inside Surveillance Station. Earlier Synology camera models were positioned as first-party devices that worked directly with the platform without the need to buy an additional camera license. For users building a Synology-based surveillance setup, that made the cameras easier to justify even when comparable third-party ONVIF cameras were available at lower prices. The camera, platform integration, AI features, and license position were effectively part of the same value proposition.

With the 2026 generation, that arrangement has changed for the BC510, TC510, and BC800Z. Users will now need to account for the cost of a Surveillance Device License when deploying these cameras beyond the default licenses included with their Synology NAS, NVR, or DVA system. Synology NAS systems generally include 2 default licenses, Network Video Recorder systems include 4, and Deep Learning NVR systems include 8. Any deployment that exceeds the available default license count will require additional license packs, just as it would when adding third-party IP cameras or other supported surveillance devices.

Device Type		License Units	Example	License required
Synology Cameras		Per camera	BC510, TC510, BC800Z	1
		Per camera	BC500, TC500	0
Synology LiveCam		Per device	Synology LiveCam app	1
Regular IP camera		Per camera	AXIS P1347	1
Panoramic (fisheye)		Per camera	AXIS M3007	1
Multi-lens	Fixed lens	Per camera	ArecontVision AV8185DN	1
	Fixed lenses with independent IP	Per channel	AXIS Q3709-PVE	3
	Removable lens	Per channel	AXIS F44	5
Video server		Per channel	Vivotek VS8801	8
I/O module		Per device	AXIS A9188	1
Intercom		Per device	AXIS A8105-E	1
IP speaker		Per device	AXIS C3003-E	1
Access controller (door)		Per device	AXIS A1001	1
Transaction device (POS)		Per device	–	2

According to Synology’s stated position around the new generation, the decision is connected to broader deployment flexibility. The BC510 and TC510 are being introduced not only as cameras for Surveillance Station, but also as devices designed to work across multiple surveillance environments. Synology states that these cameras support deployment within the native Synology ecosystem, third-party NVR and VMS infrastructures through ONVIF, and its upcoming cloud-based surveillance platform. In that context, Synology appears to be separating the camera hardware from the Surveillance Station license entitlement, rather than treating the license as implicitly bundled with the camera.

The advantage Synology presents is that this approach allows the cameras to be used more flexibly outside Synology-only deployments. In theory, a lower hardware price can reduce the entry cost for users who want to deploy the cameras in third-party systems, where a Synology Surveillance Station license would not be relevant. For mixed environments, installers, managed service providers, or businesses migrating between platforms, the cameras can be positioned as ONVIF-capable AI cameras rather than hardware tied primarily to a Synology NAS or NVR. Synology’s argument is therefore less about removing value from Surveillance Station users, and more about aligning the cameras with wider interoperability, third-party infrastructure support, and future cloud surveillance services.

Which Cameras are Affected, and What About Older Synology Cameras?

The affected 2026 Synology camera models listed as requiring a Surveillance Device License are the BC800Z, BC510, and TC510. The BC800Z is the higher-end 8MP model with PoE connectivity, optical zoom coverage, longer night vision range, IP66/IP67/IK10 protection, a 5-year warranty, and additional analytics such as License Plate Recognition and Smoke Detection. The BC510 and TC510 are 5MP PoE cameras, offered in bullet and turret designs respectively, with 2880×1620 resolution, 30 FPS video, a 110° horizontal field of view, 30 m night vision, people and vehicle detection, intrusion detection, audio detection, tampering detection, motion detection, people and vehicle counting, Instant Search, and people-based auto tracking. The CC400W is not listed as requiring a Surveillance Device License, and remains separate from the licensing change affecting the BC800Z, BC510, and TC510.

At this stage, the licensing change appears to apply to the newer 2026 generation models listed by Synology, rather than being presented as a wider retrospective change across all previous Synology cameras. Older Synology camera models are less prominent on Synology’s current product pages following the arrival of the refreshed range, so the long-term public positioning of those older models is less clear from the current camera comparison material. Based on the available details, there is no indication in the supplied information that previously released Synology cameras are being newly reclassified in the same way, but buyers and existing users should still check the official Synology Camera Support List and license documentation for their exact model before expanding or changing a deployment.

Why Has Synology Made This Decision?

Synology’s stated reasoning appears to centre on making its newer cameras more flexible across different deployment environments. The BC510 and TC510 are being positioned not only as Surveillance Station cameras, but also as cameras for third-party NVR and VMS systems through ONVIF, as well as Synology’s upcoming cloud-based surveillance platform. By separating the camera hardware from the Surveillance Station license entitlement, Synology can sell the cameras into environments where a bundled Surveillance Station license would not be useful, while also lowering the hardware entry price for users who are not deploying them directly with Synology’s own platform. There may also be a wider commercial consideration around Synology’s position as a Taiwanese camera manufacturer. In some government, education, public sector, and official institutional deployments, the country of origin of surveillance hardware can be a factor in procurement, security review, and long-term platform approval.

