Well well, perhaps we finally know what the end game is here for the Steam Play Proton compatibility layer and a lot of the other work Valve has been doing for Linux with a possible handheld Steam console. Take this with a heavy pinch of salt as we're into super speculation territory now. UPDATES WILL BE POSTED AT THE BOTTOM.
Pavel Djundik, creator of SteamDB who has often dug up interesting new strings showing up in Steam updates has a new Twitter thread up going over some new appearances that raise some eyebrows.
New appearances are seen mentioning a "Neptune" controller along with things like "GameList_View_NeptuneGames", "SteamPal Games" and more curious entries that mention things like quick access and a power menu - all of which point to something quite a bit more than just a new controller. Not only that, there's also earlier mentions of a "Callisto Developer Program" and "Device Optimized Games" going by what Djundik found.
Recently, Valve's Gabe Newell spoke at Sancta Maria College in Auckland, New Zealand that was highlighted in a since deleted Reddit posts (but you can find the video on the likes of YouTube) where Newell was asked about Steam on consoles to which Newell replied "You’ll get a better idea of that by the end of this year". Initially, we thought that might mean the likes of Half-Life: Alyx on PlayStation VR 2 but now we're not so sure going by this new set of leaks.
It would make sense for such a device to be powered by Linux, so Valve has no licensing fees to deal with and can heavily customize it to their needs. It could easily leverage all the work Valve has put into Linux graphics drivers, Steam Play, perhaps Gamescope and much more that Valve has done for Linux. Perhaps their work on sorting out "new ways for prospective users to get into Linux gaming" and the "live USB media" that we mentioned here were all efforts towards this in some way? Would be a given for it to use an AMD GPU of some kind, considering Valve's investments into the open source Mesa drivers too.
Imagine if "Device Optimized Games" were those specifically ported to Linux to work with this device, that would also work well across desktop Linux with the Steam Linux Runtime dealing with any possibly library incompatibilities. Oh the possibilities. Throw in the idea I recently brought up of a Steam Game Pass…quite exciting.
Then again, it could end up just being a Steam Controller 2 and these optimized games are just setup for it ready. I would be happy with that anyway, not quite as happy as a full Linux-powered handheld Steam console but I do love the Steam Controller. Possibly even something standalone for future VR kits and of course possibly nothing as some leaks turn out. However, with the hints mentioning an "AirplaneMode", that would only be useful for a full handheld.
Bundle a new Valve game with it like they did with the Valve Index and Half-Life: Alyx and you could get plenty of sales.
What are you thoughts on all this?
UPDATE: the website Ars Technica has reportedly spoken to "sources familiar with the matter" who have confirmed it's real and will be Linux-powered. It may even launch by the end of this year. We've reached out to Valve Press to see if they have anything to say about it for us.
As for the price of the device. They'd have to sell it for close to cost or at a loss to compete with the Switch.
Hell for a UMPC I would even take it at cost as I figure Valve has an OK-ish chance to under cut the GPD Win devices even selling at cost. While a niche market there is definitely a market and the lack of good options for Linux (Most the GPD devices have various issues with Linux when I looked into them) would probably make me pull the trigger on one finally. Plus in my case I could run a good chunk of my library on a portable device from being stuck on an iGPU system for several years so I have plenty of lighter games I really enjoy.
When using Steam Link, I personnally use my Steam Controller with a mobile phone clip for Xbox... It usually works, same form factor:
https://www.powera.com/moga/
This seems to offer a better solution than any dedicated Steam Handheld I can think of. Indeed, now that I think about it, I've got a few extra phone holders for the car that could pretty easily be modded to fit my Steam Controller ... hmmmm
The developer program mentioned might be some kind of incentive for developers to specifically optimize for/support this device. This would be generally good for us.
As for the price of the device. They'd have to sell it for close to cost or at a loss to compete with the Switch.
Yeah but no. The Switch is underpowered for VR. Because that was the discussion about, when Newell said "You'll see before the end of the year". Unless I misunderstood? And I doubt they would sell a device that can't run the Valve Index... Valve must be thinking about the Steam ecosystem as a whole, I guess.
I'd target Xbox Series X and PS5 if I were Valve (specs and prices).
Just brainstorming... Based on something similar to laptop hardware... Taking into account it's a fixed hardware configuration and adding to this scale economies... And maybe selling it at no profit or under cost... Maybe something can be done to lower the prices.
Edit: I'm working on a 17" laptop with i9-10980HK and Quadro RTX 3000 that weighs no more than 2.2kg (4.85lbs)... It's going to be smaller and and weigh less too. Convert that to gaming hardware... Ryzen 5800H? RX 6600m? Probably a couple of tweaks will be required to get the best of the battery. Just ideas... Or my personnal wishlist.

Edit2: Rumors pretends it's a Van Gogh APU (codename Aerith) 4c+SMT / 8CU rdna2... 399$ USD... Too bad for my wishlist, AAA gaming and probably VR, but better for price and battery life:
https://videocardz.com/newz/valve-steampal-handheld-gaming-console-rumored-to-feature-amd-apu
Jupiter? Neptune? What does that mean? Wrong planet or another device spec?
Last edited by Mohandevir on 26 May 2021 at 5:48 pm UTC
Well, it does seem to be a more promising idea than the Steam Machine.I would buy the shit out of this if it worked like the Switch.
And game support is certainly WAY better now via Proton than it used to be many years ago.
On the other hand, I don't really know if there is a market for such a handheld.
I wish them good luck, anyway.
