Valve Steam Machine, SteamOS Desktop, Steam Frame VR and Steam Controller

Yeah, that's pretty awesome - about to head to work so only had a chance to watch the Valve hardware overview, but I'm liking what I saw.
 
  • Like
Reactions: DPI
like this
zen4 6C, rdna3 28CU/8GB of gddr6 (can expect a significant lower (15-20%) than RX 7600 ?, 12.5% less CU, 33% less power, maybe an 6600xt type depending on the title ?, hdmi 2.0, quite "reasonable aging specs", could have been cooking for a long time and aiming for an interesting price point (steam can assume a lot of post sales return unlike anyone else in that space), better be an interesting price point to be of any interest...
 
Last edited:
6C… I wonder why they didn’t go with 8C? I keep hearing people talk about how that’s the minimum for future gaming. Maybe a thermal limitation.
 
6C… I wonder why they didn’t go with 8C? I keep hearing people talk about how that’s the minimum for future gaming. Maybe a thermal limitation.
semi-custom, money saving and with that 30 watt TDP target as you say would have limited the capacity to scale to 8-10 cores.

I doubt there is such things as minimum core count for gaming (that maybe something that could exist in realtime-security protocol scenarios ?), past 4 threads available, I imagine it is all about minimum multithread performance (and on less cores the better), 6C zen4 will be always faster than say a 12 cores 3900 in games, specially in low watt situation
 
Couldn't care less about the VR headset or the controller, but it is cool that they are going for a desktop version of a SteamOS machine.

As expected its AMD based, as AMD is still the way to go for 3D performance under Linux, but this should go a long way to drive even more improvement than the Steam Deck already has when it comes to Linux gaming, and that isn't a bad thing (unless you are Microsoft)
 
The VR setup is pretty interesting and it seems low-profile and easy to use. Those have been my issues with other solutions. I don't really need an underpowered mini-PC or another controller, though.

BUT where that Half-Life 3 at, fam?
 
6C… I wonder why they didn’t go with 8C? I keep hearing people talk about how that’s the minimum for future gaming. Maybe a thermal limitation.
6C… I wonder why they didn’t go with 8C? I keep hearing people talk about how that’s the minimum for future gaming. Maybe a thermal limitation.

Yup. I mean, this is an entry level machine.

With a semi-custom RDNA 3 chip with 28 compute units and a sustained clock of a maximum of 2.45Ghz, we are talking considerably less GPU performance than a previous gen low end Radeon RX 7600.

The RX 7600 has 32 CU's and a max boost clock of 2695mhz. Reviewing the 3DMark results browser shows that the average user only gets a sustained 2650mhz out of it though.

So we are talking a GPU that is at best 20% slower than a previous gen low end RX 7600 (non XT) model.

And that assumes the GPU isn't gimped by using main system RAM in which case it is likely even slower.

So my take is, at this level of GPU performance, 6 cores on the CPU is unlikely to hold you back,
 
The hardware looks interesting, but pricing is everything... and we won't find that out for a little while.

And yes, it's notable that there are no new games to help showcase the hardware, at least at this stage. Whether or not you believe the "Half-Life 3 is close, really" rumors, it would've been nice to have marquee titles to spur sales.
 
the eye tracking compression will be interesting (like APple do I think), maybe that kind of tech could end up on very large monitor has well one day, not just for compression but how much rendering effort is put by region on the screen to start with.

The hardware looks interesting,
For the desktop box at least, seem all about the software if it will be interesting (the most boring hardware imaginable), I am sure the headset will be top of the line great too.
 
The VR setup is pretty interesting and it seems low-profile and easy to use. Those have been my issues with other solutions.

My biggest complaint with VR has been the titles they release on them.

It has overwhelmingly been dumb casual "Nintendo Wii" style games, and no one has really solved the FPS application difficulties when it comes to moving around.

That's what it is going to take for me to adopt VR. Universal compatibility with AAA FPS titles in a way that makes sense and feels natural for movement. Not some "room scale" bs. with teleportation every few feet.

Until this happens both VR and AR are completely and totally uninteresting to me.

People like to say "but VR is different, you can't expect to play games using it like you always have, it is a new technology, you have to use it for what it is intended".

And to that I say, I'm not interested in that style of Nintendo Wii-like game at all. And I don't have to use VR at all.

I only adopt new technologies when they do what I want them to. I'm not going to buy them just because.