This may give Synology an advantage over some Chinese-made camera brands, particularly in environments where hardware from certain vendors is harder to approve or deploy. In that context, Synology may see an opportunity to position the BC510, TC510, and BC800Z as more broadly deployable surveillance cameras for institutions that want ONVIF-compatible hardware without relying on brands that may face additional scrutiny. For Synology-only users, however, the practical result is different: the license cost now needs to be considered separately when adding the BC510, TC510, or BC800Z to a deployment that has already used its default license allowance. This does not remove the cameras’ first-party integration benefits, edge AI features, or official support inside the Synology ecosystem, but it does change the overall value calculation compared with older Synology cameras that did not require a separate Surveillance Device License.

SUBSCRIBE TO OUR NEWSLETTER
[contact-form-7]
Join Inner Circle

This description contains links to Amazon. These links will take you to some of the products mentioned in today's content. As an Amazon Associate, I earn from qualifying purchases. Visit the NASCompares Deal Finder to find the best place to buy this device in your region, based on Service, Support and Reputation - Just Search for your NAS Drive in the Box Below

Need Advice on Data Storage from an Expert?

Finally, for free advice about your setup, just leave a message in the comments below here at NASCompares.com and we will get back to you. Need Help? Where possible (and where appropriate) please provide as much information about your requirements, as then I can arrange the best answer and solution to your needs. Do not worry about your e-mail address being required, it will NOT be used in a mailing list and will NOT be used in any way other than to respond to your enquiry. [contact-form-7] TRY CHAT Terms and Conditions   Alternatively, why not ask me on the ASK NASCompares forum, by clicking the button below. This is a community hub that serves as a place that I can answer your question, chew the fat, share new release information and even get corrections posted. I will always get around to answering ALL queries, but as a one-man operation, I cannot promise speed! So by sharing your query in the ASK NASCompares section below, you can get a better range of solutions and suggestions, alongside my own.

WE LOVE COFFEE

 

UGREEN NASync Celebrates 2 Years – But Is UGREEN Ready for the Big Leagues?

Two years ago, the popular battery and PC accessory company UGREEN, launched their Kickstarter campaign for the NASync personal NAS series of devices. The brand already had a steadily growing foothold in China with their DX series of NAS devices, but were still a huge outsider in the world of NAS globally. Fast forward two years, a $6 Million crowdfunding campaign, 6 new NAS releases, a new NAS kickstarter in progress (the IDX6011 AI NAS) and generally undermining long-time players who have been in the NAS market for more than a decade – UGREEN is looking like quite a beast in the world of NAS! But two years, UGREEN now finds that along with an increased market position also comes increased demand, scrutiny and expectation. I went to Shenzhen, China, to speak directly with the teams who direct and create their NASync division to ask them them questions about the development of this series, lessons that were learnt, where they are going and what they still need to do to further establish their position in the turnkey NAS market.

Full Disclosure – this Q&A has NOT been sponsored, subsidised or creatively controlled by UGREEN. These questions are my own, submitted to UGREEN 48 hours prior to the interview, and the answers provided were directly from their team.

UGREEN was already a well-established company in its own right before it expanded into NAS systems. So, currently, what is the scale of the teams and resources that your company has allocated to this? R&D, Design, Development, Technical Support, etc?

UGREEN put together its NAS team back in 2018, released its first NAS product in China in 2021, and went global for the first time in 2024, bringing its NAS products to markets around the world. NAS is one of the company’s key strategic product lines, with a team of several hundred people working on it—including product, R&D, design, testing, security, and more. This doesn’t count shared support teams like industrial design, legal, or finance; we’re only talking about people directly focused on NAS. In this whole building, every floor is filled with NAS team members—except for the third floor, which is just the cafeteria.

What has been the biggest challenge in the continued development of your NASync/UGOS services in these last 2 years?

One challenge is resource allocation. We need to support international users at the same time, which means balancing different priorities and expectations. Another challenge is localization. It’s not just about language, but also understanding different user behaviors and usage scenarios. So we had to spend a lot more time to research and validate what users actually need in each region. Based on that, we’ve been continuously adjusting our product direction and improving UGOS to better fit a global audience. It’s definitely an ongoing process, but it’s helped us build a much clearer understanding of the market.

Two years on from your initial crowdfunding, your position in the ‘turnkey NAS market’ from comparative obscurity has catapulted to effectively being in the top 5 (if not top 3) – What do you think UGREEN have brought to the market (or change in the market as a whole) that caused this?

There are a few key things behind that.

Hardware DNA, Built for AI

UGREEN is a hardware company at heart. With our NAS products, we insist on solid hardware—high-performance CPUs, ample memory, high-performance CPUs and ample memory—not just for reliability, but for computing power. AI NAS demands serious performance. Without a strong hardware foundation, AI is nothing more than a concept. Our hardware is designed to make AI run stable and fast.

User-Centric, Not Just a Slogan

We’ve always put ourselves in our users’ shoes. We listen to every voice—on social media, in forums, through user interviews. Many of our features, like snapshots and SAN Manager, came directly from users telling us, “I need this.” Our products aren’t built in a vacuum; they’re shaped together with you.

