With the growing ability of Proton and the growing popularity of PC handhelds and Steam Deck, should Steam try a console again?
- Thread starter Tony72495
- Start date
Nah. I'd rather the Deck's USP (being an affordable PC handheld on an alternative platform) be used to push what should be acceptable in the PC gaming community - minimum performance standards and options (including resolution and screen ratio), input/UI standards that are as user friendly from the moment you hit play, and a level of support for DRM/Anticheat across all platforms.
If I wanted something that played everything with all the performance, I'd use a PC. If I wanted something cheaper and away from Windows, I'd buy used/refurb and slap Chimera/Bazzite on it - because the work done on SteamOS benefits all Linux users.
If I wanted something that played everything with all the performance, I'd use a PC. If I wanted something cheaper and away from Windows, I'd buy used/refurb and slap Chimera/Bazzite on it - because the work done on SteamOS benefits all Linux users.
I have an Alienware R2 which is one of the last 'Steam Machines' and it is fantastic. I have it booting directly into big picture mode and it's a great experience. It even has 'pass through' for another video source.
Right now I use my Steam Deck in docked form 90% of the time and I would jump on a Steam Console, regardless of who makes it.
Right now I use my Steam Deck in docked form 90% of the time and I would jump on a Steam Console, regardless of who makes it.
I think a Steam Machine is the way to push SteamOS further along. 6 years of Proton and no signs of getting certain games like Fortnite or quite a few other massively popular games natively playable on SteamOS. A full console release would be a way to finally push for certain games natively on Steam(OS) as a full platform, where those big games would still end up getting a bigger cut of revenue than on the other consoles.
I think it's something like 60%+ Steam users with performance less than or equal to a PS5, so even a Steam console around or slightly stronger than a PS5 would be an upgrade for most people and give a solid baseline performance for years. I could totally see them making a custom APU with more (Zen 4) CPU power than a PS5 pro, but less GPU power. I don't know if they could get away with using GDDR6 system memory like consoles though, rather than the typical DDR5 system memory that the Deck currently uses.
I also think it's worth mentioning that a Steam console would be a perfect companion device for a Steam Deck or a Deckard VR headset because you could wirelessly stream games from it to either device.
I think it's something like 60%+ Steam users with performance less than or equal to a PS5, so even a Steam console around or slightly stronger than a PS5 would be an upgrade for most people and give a solid baseline performance for years. I could totally see them making a custom APU with more (Zen 4) CPU power than a PS5 pro, but less GPU power. I don't know if they could get away with using GDDR6 system memory like consoles though, rather than the typical DDR5 system memory that the Deck currently uses.
I also think it's worth mentioning that a Steam console would be a perfect companion device for a Steam Deck or a Deckard VR headset because you could wirelessly stream games from it to either device.
Steam Deck is cool and all but I don't really want to play at 720p with FSR Performance mode and lowest settings on my 65" TV.
Could Valve create a console using the same architecture and process as the Steam Deck. Absolutely.
From a Valve strategy/business standpoint the beauty of the Steam Deck is that it's a console that basically adapts (via Proton) to play well in the space in which Valve already has a monopoly. Building a home console version of the Steam Deck won't be able to match Xbox, PlayStation or Switch in terms of install base, developer buy-in or performance (per dollar spent by the consumer).
So should Valve do it? Nah. Better (from a business standpoint) for them to stick to improving Proton for PC where they're unchallenged rather than directly wade into competition with Sony, Nintendo and to a lesser extent MS where they'll undoubtedly lose against the barriers to entry that include the layers of in-built consumer, developer, creation tooling and publisher support for the incumbent console platforms. The barriers to entry are very high here even if Valve has a lot of knowledge about the industry.
The only version of this that I think is somewhat viable is basically a more powerful Steam Deck without the screen that sits under your TV. But I think there's a large amount of Valves audience that would either know how to build something better to serve that purpose or already own a home console that serves that purpose. It doesn't seem like there's an obvious market opportunity.