To me, the only value of VR is badass simulation of first person titles. And for that, I need a model that uses a top end PC, not some low end face mounted ARM bullshit. And I need someone to invent good controls for AAA style titles. And I need universal or near universal AAA FPS title buy-in so the titles actually work.

Until that happens, I'm out.
 
I'm surprised and disappointed by a lot of the decisions that went into these devices, especially knowing that Valve Time means we're stuck with them for years. RNDA 3 on the Steam Machine means no proper FSR 4 support, which means not much of an upscaling "conveyor belt" of games continuing to look and run better on the same hardware. A low-end device like this really needed to be on the cutting edge in upscaling, and no one was forcing them to release something before it was fully ready, both as a device and as part of an ecosystem. For the Steam Frame, I'm even more baffled. Monochrome passthrough? LCD display panels with no MiniLED? A glorified Quest 3 SoC that relies on a x86->ARM translation layer and can't even run Half Life: Alyx right now? What were they thinking?

I also have to ask "why now?" It would be one thing if HL 3 were ready and they had to get the hardware out the door, but not a peep of that either. Maybe it will all turn into something special in the long run, but this is not at all what I was expecting. It feels like a big let-down.
 
The real interesting thing here is if Valve ever decides to release a more general purpose SteamOS image users can install on their own hardware, like that old discontinued effort from 2013.

It would be really cool if a more general purpose image became available. The Polish of SteamOS would put the other efforts (Bazzite, ChimeraOS, HoloISO, etc.)
 
My biggest complaint with VR has been the titles they release on them.

It has overwhelmingly been dumb casual "Nintendo Wii" style games, and no one has really solved the FPS application difficulties when it comes to moving around.

What problems? Most FPS VR games do have fully analog movement, teleport is an option for accessibility reasons that can be disabled.
If you are referring to nausea that some feel, there is nothing VR can do to really negate it as it is purely biological issue, a disconnect between what your eyes see but your inner ear isn't feeling any motion = motion sickness. It is something the person can get used to and your brain eventually accepts it as normal situation.
 
Couldn't care less about the VR headset or the controller, but it is cool that they are going for a desktop version of a SteamOS machine.

As expected its AMD based, as AMD is still the way to go for 3D performance under Linux, but this should go a long way to drive even more improvement than the Steam Deck already has when it comes to Linux gaming, and that isn't a bad thing (unless you are Microsoft)
I'm interested in the headset because I can watch movies, sports, play my Steam Deck, stream a Steam Machine, or whatever else I want on it as it is a monitor. Games should scale quite nicely on it to give a tremendous view. :) Of course I want to check out some VR titles also.


View: https://youtu.be/bWUxObt1efQ?t=1604

26:46 into the GN video.
 
It has overwhelmingly been dumb casual "Nintendo Wii" style games,
and sitted down sim, (car, plane, spaceship) which seem to me a much more obvious use case than an shooter, as your are in-game characther is staying sitted down just like you are, way more natural and has been working quite well for more than a decade now.
 
I'm surprised and disappointed by a lot of the decisions that went into these devices, especially knowing that Valve Time means we're stuck with them for years. RNDA 3 on the Steam Machine means no proper FSR 4 support, which means not much of an upscaling "conveyor belt" of games continuing to look and run better on the same hardware. A low-end device like this really needed to be on the cutting edge in upscaling, and no one was forcing them to release something before it was fully ready, both as a device and as part of an ecosystem. For the Steam Frame, I'm even more baffled. Monochrome passthrough? LCD display panels with no MiniLED? A glorified Quest 3 SoC that relies on a x86->ARM translation layer and can't even run Half Life: Alyx right now? What were they thinking?

I also have to ask "why now?" It would be one thing if HL 3 were ready and they had to get the hardware out the door, but not a peep of that either. Maybe it will all turn into something special in the long run, but this is not at all what I was expecting. It feels like a big let-down.

I think there are many reasons for that.

First off, AMD hasn't made a scaled down version of RDNA4 yet. The smallest is Navi 44. The Radeon 9060 (non Xt) does have only 28 CU's, but it is a scaled back version of Navi 44 on a costlier TSMC N4P process node.

Using a semi custom Navi 33 - however - on TSMC N6 is likely dirt cheap by comparison.

Remember, this machine is supposed to be budget friendly.