R&D Investment, Bringing NAS to Everyone

We established our software R&D team, including an AI pre-research team, early on. User feedback has driven us to keep investing, with one goal in mind: to shorten the learning curve. NAS shouldn’t be just a toy for tech enthusiasts. We believe the future of NAS is for everyone—simple, smart, and accessible. This is the path we’re on, and it’s one we want to walk together with you.

The UGREEN IDX6011 AI NAS series has been in development for a long time, and will be headed to its own crowdfunding campaign shortly. What was the biggest challenge you faced in its development and/or lesson that you learned about this new profile of solution?

The biggest challenge was finding the right balance between AI capabilities and real user value. It’s relatively easy to add AI features from a technical perspective, but making them actually useful, stable, and well-integrated into everyday workflows is much harder. Especially on a NAS, NAS is essentially a local storage product, everything runs locally, so for us, it was important that AI features also run locally. But hardware resources and compute power are limited. So the question is, how do we build useful and stable AI features without affecting NAS core functionality like storage, backup and overall system performance? That’s very difficult.

And from product design perspective，it’s also challenging to define the right AI use cases. It’s not about adding more AI features, but making sure they are scenario-driven and actually solve real problems, like better file organization, smarter search, easier intraction… We need to keep the experience simple. Many users are still new to AI on a NAS, so we wanna make things natural and do not add extra complexity. So right now, we’re still in the process of refining and validating these ideas, and making sure we deliver something that’s both practical and reliable for users.

I think it would be fair to say that UGREEN has chiefly focused on Desktop NAS server ownership in their portfolio of solutions to date. But have you explored rackmount solutions, and/or is this something that could happen in the near future?

Actually we’ve done some internal research on rackmount solutions, it’s quite different from desktop NAS in terms of target audience, hardware, software and sales. So it’s not just an extension of what we’re doing now, it requires a different product strategy. For now, our focus is still on improving and expanding our desktop NAS lineup, in the short term, we don’t have a concrete plan for rackmount products, but it’s something we’ll continue to evaluate over time.

Now that UGREEN is a largely established player in the turnkey NAS market, there is a lot more scrutiny on the extent to which your brand preemptively prepares against cyber security threats. What is UGREEN doing to address (in terms of foundations on this platform and broader services) this to avoid a potential slow moving snowballing security incident?

Security is something we take very seriously. At the product level, we provide a range of built-in security features. For example, users can enable DoS protection to defend against network attacks, automatically block IP addresses after multiple failed login attempts, and set up firewall rules to control access from specific IP addresses and built-in Security app to scan for suspicious files We also have a dedicated internal security team and a well-established vulnerability handling process, so critical issues can be identified and resolved quickly. We also provide a vulnerability reporting channel on our official website. If users discover any potential security issues, they can report them to us, and we will assess and respond accordingly.

(Below is a snippet of the Security Disclosure page from UGREEN, available HERE)

At CES 2026, UGREEN unveiled its surveillance platform and edge AI cameras. At that time, your team was kind enough to allow me to see the early development of your NAS surveillance application that will allow full management, direct control and storage of these new Surveillance services. Is development still continuing on this and will this be a service that existing NASync owners will have access to in the future

Yes, development is still ongoing. The surveillance platform you saw at CES is actually a part we are exploring, and is still under active development. From what I understand, AIOT is a broad ecosystem that is going to feature ai NAS, ai-based home security cameras, and many more AI-empowered hardware devices for a smarter lifestyle. On the NAS side, we’re also building our own surveillance application for NASync. We plan to launch it within this year. In terms of compatibility, we aim to support both UGREEN cameras and third-party cameras, so users have more flexibility to build their setup. So overall, both sides are moving forward, but they are different products within the UGREEN ecosystem.

In the last year, UGREEN released two ARM RK chip-powered NAS solutions in the DH2300 and DH4300 – How easy/hard was scaling UGOS onto this more modest hardware base, and were there any useful lessons learned that have benefited your NAS development as a whole?

DH series is our entry-level lineup, designed mainly for NAS beginners and users with simpler needs. From a technical perspective, running UGOS on an ARM-based platform is definitely more constrained compared to x86, especially in terms of performance and resource availability. A lot of things can’t just be directly carried over, we need to re-adapt them for the ARM architecture, including the kernel, system services, and many core features. So we had to be more selective and thoughtful about which features to include and how to optimize them. And from a product perspective, it actually helped us become more focused. With the DXP series, we already emphasized user-friendliness, but with the DH series, we really wanted to take that further and make it as simple as possible, essentially positioning it as a user’s first NAS.

So in practice, we streamlined certain features based on the hardware and target users. For example, we simplified or did not include things like virtual machines and some AI capabilities, and instead focused on delivering a smooth and reliable core experience. One key lesson we learned is that not every product needs to do everything. It’s more important to match the right experience to the right user group. And that thinking has also helped us better define our overall NAS product lineup.