That said if Valve wanted to go 'all-in' on building the fourth console platform I wouldn't be against more competition in that space but I don't really think it's viable.
From a Valve strategy/business standpoint the beauty of the Steam Deck is that it's a console that basically adapts (via Proton) to play well in the space in which Valve already has a monopoly. Building a home console version of the Steam Deck won't be able to match Xbox, PlayStation or Switch in terms of install base, developer buy-in or performance (per dollar spent by the consumer).
So should Valve do it? Nah. Better (from a business standpoint) for them to stick to improving Proton for PC where they're unchallenged rather than directly wade into competition with Sony, Nintendo and to a lesser extent MS where they'll undoubtedly lose against the barriers to entry that include the layers of in-built consumer, developer, creation tooling and publisher support for the incumbent console platforms. The barriers to entry are very high here even if Valve has a lot of knowledge about the industry.
The only version of this that I think is somewhat viable is basically a more powerful Steam Deck without the screen that sits under your TV. But I think there's a large amount of Valves audience that would either know how to build something better to serve that purpose or already own a home console that serves that purpose. It doesn't seem like there's an obvious market opportunity.
That said if Valve wanted to go 'all-in' on building the fourth console platform I wouldn't be against more competition in that space but I don't really think it's viable.
This is where I'm at, just keep iterating on the Dock.
Desktop just generally works and I think it's helpful to have open source and a GitHub presence to allow hacking that can creatively expand the platform in unexpected ways, but the turnkey device under your TV really is better suited by a docked Steam Deck vs. an entirely new product.
I sort of see the "steam box" like the hackintosh movement, something that the community can self-document and if it gets big enough then react to it with products.
Contributors to FEX-Emu/FEX
A fast usermode x86 and x86-64 emulator for Arm64 Linux - Contributors to FEX-Emu/FEX
github.com
alyssarosenzweig is a contractor for Valve and FEX being an x86 to ARM translation tool for Linux. This person is well known for their work on Linux on Apple ARM hardware
FEX-Emu - YouTube
I play games on it listed as not working all the time. Do they not update what the deck can play often enough? It plays a lot more than people think.
I feel like most of Steam Deck's success comes from being the first real PC gaming handheld with mass market presence. You couldnt really "build" your own gaming handheld.
The desktop market is already oversaturated and there is nothing new they can bring to it. Nothing has changed since they last tried their hands at it. Why would this be different?
The desktop market is already oversaturated and there is nothing new they can bring to it. Nothing has changed since they last tried their hands at it. Why would this be different?
Nah. Building an APU-based mini PC is easy and relatively inexpensive (compared to "full-sized" PCs) these days.
View: https://youtu.be/uzQcyewsEcA?si=fyFvF_wSYe1xaS-r
Valve does need to make SteamOS available for download though.
View: https://youtu.be/uzQcyewsEcA?si=fyFvF_wSYe1xaS-r
Valve does need to make SteamOS available for download though.
The sales proposition of a new Steam Machine would be less clear than the Steam Deck's obvious advantage of being the first popular handheld gaming PC, but a "Steam Console" with Proton's recent improvements could still have some benefit.
This device would pretty much be a SFF pre-built, and people in this thread are saying that Valve's existing audience would probably rather just make their own SFF and maybe download a licensed-out edition of SteamOS (which I imagine Valve will eventually offer someday). The first attempt at Steam Machines was basically a bunch of SFF pre-builts that shipped with early SteamOS, but one of the main criticisms of the project was that Valve didn't provide a reference model -- an example for the rest of the Steam Machines to follow.
While the Steam Deck's portability is its most obvious benefit, its status as a measuring stick for portable/potato PC gaming is possibly getting overlooked here. Steam Deck verification has forced some developers to optimize their PC games which has improved performance on similarly low-spec systems. If Valve actually does license out SteamOS, then it will have essentially repeated the entire Steam Machine initiative but with handheld PCs this time.