It also probably has something to do with how hackish FSR4 support is under Linux today. If mature FSR4 support for Linux were ready, they might have considered current generation, but still, that would have been expensive, and they would have been fighting for latest gen supply.
 
Every bit of it, pretty awesome! Valve just single handed dragged VR forward again as opposed to everything being predicated on Meta's proprietary platforms. The hardware looks good too and running SteamOS is even better. FOVeated rendering AND Streaming! Pancake lenses! Eye tracking built in! A standardized, open source/spec API for FOViated rendering much less streaming will be transformational and let many devs use a standard approach. Now we're getting somewhere. I am curious about the inside-out tracking to be honest; I am wondering if they will be able to increase fidelity or whole body tracking for those who have Lighthouse 2.0 from their Indexes, but if the new system works as well and can be expanded sufficiently then so much the better. New controllers with TMR will put an end to everyone worried about potentiometers. Also the controller design being useful for "standard" gamepad titles seems smart as well; one of the very few things Steam Input lacks is support for independent dual motion control such as used by JoyCons; hopefully this will fix that. A discreet 6ghz streaming dongle goes all the way back to the 802.11ad wireless prototype for the Vive and Index, so glad things have moved along well enough. I'm a little worried about the onboard mobile SoC, but considering that pretty much every other device that's stand alone as something similar, the actual "power" can come from connecting it directly to your PC rather than trying to run everything on an ARM chipset so I can't be too upset. After all the alternative was an ARM chipset controlled by an OS reporting everything you did to Meta, instead of a Linux FOSS SteamOS one with full user control.

Steam (Deck) Controller is pretty much everything we hoped. I admit a bit of trepidation with a symmetrical layout and the position of the trackpads, but it turns it they may be just fine depending on the structure of the controller/size itself. Lets hope the haptics are equal or better than a DualSense / Edge, and its other features continue to move things forward. I'd say I'd hope to see a premium / luxury variant similar to a Xbox Elite S2 / DualSense Edge, but they pretty much covered most of in the default version.

The Steam Machine looks like a winner overall, but I have a few elements I wonder about. Most primarily, I HOPE they do not stick exclusively with a RDNA3.x GPU. RDNA4 or beyond is the way to go for no other reason than native support for FSR4. Now maybe this is one of those custom bits that after Redstone AMD will backport FSR4 to working on this chip reasonably well, but it would be better if they didn't need to do that. During the release of the original Steam Deck the use of a RDNA2 GPU was a big deal compared to how they were still using crappy Vega in most APU-ish things of that type, so I'd expect they'd need to clean things up a bit here especially if this is not a PC that has any expansion potential because of its small form factor. It IS a PC and that's great, but if theyt would just change it to perhaps Zen5 and 8C instead of Zen4 6C, ensure the GPU is RDNA4 based and thus FSR4 compatible if you're depending on FSR for the noted 4K 60FPS, and perhaps some RAM upgrades. On a smaller scale, come on now with the Display Port 1.4 and HDMI 2.0 - maybe its the older GPU but those are not viable. Hell, even my old OLED LG C2 monitor requires HDMI 2.1 in order to get 4K 120hz HDR 4:4:4 chroma (something many Linux users have become frustrated regarding when HDMI consortium decided granting that bandwith to FOSS drivers wasn't accessible, leaning to varied workarounds. Fuck off, HDMI Consortium!) so Display Port 2.1a and HDMI 2.1b are better alternatives for a more modern capability at this point. Likewise, upgrade to some USB4 ports (at least one) - its not even new hardware. Of all of it, this is perhaps what is most disappointing in that some of the hardware feels a bit too old for the unique cutting edge stuff that Valve pushes. When the Deck launched, only the PS5 / XSX had a similar Zen2 + RDNA2 SoC - the Steam Machine needs to be forward thinking by far. Valve is a MUCH larger company compared to something like Framework (who make great modular laptops) but the latter had custom AI focused SFF boards made for their Framework Desktop, so I'd think AMD can really push the envelope to help Valve launch and showcase its latest and greatest.