I canvased a large group of UGREEN NAS users (many of whom were part of your original Kickstarter campaign) who are still using their NASync systems to this day, and have followed you on your journey so far. I asked them which features or improvements they would like to see in future updates and revisions to UGOS. Are you able to share if these are features that are on the roadmap, or have been explored?

• Full Volume Encryption
• WORM support
• A mixed drive RAID storage system (comparable to Synology Hybrid RAID or Terramaster TRAID)
• A tiered storage system (unlike the copy system of ‘caching’, but a SSD+HDD composite pool that intelligently moves ‘hot’, ‘warm’ and ‘cold’ data to appropriate storage areas)
• ZFS as a file system choice
• A native Plex Media Server Application
• A local client application for Mac/Windows for file pinning, streaming, intelligent 30-day deletion (see Synology Drive, QNAP Qsync, etc)
• A more comprehensive security scanner (eg scanning for unsecure open ports, SSH being open, weak passwords, admin accounts, auto blocks disabled, etc)

 

We’ve actually seen many of these requests from our users as well, these are very valuable suggestions and we’ve already had internal discussions around most of them. But many of these features, like full volume encryption, hybrid RAID, or tiered storage are quite complex, they take time to design, develop and validate, especially we wanted to make them stable and reliable.

So at this stage, we don’t have a specific timeline we can share yet. But these are definitely things we take seriously, and we’ll plan them carefully based on user demand and overall product direction. If we see strong demand from users, we’ll absolutely prioritize them accordingly.

Thank you to the team at UGREEN for their time in this interview. As mentioned, the answers about were provided in their entirety and without prejudice. This will be a video soon that covers this, the tour of the facilities, as well as further discussion around the IDX6011 NAS Kickstarter and how this has been managed.

 

Want to Support the work me and Eddie do at NASCompares? If you found this article helpful and are going to buy a UGREEN NAS from the brand’s official site or from Amazon, use the links below. Using these links will result in a small commission coming to us (which costs you nothing extra) and it allows us to keep doing what we do! Thank you for keeping the internet a fair and sustainable place!

Buy the UGREEN NAS from your local Amazon Store	Buy the UGREEN NAS solutions Directly from UGREEN.COM

 

 

SUBSCRIBE TO OUR NEWSLETTER
[contact-form-7]
Join Inner Circle

This description contains links to Amazon. These links will take you to some of the products mentioned in today's content. As an Amazon Associate, I earn from qualifying purchases. Visit the NASCompares Deal Finder to find the best place to buy this device in your region, based on Service, Support and Reputation - Just Search for your NAS Drive in the Box Below

Need Advice on Data Storage from an Expert?

Finally, for free advice about your setup, just leave a message in the comments below here at NASCompares.com and we will get back to you. Need Help? Where possible (and where appropriate) please provide as much information about your requirements, as then I can arrange the best answer and solution to your needs. Do not worry about your e-mail address being required, it will NOT be used in a mailing list and will NOT be used in any way other than to respond to your enquiry. [contact-form-7] TRY CHAT Terms and Conditions   Alternatively, why not ask me on the ASK NASCompares forum, by clicking the button below. This is a community hub that serves as a place that I can answer your question, chew the fat, share new release information and even get corrections posted. I will always get around to answering ALL queries, but as a one-man operation, I cannot promise speed! So by sharing your query in the ASK NASCompares section below, you can get a better range of solutions and suggestions, alongside my own.

WE LOVE COFFEE

 

Je viens tout juste de rentrer de vacances. Nous sommes partis en Sicile avec des amis, en explorant principalement l’Est de l’île : Syracuse, Taormine, l’Etna, etc. Un décor idéal pour tester un nouvel équipement que j’avais emporté dans mes bagages : le drone DJI NEO 2. Et clairement, je ne regrette pas mon choix…

Prise en main Neo 2

Je ne vais pas vous inonder de photos ou de vidéos ici. L’objectif est plutôt de partager mon retour d’expérience.

Il y a quelques années, j’avais acheté un drone Parrot en 2013. Je l’utilisais peu et je l’ai finalement revendu après 2 ans. Depuis, le marché a bien évolué… Les drones sont devenus bien plus accessibles et simples à utiliser. J’ai toujours été attiré par les images impressionnantes (vues aériennes) qu’ils permettent de prendre… mais je n’avais pas envie de passer des heures en réglage pour prendre une photo ou une vidéo. Mon besoin était simple : obtenir quelques prises de vue en hauteur facilement.

J’ai profité d’une promotion pour acheter le DJI NEO 2 à environ 200 €. Ce qui m’a attiré, c’est sa promesse de fonctionner de manière quasi-autonome.

Plusieurs modes de pilotage

Le drone peut être utilisé de plusieurs façons : via les boutons intégrés sur le côté (voir photo), l’application mobile, des commandes vocales et même avec des gestes de la main. Il propose une douzaine de modes automatiques (cercle, boomerang, rocket, suivi…), qui permettent de capturer facilement des séquences propres et dynamiques. Grâce à son LIDAR embarqué, il peut détecter et éviter les obstacles. Aussi, il suffit de présenter sa main, paume ouverte vers le haut pour que le drone viennent se poser. Sa vitesse maximale peut atteindre 12m/s (43,2km/s) !