This is what a SteamOS console could be fore pre-builts or even desktops in general. Even if sales get nowhere near consoles, if there ends up being one specific desktop/SFF PC specification that outsells the others, It turns into an agreed-upon lowest-common-denominator spec for PC gaming in general which gives developers, players, and even builders a clear measuring stick to target. BTW, Microsoft could do this too if it came up with an alternate Windows interface for handhelds/controllers.
This device would pretty much be a SFF pre-built, and people in this thread are saying that Valve's existing audience would probably rather just make their own SFF and maybe download a licensed-out edition of SteamOS (which I imagine Valve will eventually offer someday). The first attempt at Steam Machines was basically a bunch of SFF pre-builts that shipped with early SteamOS, but one of the main criticisms of the project was that Valve didn't provide a reference model -- an example for the rest of the Steam Machines to follow.
While the Steam Deck's portability is its most obvious benefit, its status as a measuring stick for portable/potato PC gaming is possibly getting overlooked here. Steam Deck verification has forced some developers to optimize their PC games which has improved performance on similarly low-spec systems. If Valve actually does license out SteamOS, then it will have essentially repeated the entire Steam Machine initiative but with handheld PCs this time.
This is what a SteamOS console could be fore pre-builts or even desktops in general. Even if sales get nowhere near consoles, if there ends up being one specific desktop/SFF PC specification that outsells the others, It turns into an agreed-upon lowest-common-denominator spec for PC gaming in general which gives developers, players, and even builders a clear measuring stick to target. BTW, Microsoft could do this too if it came up with an alternate Windows interface for handhelds/controllers.
I mean... I guess developers could do something like making their games automatically boot in the "most optimized" graphical settings, like a "Deck console" mode if it detects the hardware.
It's just that I think console gamers want the simplicity of console games. I guess on consoles now there are actually graphical modes, but that's nothing compared to a good settings page in a PC game.
Even if the hardware is "consolized," they'd still be running PC games, which would need user input for setup in each game and twiddling to the user's preferences to make the games most enjoyable.
I just... I don't think console gamers would take a (compromised) consolized PC when the simplicity of actual consoles exist, and I don't think PC gamers would take the same over an actual gaming PC.
It's just that I think console gamers want the simplicity of console games. I guess on consoles now there are actually graphical modes, but that's nothing compared to a good settings page in a PC game.
Even if the hardware is "consolized," they'd still be running PC games, which would need user input for setup in each game and twiddling to the user's preferences to make the games most enjoyable.
I just... I don't think console gamers would take a (compromised) consolized PC when the simplicity of actual consoles exist, and I don't think PC gamers would take the same over an actual gaming PC.
The thing about the Deck is it was in a niche that didn't really exist before it came along. It birthed a whole new product category, and because of that they were able to find a lot of success.
Whereas if I wanted to play my Steam PC games on the TV, all I have to do is connect the HDMI cable to it instead of my monitor and turn on big picture mode. It's a solved problem already, and doesn't need new hardware or software solutions. And that's before you consider that consoles exist, providing a really similar but more refined experience compared to this hypothetical Steam machine.
The only reason I see Valve bringing back Steam consoles is if Microsoft starts looking like they might lock things down again like they did in the Win 8 days. That's what kicked off the whole Steam OS thing to begin with. But beyond I just don't think there would be a ton of interest.
Whereas if I wanted to play my Steam PC games on the TV, all I have to do is connect the HDMI cable to it instead of my monitor and turn on big picture mode. It's a solved problem already, and doesn't need new hardware or software solutions. And that's before you consider that consoles exist, providing a really similar but more refined experience compared to this hypothetical Steam machine.
The only reason I see Valve bringing back Steam consoles is if Microsoft starts looking like they might lock things down again like they did in the Win 8 days. That's what kicked off the whole Steam OS thing to begin with. But beyond I just don't think there would be a ton of interest.