Overall, I'm quite pleased. I'm definitely in for a Steam Frame (If somehow Gaben is reading, a special cyberpunk-themed Deckard edition would be a nice option) and Steam Controller, but we've got time before 2026. They should upgrade everything as much as possible in that interim - some devices have more room than others. Even the specs have an asterisk that says "May be subject to change" ; good! Change, improve things in the meantime, patch holes, make it better since the announcement and between now and release! THe hardware should be phenomenal as we already know they have the capability to push the OS and software forward. Still overall, a major leap for 2 out of the 3 with a few little tweaks and the last one can be a step forward if done correctly with a few improvements.
 
Last edited:
What problems? Most FPS VR games do have fully analog movement, teleport is an option for accessibility reasons that can be disabled.
If you are referring to nausea that some feel, there is nothing VR can do to really negate it as it is purely biological issue, a disconnect between what your eyes see but your inner ear isn't feeling any motion = motion sickness. It is something the person can get used to and your brain eventually accepts it as normal situation.

I never said there were problems per se, but there are challenges in getting the only types of titles I am interested to work well.

For instance, how do you handle a title where look around and hand movement are separate from eachother.

And if you are stuck in a room scale setup, when using an open world walk around title, how do you travel in game more than 10-15ft without needing to teleport to the next 10-15 x 10-15ft square on the map.

There needs to be some sort of next gen controller setup that can both handle player movement, and crosshair movement, and the software needs to be written to allow the camera to move freely with the VR headset inputs, while at the same time allowing the crosshairs (probably mouse) to handle direction.

and sitted down sim, (car, plane, spaceship) which seem to me a much more obvious use case than an shooter, as your are in-game characther is staying sitted down just like you are, way more natural and has been working quite well for more than a decade now.

I'll agree that these are pretty good applications for the tech. They just aren't very interesting to me.

I want to get as close to the holodeck experience as I can, while at the same time play in a seated position.
 
Using a semi custom Navi 33 - however - on TSMC N6 is likely dirt cheap by comparison.
this look a lot like those rdna3/zen4 Phoenix apu, they were on TSMC 4nm generally and the gpu part is probably much smaller than the old navi 33 (204mm) gpu on tsmc was on that node (navi 44 was a bit smaller than that).

A bit something like this could have been the base ?
https://www.techpowerup.com/cpu-specs/ryzen-5-7640u.c3205, with boosted CU count

they might have considered current generation
Depending on how semi-custom we are talking, you probably need to lock some core generation down on a system 12-18 month before launch (maybe they aimed for this Christmas) ? Things like FSR4 could have been up in the air until this spring and speculative at the time, RDNA 4 yet to launch, etc....
 
Personally, I don't want to play VR titles for long stretches of time. It's claustrophobic, isolating, clunky, and (thus far) pretty uncomfortable. It's a fun way to experience something new and unique for 15-30 minutes, and then I'm done. The titles that do that really well are what I want more of. The PSVR and PSVR2 libraries have gotten pretty robust, although you'd never know it based on how little publicity they get. If I had a better way to play those sorts of games, I'm all for it.
 
For instance, how do you handle a title where look around and hand movement are separate from eachother.

And if you are stuck in a room scale setup, when using an open world walk around title, how do you travel in game more than 20 ft without needing to teleport to the next 20x20ft square on the map.

There needs to be some sort of next gen controller setup that can both handle player movement, and crosshair movement, and the software needs to be written to allow the camera to move freely with the VR headset inputs, while at the same time allowing the crosshairs (probably mouse) to handle direction.

If you are referring to if the movement is headset relative or controller relative, then yes it is a minor issue. Controller relative is a mess if you have to do a lot of things with your hands, I never use it. Headset relative feels weird at first but you quickly learn to compensate with the analog stick when looking around. If I walk forward and look left I automatically tilt the stick to the right so that I keep on walking the same direction I was going to.

There is also a third option in some games, body relative which is entirely natural but that requires an extra tracker on your hip.
 
Personally, I don't want to play VR titles for long stretches of time. It's claustrophobic, isolating, clunky, and (thus far) pretty uncomfortable. It's a fun way to experience something new and unique for 15-30 minutes, and then I'm done. The titles that do that really well are what I want more of. The PSVR and PSVR2 libraries have gotten pretty robust, although you'd never know it based on how little publicity they get. If I had a better way to play those sorts of games, I'm all for it.
I suspect Valve is addressing part of this with the Steam Frame. Games on the PSVR2 are awesome; what's not awesome is wearing a bulky headset that's physically tethered to a console and requires extensive setup.