À noter : il est compatible avec plusieurs radiocommandes DJI et casques FPV, à condition d’ajouter un émetteur-récepteur à l’arrière du drone (environ 20 €).

Comme vous pouvez le constater, il s’agit d’un petit drone, qui ne rivalise pas avec un modèle à plus de 1 000 €. Ce n’est pas l’objectif. Je cherchais un appareil léger (151 g), compact (147 × 171 × 41 mm) et accessible à toute la famille. Il dispose d’une mémoire interne de 49 Go.

En moins de 5 minutes (et avec quelques explications), ma fille cadette était capable de l’utiliser seule.

Qualité des photos et vidéos

Je n’ai utilisé le drone qu’en plein jour (pas testé la nuit). Par défaut, il est paramétré pour capturer des photos de 12 Mpx (format JPEG) et des vidéos 4K en 60 images par seconde (il peut monter jusqu’à 100 IPS). La plage ISO s’étend de 100 à 12 800 selon le fabricant, ce qui laisse à penser qu’il peut être polyvalent avec peu de luminosité.

Pour un usage en plein jour, les résultats sont franchement très très bien.

Autonomie : le point faible

L’autonomie est probablement le principal point faible de l’appareil. Sur le papier, on pourrait souhaiter mieux. En pratique, cela ne m’a pas posé de problème. Mes séquences n’ont jamais dépassé trois à quatre minutes : j’activais un mode et lançais le drone… il revenait et je le rangeais. Comptez environ 15 minutes en vol en tout sur une charge.

Pour le voyage, j’avais prévu une seconde batterie et une boîte de rangement. L’ensemble tenait sans problème dans mon sac Eastpak, à côté d’autres affaires. J’avais également prévu une batterie externe pour recharger (au cas où), mais je n’en ai pas eu besoin.

Mes accessoires indispensables pour voyager léger (sac, batterie, chargeurs…)

Attention au vent

Sur le terrain (que ce soit sur l’Etna ou en bord de mer), je n’ai rencontré aucun souci lié au vent. Mais il a des limites ! Selon le fabricant, sa résistance maximale est de 10,7 m/s (force 5 sur l’échelle de Beaufort).

En synthèse

Non, je n’ai pas prévu de devenir influenceur spécialisé dans les drones. Cet achat répond à un usage simple et efficace. Oui, c’est un gadget. Non, ce n’est pas un outil professionnel. Et non, il ne remplacera jamais un drone haut de gamme. En revanche, il coche toutes les cases pour un usage grand public : léger, simple, rapide à prendre en main, avec une qualité photo et vidéo très satisfaisante pour ce segment de prix.

Depuis quelque temps, un mouvement de fond s’observe : un retour progressif vers les infrastructures locales. La raison ? Même s’il a de nombreux atouts, le Cloud ne coche plus toutes les cases dès que l’on parle de confidentialité des données, de latence ou simplement de coûts. C’est sur ce créneau que QNAP positionne son QAI-h1290FX. Un serveur de stockage pensé pour les charges de travail IA : LLM, architectures RAG, inférence en temps réel. Ici, il ne s’agit pas d’un simple NAS avec un logo IA collé dessus…

QNAP QAI-h1290FX

Le QAI-h1290FX est un boitier 12 baies SSD U.2 NVMe/SATA. Pas de disques rotatifs ici, on est clairement dans le registre des IOPS élevées, indispensables pour alimenter des pipelines de données intensifs ou soutenir l’inférence en temps réel sans créer de goulot d’étranglement côté stockage. Il est animé par un AMD EPYC 7302P (16 cœurs / 32 threads) pouvant atteindre 3,3 GHz. A noter que ce processeur a obtenu 32 114 points selon PassMark. Ce dernier est épaulé par 128 Go de RAM RDIMM DDR4 ECC extensible jusqu’à 1 To (8 × 128 Go)

Le QAI-h1290FX n’est pas un NAS comme les autres. Il est compatible avec les cartes NVIDIA RTX, notamment la RTX PRO 6000 Blackwell Max-Q, embarquant jusqu’à 96 Go de VRAM. Une capacité mémoire GPU qui change la donne pour quiconque veut faire tourner des LLM de taille respectable en local. La prise en charge de CUDA, TensorRT et du Transformer Engine vient confirmer l’orientation IA-first de la machine. On n’est pas sur un gadget, mais un outil capable d’accélérer des modèles de deep learning, de génération d’images ou de traitement du langage naturel.

Connectique

La connectivité est à la hauteur des ambitions du boîtier :

• 3 ports USB 3.0
• 2 ports réseau 2,5 Gb/s
• 2 ports réseau 25 Gb/s en SFP28

A noter la présence de 4 emplacements PCIe (3* Gen 4 x16 et 1* Gen 4 x8)

QuTS hero et l’écosystème logiciel

Côté système, on est sur du QuTS hero, basé sur ZFS. On retrouve les fonctionnalités attendues pour un usage professionnel : déduplication, snapshots, intégrité des données. Rien de révolutionnaire pour les habitués de la gamme… Container Station et Virtualization Station permettent de gérer des environnements bénéficiant d’un accès direct au GPU, ce qui permet aux équipes de déployer des modèles sans friction et sans reconfiguration complexe.