They'd only do it if it served a meaningful purpose.
The point of the original machines was to scale up steam and protect themselves from being stuck on Windows.
Tons of issues with the approach, but they went on to tackle and evolve each facet (funding and pushing VK, Linux and SteamOS improvement, and investment in Wine>proton, Steam Controller/Input, overall of clients etc).
Steam Deck serves the "console" aspect already; point being to extend the whole steam library in a mobile form and establish a new form factor of handheld PCs.
Having a "console" on top of that, is pretty pointless on its own. It doesn't extend anything that Steam already does, and people already build for themselves.
SFF PCs are already getting attention from nvidia and MS, so I would expect them to push on that direction rather than Valve.
IMO people are overly obsessed with "consoles" since that's what they know. It's not necessarily the needed direction or form factor.
Next stop for Valve seems to be their own standalone VR/AR hardware. Maybe they have an external hardware solution for that, but I think it is more likely not the case. Having a box at home as a compute source makes sense, but PCs do that already, Deck can and eventually will do that better and better over time. Other vendors are already pushing into a more SFF "console" direction. Who knows. I'm leaning against that being a thing near term.
The point of the original machines was to scale up steam and protect themselves from being stuck on Windows.
Tons of issues with the approach, but they went on to tackle and evolve each facet (funding and pushing VK, Linux and SteamOS improvement, and investment in Wine>proton, Steam Controller/Input, overall of clients etc).
Steam Deck serves the "console" aspect already; point being to extend the whole steam library in a mobile form and establish a new form factor of handheld PCs.
Having a "console" on top of that, is pretty pointless on its own. It doesn't extend anything that Steam already does, and people already build for themselves.
SFF PCs are already getting attention from nvidia and MS, so I would expect them to push on that direction rather than Valve.
IMO people are overly obsessed with "consoles" since that's what they know. It's not necessarily the needed direction or form factor.
Next stop for Valve seems to be their own standalone VR/AR hardware. Maybe they have an external hardware solution for that, but I think it is more likely not the case. Having a box at home as a compute source makes sense, but PCs do that already, Deck can and eventually will do that better and better over time. Other vendors are already pushing into a more SFF "console" direction. Who knows. I'm leaning against that being a thing near term.
Last edited:
what's the failure rate on decks? even considering a worst case scenario nearing a 3% failure rate we're looking at less than 100k units.
we're talking about hardware that's sold ~3m total
Having been through this with both the Playstation TV and the Switch, I don't think it works well. The difference in resolution between the handheld and TV is too big.
It would be interesting to see, especially since it would mean a fairly cheap but capable preibuilt PC with a smaller form factor that isn't a NUC variety thing or put out by one of the usual PC OEMs you see at BB or Wal-Mart. I wouldn't want it to cannibalize the Deck, but the deck has a particular niche it fits into with what it is (portable). It would probably still be all AMD stuff, which means it'd be going along the same range as the PS5 and XB<whatever>.
Even right now it's fine if you don't use Wayland, works okay for some people who use Wayland and is still the preferred choice if you want to use DLSS/RT/Etc.
Nvidia isn't a no-go on Linux. It certainly won't be nearly as much of a pain for some users some time this month once Explicit Sync is available in the driver. NVK is gaining a lot of dev work and, well, steam with the updates it's had and still be worked on.Making a custom APU isn't cheap.
AMD is making bigger laptop SoCs in the second half of the year.
Memory bandwidth perhaps is the thing lacking compared to consoles.
There's a growing ecosystem of companies making mini pcs based on AMD chips already.
Nvidia is essentially a no-go zone for desktop / gaming on Linux.
Things are broken in general.
Whilst AMD's Linux drivers are improving quite a bit.
It's things like HDMI forum blocking open source 2.1 driver support that are the annoying things:
In the interview with Tested on youtube with Valve announcing the OLED launch, their engineer mentioned still working on opening up the OS to third party vendors.