We don't need raw computing power now so much as we do a simpler, more elegant experience. Valve's headset is relatively light, and a wireless PC link is fairly seamless. SteamOS will hopefully take some of the software complexity away from this kind of VR. My worry is that Valve is still catering to enthusiasts more than the mainstream at the moment. And for all the gripes about the costs and weight of the Vision Pro and Galaxy XR, they are easier to use.

And the bigger concern is simply that headsets aren't really the future of XR — smart glasses are (assuming they catch on in a big way). They keep you grounded in the real world, they're more practical in daily life, and roughly as comfortable as wearing regular glasses. And while it might take a long time, I could see a day when they're as powerful as headsets without having to pair to another device. In that light, the Steam Frame is really just one step on a long road.
 
RDNA 3 in the steam machine is a let down tbh, everything should be pushing FSR 4 at this point. Headset basically seems like a Quest 3 with better streaming.

It's all going to depend on price, none of these products seem like they should be super expensive, but I'm guessing they'll be priced pretty high.
 
Last edited:
Mannn I was really excited for the steam machine, was really wanting a small device to put in my living room. 8GB of VRAM though is going to age like milk. It would have been an instant buy for me at 16GB VRAM or atleast 12GB VRAM, but 8GB??? What were they thinking?

I'm excited for the steam controller, looks awkward but I'll give it a shot. Also excited for the Frame, been wanting to get into VR since starting my job, and ideally I don't want to buy from Meta.
 
I think there are many reasons for that.

First off, AMD hasn't made a scaled down version of RDNA4 yet. The smallest is Navi 44. The Radeon 9060 (non Xt) does have only 28 CU's, but it is a scaled back version of Navi 44 on a costlier TSMC N4P process node.

Using a semi custom Navi 33 - however - on TSMC N6 is likely dirt cheap by comparison.

Remember, this machine is supposed to be budget friendly.

It also probably has something to do with how hackish FSR4 support is under Linux today. If mature FSR4 support for Linux were ready, they might have considered current generation, but still, that would have been expensive, and they would have been fighting for latest gen supply.
This really feels like something that they should be driving before the device's release, rather than release a device that's significantly outdated. Valve, perhaps more than anyone else in or debated outside the gaming sphere, has the ability to "make shit happen(tm)" on Linux. Taking Wine and DXVK and turning it into the polished Proton and Steam Deck Verified play for compatibility? SteamOS needing better driver support for GPUs, resume/suspend etc... they just plain made it happen in both the Linux kernel and the AMD FOSS driver etc. Even when we look at the time between the end of the year, the 2026 delivery of these items gives them some time though they should have been thinking about it much before.

Valve, working with AMD like their massively successful Steam Deck custom-ish SoC, should mean that both of them benefit when they get it right. As I said, back in the day a Zen2 + RDNA2 was unheard-of for 'normal' APU or custom PC boards; it was only something we saw on the PS5 era of consoles. Thus, the idea that there's no other easily available RDNA4 GPU shouild be a great opportunity for this to be the first; that's the point of a custom partnership rather than just off the shelf normalcy. RDNA4's usage of FSR4 is pretty much the first time there's a parity on image quality and features for generic upscaling performance (even frame gen, and with redstone multi-frame gen). They can't afford to launch something in users hands in 2026 that is already out of date, especially without the "its a portable thing, please excuse its not supposed to have full PC performance". Especially given how long Valve supports these devices - and well - its going to look antiquated if they aren't fcutting edge at the start.. This looks bad for both Valve and AMD and neither one need to let that happen - they've seen success when things are done right.

Use some Zen5 cores (i'd prefer 8C if possible), on the CPU side, put an RDNA4 (maybe a 9060XT equivalent which has both 8 and 16GB sides) on the GPU side, consider an option with 32GB RAM and 16GB VRAM on the high end and let the 16 + 8 be for the cheaper lower end. Upgrade some of the other components and standards as I mentioned (ie USB4, DP2.0 + HDMI2.1b etc) and this could be a winner. I think Valve saw that more users ended up buying the middle and high tier Steam Deck models rather than the absolute cheapest, so worrying too much about cost may well..cost them in sales if the hardware isn't up to snuff.

On the software side, it should be easier for Valve and AMD to make shit happen(tm) with improved Linux support for FSR4; in my experience it already isn't bad especially if you use it on the FOSS drivers and the game originally had at least FSR3.x support first, but they can improve things. Getting the software implementations more smoothly seems an element of focus on time and resources but with the new Steam Hardware its a great motivation to put some time into smoothing out the rough edges, which will have benefits both to Valve with Linux/SteamOS and AMD far beyond.
 