Plusieurs outils populaires dans l’écosystème IA open source sont préinstallés :

• AnythingLLM, OpenWebUI, Ollama : pour monter rapidement un LLM privé ;
• vLLM* : moteur d’inférence LLM ;
• Stable Diffusion*, ComfyUI* : pour la génération d’images ;
• n8n* : pour l’automatisation et les workflows sans code.

C’est une approche « prêt à l’emploi » qui tranche avec les serveurs IA nus que l’on retrouve chez certains concurrents.

En synthèse

Le QNAP QAI-h1290FX est un serveur de stockage conçu de bout en bout pour répondre aux besoins d’IA on-premise. L’alliance d’un stockage full-flash NVMe, d’un processeur EPYC et d’une compatibilité GPU NVIDIA en fait une plateforme intéressante pour les entreprises qui souhaitent reprendre la main sur leur stratégie IA (sans dépendre du cloud ou exposer leurs données).

Les logiciels faciles à installer (Ollama, OpenWebUI, n8n…) abaissent la barrière à l’entrée, ce qui est un point fort pour les équipes IT non spécialisées. Reste à connaître son prix et la date de disponibilité…

* en cours d’intégration

Sharge Go Back to the Drawing Board, with the Disk Pro 2

The Sharge Disk Pro 2 is an upcoming portable storage and connectivity device that combines the functionality of a USB hub with external SSD support in a compact, credit card-sized form factor. Developed by Sharge, the device is designed to address the increasing demand for high-speed data access, external storage expansion, and multi-port connectivity across mobile and desktop platforms. Unlike conventional USB-C hubs or portable SSDs, the Disk Pro 2 merges both roles into a single unit, while also incorporating active cooling to maintain consistent performance under sustained workloads. At launch, it will be available in two variants, Lite and Ultra, which differ in display capability and power efficiency, introducing a tiered approach not seen in the previous model.

Positioned as a follow-up to the earlier Sharge Disk Pro, this new iteration shifts away from fixed internal storage and instead introduces support for user-installed SSDs in multiple M.2 form factors. Alongside this change, the device retains key characteristics such as 10Gbps data throughput, integrated power delivery, and video output capabilities, while adding refinements including magnetic attachment and a lanyard-style data cable. The Lite version features HDMI 2.0 and a higher power draw of around 4W, while the Ultra version includes HDMI 2.1 and operates at approximately 1W, providing a more efficient option with expanded display support. The Disk Pro 2 is scheduled to launch via Kickstarter, continuing the company’s established approach of introducing new hardware through crowdfunding platforms.

Sharge Disk Pro 2 – Design & Storage

The Sharge Disk Pro 2 maintains a compact footprint, measuring approximately 90 × 61 × 11 mm, aligning closely with the dimensions of a standard credit card. This size places it firmly in the category of ultra-portable accessories, designed to be carried alongside a smartphone or laptop without requiring additional space typically associated with external drives or docking stations. The chassis follows a flat, rectangular layout with integrated components distributed to maximize internal efficiency while preserving a slim profile. A defining aspect of the design is its transparent enclosure, which exposes internal components in a style often associated with “cyberpunk” aesthetics. This approach is not purely cosmetic, as it also highlights the inclusion of active cooling hardware within a device of this size. The visible fan and internal layout reinforce the product’s positioning as a performance-oriented device rather than a passive accessory, distinguishing it from more conventional sealed USB hubs. The external design remains consistent across both Lite and Ultra variants, with no physical differentiation beyond internal configuration.

The Disk Pro 2 introduces a magnetic mounting system intended for direct attachment to compatible devices. This includes native support for MagSafe-enabled smartphones, as well as the option to use included magnetic rings for broader compatibility with non-MagSafe hardware. The goal is to reduce cable strain and improve portability by allowing the hub and connected device to function as a single unit during use, particularly in mobile workflows such as handheld video capture or on-the-go file transfers. Another physical design element is the inclusion of a detachable lanyard-style cable that supports both data and power delivery. This integrated approach removes the need for users to carry separate cables for connectivity, while also doubling as a carrying mechanism. The included cable is specified as a 24-pin pure copper design, supporting up to 10Gbps data transfer, power delivery, and DisplayPort signal passthrough.

In terms of storage, the Disk Pro 2 departs from the fixed-capacity approach of the earlier Sharge Disk Pro. Instead of pre-installed flash memory, it supports user-installed M.2 SSDs in 2230, 2242, and 2280 form factors, with a maximum supported capacity of up to 8TB. This change introduces flexibility in both capacity selection and potential future upgrades, allowing users to tailor storage based on their requirements rather than being limited to predefined configurations. The choice between Lite and Ultra models does not affect storage compatibility, with both versions offering the same SSD support and expansion capabilities.