So why not have gaming PCs, that boot straight into SteamOS in the future.
Drivers are all packaged up with the OS etc
I was thinking about how AMD has announced that they're making their AGESA BIOS firmware opensource by 2026:
AMD Open-Source Firmware is Coming, openSIL will Replace AGESA by 2026
During the OCP Regional Summit, AMD has shared plans to replace AGESA with openSIL. The change will not come soon, and according to details, it will be a slow process starting in 2026. AGESA firmware updates are quite important but also vulnerable to cyber-attacks, which is one of the key points...www.techpowerup.com
So maybe that'll inspire some innovation in the PC space.
Like perhaps if you press the PS button on a paired controller, it'll boot directly into steam OS.
Whilst normally booting in to windows etc.
Or how the Xbox has a hypervisor and virtual machines for games and the os.
You could make steam OS a guest that you could easily switch to, suspending your windows/linux desktop.
So being able to switch between a system optimised for power consumption and another for high performance.
Even right now it's fine if you don't use Wayland, works okay for some people who use Wayland and is still the preferred choice if you want to use DLSS/RT/Etc.
I could see this happening in the next refresh, along with a new steam controller. I think it will also depend on how good proton gets and if developers continue to focus on getting their games Deck certified. Both Proton and the deck certification are great now, but I think nearly every new game in Steam is going to need be close to playable before this makes sense.
Console gamers expect for things to just work. I don't think normal people will be happy to ever look at their settings (as much as this pains me to admit).
Console gamers expect for things to just work. I don't think normal people will be happy to ever look at their settings (as much as this pains me to admit).
What even comprises Deck Certification? Last DF ep they talk about how Fallout 4 is now "Deck Certified" and the only thing that changed is they removed the launcher popup that allows for game configuration.
Would it fracture the Steam Deck initiative to pretty much copy the Steam Deck, but make it not battery powered, with the latest APU possible, no screen, but with a Steam Deck controller?
I guess the official name is Steam Deck Verified. Here is a link to their compatability requirements: https://partner.steamgames.com/doc/steamdeck/compat#DeckCompatibilityChecklist
To summarize, Valve requires games to have:
- Controller support
- Controller glyphs (Xbox = A, PlayStation = X)
- Easy text input
- 1280x720 or 1280x800 resolution
- Legible text
- No compatability warnings
- No launcher or controller compatible launchers
- Settings that offer a "playable framerate"
EDIT: I'm also wondering if Valve actually tests these games with a real person. They may have a script or program that can quickly test these features or at least try to weed through most games.
Keep it simple stupid. Valve is a very rich but not massive company. To me, the strength of the Steam Deck stems from their ability to focus on just one or one-and-a-half models (OLED). It's why I think it's best to just improve the software until hardware gens improved so much, that a Steam Deck 2 would have drastically better performance. I'd rather they concentrate all their effort on a Steam Deck 2 rather than some console in addition. I imagine they try to develop a lot of things hardware-agnostic over time, but I bet all the frequent and great upgrades the Steam Deck received so far weren't possible if they had to support numerous models already.
Let's also keep in mind that Steam Deck sales are minuscule compared to PS5 and Switch. Branching out more before the Steam Deck (2) turns into a tens of millions seller probably doesn't even do much financially.
I'd also want them to stay far away from the original Steam Machine idea. Third-party manufacturers à la Asus and MSI just won't deliver the same level of quality at the same cost-efficiency. Nor software support. The Steam Deck stands far above the other handheld PCs and with those not benefitting from Steam sales anyway, what exactly is the incentive to put in as much production effort? They realistically just can't price-match Valve, business connections, and certain experiences aside. It's great that SteamOS is going to be available openly eventually, but I wouldn't dilute the Steam Deck name or anything similar by allowing Alienware to slap a label on a shitty, expensive PC handheld.