I for one would be interested in such a thing if it was reasonably priced and had a nice battery life. On the other hand if it's 'play Cyberpunk 2077' on a handheld it's probably not for me, but i'd like to play Dead Cells and the like on the go.I mean you can kind of already play Dead Cells on the Switch...
It would be also cool if they kept it open like SteamOS (with a debian based system perhaps) so you could install Retroarch etc. on it or use it as a pdf reader.
Would a device with swappable controller bits work? Like a way to have the section where a D pad/stick would go instead be able to have it unclipped and fitted with a physical D pad or stick or the touchpad like the Steam controller or potentially even something else. Customise the controller setup to your preference.Like the Thrustmaster eSwap Pro?
https://www.youtube.com/watch?v=Vg9svT9VsQo
I love mine.
i dont think the average joe would give a second chance to valve, imagine knowing the disaster it was the last time and beting your money at it?The average joe does not remember what happened two years ago, much less as far back as the Steam Machine is now. Otherwise the state of politics would be very different.
The developer program mentioned might be some kind of incentive for developers to specifically optimize for/support this device. This would be generally good for us.That is true. I'm sure the Switch itself sells at a loss, after all, and the other consoles. Yet another reason it was always important for Steam to make the dashed things themselves: They can make plans based on the revenue from additional game sales making up for losses on selling the hardware, whereas companies in a third party ecosystem cannot. So a bunch of third parties were never going to be able to sell at a price point competitive with consoles who also have that sell at a loss, make it up on game sales model.
As for the price of the device. They'd have to sell it for close to cost or at a loss to compete with the Switch.
Would a device with swappable controller bits work? Like a way to have the section where a D pad/stick would go instead be able to have it unclipped and fitted with a physical D pad or stick or the touchpad like the Steam controller or potentially even something else. Customise the controller setup to your preference.Valve do hold a patent on doing exactly that.
Edit: found [the patent](https://patents.google.com/patent/US20110105231A1/en).
As well as having an ecosystem of widgets so that people can customise the controls how they want, that would also make replacing them easy if they got faulty or gunged up or whatever.
Last edited by CatKiller on 27 May 2021 at 1:00 am UTC
That is true. I'm sure the Switch itself sells at a loss, after all, and the other consoles.
That is most likely true at console launch, when it's of paramount importance to get the majority share of the market to reap the benefits of network effect. You only have 1 console launch. But after that it's not a given anymore. There was a lot of speculation back then at xbox 360 and ps3 time that m$ and Sony stopped selling at loss after 3-4 years and were in active territory since ages.
The last generations though suffered pc competition greatly, they had to reduce iteration to just two years so probably they never had the time to compensate.
Nintendo however is a different player. It doesn't use superior hardware as it's selling point. It's totally possible Nintendo sells at production cost at launch and then start to gain something already after a few months.
Funny thing. Everyone is 100% sure consoles are sold at loss, nobody challenge this accepted fact. Yet in EpicVSApple Epic couldn't provide any piece of evidence or even some clue that this is a fact, despite this being a fundamental piece of its theory and despite having both Sony and M$ support. This at least should suggest something on consoles real manufacturing costs.
he may not remember, but the gaming journalist will remember themi dont think the average joe would give a second chance to valve, imagine knowing the disaster it was the last time and beting your money at it?The average joe does not remember what happened two years ago, much less as far back as the Steam Machine is now. Otherwise the state of politics would be very different.
Certainly one of the things that Valve will absolutely have to push much harder on and get right this time ishe may not remember, but the gaming journalist will remember themi dont think the average joe would give a second chance to valve, imagine knowing the disaster it was the last time and beting your money at it?The average joe does not remember what happened two years ago, much less as far back as the Steam Machine is now. Otherwise the state of politics would be very different.
With how much advertising and availability for the Steam Machines, I don't think the average joe even knew about it.i dont think the average joe would give a second chance to valve, imagine knowing the disaster it was the last time and beting your money at it?The average joe does not remember what happened two years ago, much less as far back as the Steam Machine is now. Otherwise the state of politics would be very different.
I bet more average people have heard of the Atari VCS!
Funny thing. Everyone is 100% sure consoles are sold at loss, nobody challenge this accepted fact. Yet in EpicVSApple Epic couldn't provide any piece of evidence or even some clue that this is a fact, despite this being a fundamental piece of its theory and despite having both Sony and M$ support. This at least should suggest something on consoles real manufacturing costs.I've always questioned this. Most people don't realize the cost of materials significantly drops when millions are ordered. With just about every console maker just making deals with AMD, it's hard not to think that they're making some serious amounts of bank, and that there's a good mark-up on things.
Funny thing. Everyone is 100% sure consoles are sold at loss, nobody challenge this accepted fact. Yet in EpicVSApple Epic couldn't provide any piece of evidence or even some clue that this is a fact, despite this being a fundamental piece of its theory and despite having both Sony and M$ support. This at least should suggest something on consoles real manufacturing costs.
Well we know the Xbox Series X (IIRC) at the very least is being sold at costs from Xbox's divisions VP saying as much in testimony and I wouldn't be surprised if at cost has been the case for the other two console manufacturers most the time. The only clear case of taking a loss I can really recall is the original 3DS model post price cut as it was definitely taking a loss just from some basic BOM costs even accounting for scale as a few people with electrical engineering experience figured it up on a few of the console centric gaming forums. But at costs makes sense as selling at a lose even when you factor in games sales couldn't be that large of a loss, as from many outside data firms many sold consoles only see about 10 game sales each on average with the people that buy a lot of new games helping balancing out the people that generally buy used or simply don't end up buying many games.
Processing Vulkan Shaders
zzz
Can easily be done on Valves servers. The target is exactly known, so nothing has to be done on the device itself.
See more from me