The Steam Machine seems unnecessary, as pretty much everyone who knows and uses Steam already has a machine. The controller might be nice, but I already have so many controllers, already. Maybe the Steam Frame VR will be worth it, regardless if it's good or not, competition is good.
 
Whats our predictions.. $499 for the 512gb option... $799 for the 2tb option ?
(on the Steam Machine that is ..)
I'm saying $399-$450, something puts it into direct contention with ALL the current consoles and not just the PS5/Series X.

As to the controller and frame headset pricing, no idea. I'm probably game to buy either no matter the cost though. I've really like my original Steam controller and Valve Index over the years, I'll gladly pay the ask for any follow up products. There's an opening in the market for mid to high -end VR that is empty right now. They can price the Frame however they want if it is good enough. Consider the Meta headsets are highly "enshittified" products requiring facebook accounts, 3rd party headsets are (mostly) glorified/higher-res Index designs, and the Apple Vision is sort of a joke in terms of price to features.
 
The Steam Machine seems unnecessary, as pretty much everyone who knows and uses Steam already has a machine. The controller might be nice, but I already have so many controllers, already. Maybe the Steam Frame VR will be worth it, regardless if it's good or not, competition is good.
You'd be amazed at the number of kids playing on trash computers or younger people who are apprehensive about buying a PC for gaming. I think our hobby has gotten a bit less accessible since some families don't have a desktop of any sort. I'm guessing the play is to make the Steam Machine an extremely competitive price - something obtainable to even a teen.

the Steam controller looks kinda chunky
I'd put that in the "pro" column. The old steam controller was chunky too, pleasantly reminded me of some chunky older controllers like the DC or OG Xbox.
 
Last edited:
You'd be amazed at the number of kids playing on trash computers or younger people who are apprehensive about buying a PC for gaming. I think our hobby has gotten a bit less accessible since some families don't have a desktop of any sort. I'm guessing the play is to make the Steam Machine an extremely competitive price - something obtainable to even a teen.
Maybe, it will really depend on the specs and the price, but new hardware is never going to be cheap, especially with RAM and SSD prices going through the roof because of AI, and graphics card prices having been ridiculous the last half decade.
 
Rembrant had zen3+ and RDNA 2 and were out january 2022, a bit before the steamdeck
https://www.techpowerup.com/cpu-specs/ryzen-5-6600h.c2525

Spring 2022, zen2-rdna2 was already quite mature, just less than zen4-RDNA 3 is now too.
I meant specifically the combination of the two. Remember that the Steam Deck and its specs were originally announced nearly half a year prior, around July of 2021 as I recall. The PS5/XSX launched in Nov 2020 if my timing is right as well. I know there were lots of projects back in the day after the Deck was announced (but in manufacture) where trying to get ahold of comparable mobile APUs meant you couldn't get RDNA2 IGP and instead had to settle for some Vega variant . I remember as I was building a cyberdeck project that was stuck to AMD 4x4 sized boards (or comparable embedded, NUC-style etc) using Zen2 SoC, but without anything as powerful as RDNA2 on the GPU side.
 
Last edited:
pricing seem hard to predict, but everything seem to be on the aggressive direction, 8gb of vram, 512GB (not 1TB) entry drive, 1gbps ethernet , maybe they do not know yet, but the plan was for agressive pricing when those decision were made.

7640u do fit under 180mm, you need to add 20 cu but maybe they could cut a little bit of space (2 cpu core, less io capability, npu, etc...), fit under 200mm ?
 
Sat through some more preview videos and I'm really liking it. Also, the early rumored leaks were pretty much spot-on too. Spec-wise, I'd for sure be happier if it rocked a 7600 with 12GB (or even better some custom RDNA4 part), but really, I just want to play the JRPGs and other console-type fare on a console-like box. I'm going to assume that this Steam Machine will create more powerful spec'd clones running SteamOS (or Bazzite) with similar functionality; IE: Minisforum, Beelink, etc.

I'm really liking the 'GabeCube' nickname I'm seeing being used for this thing. I'm hoping somebody finds a way to mod a handle on the back of this thing to really give it that GC feel.
 
Back
Top