Sharge Disk Pro 2 – Internal Hardware

At the core of the Sharge Disk Pro 2 is a multi-controller architecture described as an independent 4-chip control system. Each major function, including storage access, USB expansion, video output, and power delivery, is handled by a dedicated controller. This separation is intended to improve stability and reduce bandwidth contention when multiple ports are in use simultaneously, particularly under sustained workloads such as file transfers while outputting video and supplying power. A central feature of the internal design is the active cooling system, referred to as the “Ice-storm” fan. Operating at speeds of up to 10,000 RPM, the fan is designed to maintain consistent thermal conditions during extended data transfers. The system includes three operational modes: OFF, Auto, and Turbo. In Auto mode, fan speed adjusts based on internal temperatures, while Turbo maintains maximum cooling performance. This approach addresses a common limitation in compact hubs and SSD enclosures, where passive cooling can lead to thermal throttling under load. The cooling system is consistent across both Lite and Ultra variants, with no differentiation in thermal hardware between the two models.

The storage interface supports M.2 NVMe SSDs across multiple physical formats, with a maximum capacity of up to 8TB. Data transfer is handled over a 10Gbps USB interface, setting an upper limit on throughput but aligning with typical USB 3.2 Gen 2 performance expectations. The combination of active cooling and dedicated controllers is intended to sustain transfer speeds closer to this ceiling over longer periods, rather than allowing performance to degrade as temperatures increase. Differences between the Lite and Ultra versions are not related to storage or controller design, but instead focus on power efficiency and display output, meaning internal data handling performance should remain consistent regardless of variant selection.

Sharge Disk Pro 2 – Ports & Connections

The Sharge Disk Pro 2 integrates a total of 6 ports, combining data transfer, display output, and power delivery within a single device. These include 2 USB-C ports, 1 USB-A port, 1 HDMI output, and dual card reader slots for SD and microSD media. This configuration positions the device as a compact alternative to larger desktop docking stations, while maintaining compatibility with a wide range of peripherals and storage formats. The primary USB-C interface (USB-C1) supports 10Gbps data transfer alongside up to 100W power input, allowing the connected host device to be charged while the hub is in use. A secondary USB-C port (USB-C2) provides up to 80W power output, enabling downstream charging for connected devices. The inclusion of both input and output power delivery allows the hub to function as an intermediary between a power source and multiple connected devices without interrupting data throughput. This overall port layout remains consistent across both Lite and Ultra variants.

Video output capabilities differ between the two versions. The Ultra model includes HDMI 2.1, supporting resolutions up to 4K at 144Hz or 8K at 30Hz, depending on the host system and display compatibility. In contrast, the Lite version is equipped with HDMI 2.0, which reduces maximum output capabilities accordingly. Outside of this distinction, additional connectivity is provided through a USB-A 3.0 port operating at up to 5Gbps, alongside SD and microSD card slots with rated read speeds up to 180MB/s and write speeds up to 120MB/s. The included lanyard cable also functions as a full-featured USB-C connection, supporting 10Gbps data transfer, up to 100W power input, and DisplayPort signal transmission, reducing reliance on separate cables during use.

Sharge Disk Pro vs Sharge Disk Pro 2

The transition from the original Sharge Disk Pro to the Sharge Disk Pro 2 represents a shift in both hardware architecture and product segmentation. The Disk Pro is fundamentally an all-in-one device, combining fixed internal NVMe storage with a compact multi-port hub and active cooling, positioned as a self-contained solution for users who want storage and connectivity without additional components. It integrates storage capacities up to 4TB and was originally sold in tiered pricing depending on capacity . In contrast, the Disk Pro 2 removes onboard storage entirely and instead supports user-installed M.2 SSDs up to 8TB, changing the device into a modular enclosure and hub hybrid rather than a pre-configured storage product. This also alters the pricing structure significantly, as the Disk Pro 2 is sold as a standalone unit starting at $49 for the Lite version and $69 for the Ultra version, separating the cost of storage from the hardware itself.

Beyond storage, the Disk Pro 2 introduces clearer product tiering with Lite and Ultra variants, something not present in the original model. The Lite version reduces cost by using HDMI 2.0 and operating at a higher power draw of around 4W, while the Ultra version includes HDMI 2.1 and lowers power consumption to approximately 1W. Both retain the same core concept of combining data transfer, display output, and power delivery into a compact device, but the newer model expands connectivity with additional ports, including SD and microSD slots. Both generations maintain active cooling as a central feature, designed to prevent thermal throttling during sustained transfers, a capability that has been demonstrated in testing of the original device where performance remained stable under load . Physically, both devices share a similar credit card-sized footprint and transparent design, but the Disk Pro 2 refines usability with a detachable lanyard cable and broader magnetic compatibility. Overall, the original model prioritizes simplicity and integration, while the newer version emphasizes flexibility, lower entry cost, and configurable storage.

 

Sharge Disk Pro 2 – Launch Date and Price?