Medium- and long-term I'm all for Valve attempting a console and so forth. But rushing it now doesn't seem wise.
Let's also keep in mind that Steam Deck sales are minuscule compared to PS5 and Switch. Branching out more before the Steam Deck (2) turns into a tens of millions seller probably doesn't even do much financially.
I'd also want them to stay far away from the original Steam Machine idea. Third-party manufacturers à la Asus and MSI just won't deliver the same level of quality at the same cost-efficiency. Nor software support. The Steam Deck stands far above the other handheld PCs and with those not benefitting from Steam sales anyway, what exactly is the incentive to put in as much production effort? They realistically just can't price-match Valve, business connections, and certain experiences aside. It's great that SteamOS is going to be available openly eventually, but I wouldn't dilute the Steam Deck name or anything similar by allowing Alienware to slap a label on a shitty, expensive PC handheld.
Medium- and long-term I'm all for Valve attempting a console and so forth. But rushing it now doesn't seem wise.
The problem there is that, PC building knowledge aside, a mid-range GPU by itself can be $300+ by itself. I think somewhere there is a price / power point where expecting PC players to just DIY everything could be replaced with an alternative solid Steam box that isn't just the Deck's underpowered hardware without the screen.
They could make custom APUs at scale that outperform what you would get in a comparable PC for the price, and perhaps at the same time attract more big games to steam for more software sales.
Designing and manufacturing an APU that can outperform alternative pc hardware at the same price point is a lot easier said than done; unless they adopt a console model where making a loss on hardware in exchange for being locked in to a software platform is feasible...
there are things like the ROG NUC coming out but they're not exactly cheap, nor do they compare favourably to PC's at the same price point. the most favourable comparison is performance per watt, but most people aren't particular concerned about that in the PC space
the Steam Deck filled a perceived gap in the market. There is no such gap for a console like device because that is called a PC.
I'd be willing to bet that after some specific Nvidia updates we're getting on the Linux side this month happen and more things iron out, possibly moreso with the newer open-source nvidia driver eventually getting more than just "workable" for some games, we may, finally, see SteamOS 3.0 or whatever they end up renaming it release as a thing you can just install.
Do that with some cheap ITX hardware, especially with nvidia getting on board with having ITX configurations for some of their cards and we may have a better path to not having the "PC bullshit" you mention.
Yeah, absolutely. At the end of the day if Valve delivers the OS that's all that's really needed.
I'd be surprised if someone, anyone, hasn't kiiind of tried to already do it, actually. It wouldn't be at all difficult (and I mean this, from experience) to make a custom Arch installation that has the same packages Valve uses, the same versions, etc, and then also include easy ways to install extra pieces someone may need like the closed-source nvidia driver (or............ the nvidia-all repackaging from TKG, which everyone should be using >_>) or easier pointing to things in Discover to grab various things like Slack, Teams, Zoom, etc.
100% this.
I'm playing docked like 90% of the time.
An actual machine with my STEAM games would be bliss.
Wouldn't be a bad idea. I wouldn't use it, but I know a couple of people who would. Steam OS is pretty good now and handles VRR/HDR and all that too.
Personally, I just want a Steam Controller 2 that is setup exactly the same as the Deck to use with my current PC.
Personally, I just want a Steam Controller 2 that is setup exactly the same as the Deck to use with my current PC.
If they could provide it at the value the steam deck offers I would buy one. Would be great, but I don't think Value is setup to support two major platforms, or really have any incentive to reach a console price for pc hardware.
Improving and releasing Steam OS is probably their best course of action.
Improving and releasing Steam OS is probably their best course of action.
I think theres a fantastic opportunity for them to make a entry level device to PC gaming. A TV ready console, with the Steam Deck innards, running at double the TDP , running a vast majority of the library fine at 1080p for 249$ could be frankly amazing
They just need to release Steam OS and get all that Steam money on more devices.
No need for hardware
No need for hardware