The Sharge Disk Pro 2 is scheduled to launch via a crowdfunding campaign on Kickstarter, with the campaign planned to go live on June 9. As with previous releases from Sharge, this approach places the product in an early-access phase prior to wider retail availability, meaning final specifications and delivery timelines may still be subject to change. The device will be offered in two distinct variants, allowing users to choose between different feature sets and efficiency profiles at launch. The entry-level Lite version is priced at $49 and features HDMI 2.0 output with a higher reported power consumption of around 4W. The higher-tier Ultra version is priced at $69 and includes HDMI 2.1 support, alongside a lower power draw of approximately 1W. Both versions are expected to ship with a 24-pin pure copper lanyard-style cable that supports data transfer, charging, and DisplayPort signal transmission. This tiered pricing structure introduces a lower entry point compared to earlier expectations, while still separating features such as display capability and power efficiency between the two models.

 

SUBSCRIBE TO OUR NEWSLETTER
[contact-form-7]
Join Inner Circle

This description contains links to Amazon. These links will take you to some of the products mentioned in today's content. As an Amazon Associate, I earn from qualifying purchases. Visit the NASCompares Deal Finder to find the best place to buy this device in your region, based on Service, Support and Reputation - Just Search for your NAS Drive in the Box Below

Need Advice on Data Storage from an Expert?

Finally, for free advice about your setup, just leave a message in the comments below here at NASCompares.com and we will get back to you. Need Help? Where possible (and where appropriate) please provide as much information about your requirements, as then I can arrange the best answer and solution to your needs. Do not worry about your e-mail address being required, it will NOT be used in a mailing list and will NOT be used in any way other than to respond to your enquiry. [contact-form-7] TRY CHAT Terms and Conditions   Alternatively, why not ask me on the ASK NASCompares forum, by clicking the button below. This is a community hub that serves as a place that I can answer your question, chew the fat, share new release information and even get corrections posted. I will always get around to answering ALL queries, but as a one-man operation, I cannot promise speed! So by sharing your query in the ASK NASCompares section below, you can get a better range of solutions and suggestions, alongside my own.

WE LOVE COFFEE

 

Avec le ZimaCube 2, IceWhale ambitionne de bousculer le marché des NAS. Fini le simple boîtier que l’on installe dans un coin pour stocker ses données. Le fabricant veut transformer le NAS en un véritable serveur domestique polyvalent, capable de gérer le stockage, la domotique, mais aussi des usages plus avancés comme l’IA en local…

ZimaCube 2

Le ZimaCube 2 est un boîtier avec 6 baies pouvant recevoir des SSD et disques durs 3,5 pouces. À cela s’ajoutent 4 emplacements M.2 NVMe. Le système (voir ci-dessous) est préinstallé sur un SSD interne de 256 Go et sur un autre emplacement dédié. Ce nouveau NAS est construit autour d’un processeur Intel Core de 12e génération (Core i3 ou Core i5) épaulé par de la mémoire DDR5.

3 configurations = 3 usages

La gamme se décline en 3 modèles :

• ZimaCube 2 est un boitier gris avec un Core i3-1215U et 8 Go de DDR5
• ZimaCube 2 Pro est un boitier noir avec un Core i5-1235U et 16 Go DDR5
• Creator Pack identique au Pro avec 64 Go DDR5, 1 To de NVMe et une NVIDIA RTX PRO 2000

Pour rappel, l’Intel Core i3-1215U a obtenu un score Passmark de 10 196 points. De son côté, l’Intel Core i5-1235U a obtenu 12 595 points.

Connectique

Pour les interfaces de connexion, le ZimaCube 2 propose:

• À l’avant : 3 ports USB 3.0 (dont 1 Type-C), 1 sortie audio 3,5 mm ;
• À l’arrière : 2 ports USB 3.0, 2 ports Thunderbolt 4, 1 sortie HDMI 2.0, 1 sortie DP 1.4, 2 ports réseau 2,5 Gb/s.

Les modèles Pro et Creator Pack disposent également d’un port réseau 10 Gb/s.

Le NAS dispose également de 2 emplacements PCIe (PCIe 4.0 x16 et PCIe 3.0 x8), permettant d’installer une carte graphique, un accélérateur IA ou encore une carte réseau supplémentaire.

ZimaOS

Le système maison se nomme ZimaOS (intégration native de CasaOS). Il est basé sur un Linux Debian et Docker. Il propose une boutique d’applications (800 différentes) permettant de déployer en un clic Plex, Jellyfin, Immich, Nextcloud, Home Assistant… Le gestionnaire de fichiers regroupe NAS local, stockage cloud et périphériques USB dans une même interface.

Pour les utilisateurs qui le souhaitent, TrueNAS, Proxmox, ou Unraid sont compatibles sur ce matériel.

Prix et disponibilités

Les machines sont disponibles dès maintenant sur shop.zimaspace.com :

• ZimaCube 2 = 681,70€
• ZimaCube 2 Pro = 1 108,30€
• Creator Pack = 2 132,13€

Un positionnement tarifaire intéressant, au regard des solutions concurrentes… À voir si le public sera au rendez-vous !

source