Valve is making a Switch-like portable gaming PC | Hacker News Reader

Valve is making a Switch-like portable gaming PC

arstechnica.com

190 pointssigna11143 comments

Valve's track record with non-VR hardware is, to put it politely, not good.

Seven years after launching the Steam machine (2014) initiative, there are no Steam machines available for sale. [1]

Both the Steam Controller (2015-2019) and Steam Link (2015-2018) are discontinued and cannot be purchased, despite consistently positive reviews from users. [2]

As other commenters have noted there are already numerous "Switch-like, but x86" products that exist, and it's unclear to me what innovation Valve can bring to the space.

IMHO if Valve wants to be taken seriously in the hardware market, they need to stop pulling a Google and actually support/iterate/not discontinue their existing hardware products (before someone adds that they did support these through firmware updates, I mean updating the hardware to keep up with the times, Steam Link with apps or 4K, etc).

[1] https://store.steampowered.com/sale/steam_machines

[2] https://store.steampowered.com/search/?category1=993

That being said, they did support both the Steam Controller and Link for a very long time, and the Steam Controller continues to be used by many (The configurator and services for sharing the configurations work all the same today). Steam Link has mostly been replaced by software solutions like an Android App, or it being straight up built into some TVs.

I've got a hardware Link - it's still being supported now. It had an update a couple of weeks ago.

Yes, but I've had 4K monitors and 4K TV for almost as long as the Link has existed. It very quickly became useless to me the moment I upgraded to a 4K TV. The Steam controller is also an utterly horrible UX so the pair were just a waste of money that sat in a drawer until I stuck them on eBay.

A 4K Steam Link should have been an obvious iteration, and a controller with actual buttons and sticks (I half-joke about the controller, I know some love its touch nature, I just hated it, could never keep my thumbs in the right spot).

Then don't buy a steam controller? Why release something you can buy. And you can choose the layout (analog stick) or size xbox/ps4,ps5

The controller is amazing and enables non controller games to be played with one. This doesn't stop me from using another controller for emulators.

To be clear, it's not like there were never steam machines for sale, but manufacturers stopped making them.

Here, you can see five for sale in 2016. https://web.archive.org/web/20160225105722/https://store.ste...

I don't know that Valve were ever going to make physical devices themselves (other than the 100 they made for beta testers.) From pretty early on, and definitely by Steam Dev Days 2014, it seemed to me like they were trying to make a platform for hardware manufacturers to build linux computers made for gaming, rather than manufacturing their own computers.

As for the Steam Controller, and Steam link, I completely agree. Steam Link was a poor showing, and the Steam Controller, while neat, really could have used a V2 upgrade.

More generally, I don't know that Valve has made a truly successful product or game other than maybe Halflife Alex in well over a decade. And even with Alex, the player base is so limited due to being a VR exclusive.

It's pretty sad.

I was going to say Portal 2 but realized that it is from 2011 ...

Dota 2 is atleast barely in a decade from 2013. But ye, it is underwelming.

> Halflife Alex

Alyx. Sounds the same, but this spelling has the right gender.

> And even with [Alyx], the player base is so limited due to being a VR exclusive.

I have played it, and am currently replaying it with developer comments on. There's a reason it's a VR exclusive. There is no way this could transfer over a flat screen.

VR interactions are fundamentally different. Motion sickness severely constrain how many players may move about (it gets better with practice, but Alyx is aimed at beginners). The ability to pick up objects and throw them about is crucial to immersion. The kind of weapons that work in VR are different (1 handed mostly). Pacing is different we move slower in VR than we do in a flat screen FPS, often by a factor of 5. This modify level design and how to manage tension. Even combat must be paced slower, especially at the beginning where players have yet to reload their guns seamlessly (requiring gestures is more immersive, but there's a learning curve, and doing it under duress requires serious training).

A flat screen port of Half Life: Alyx would not be a good game. Most rooms would feel cramped. Some parts of the game would be cumbersome. Others like the Jeff chapter (Jeff is an unkillable monster with very good hearing) would swing between being overly punishing or way too easy. Many puzzles would have to be cut or replaced by something else. Some weapon features like an aiming laser would not even make sense (unless you remove the aiming dot at the centre of the screen, making aiming way too punishing). An inventory of 2 items (plus ammo) would feel limited for no good reason the player can see (one item per wrist is an obvious good reason in VR).

Some games, like vehicle simulators (car, plane, spaceship…) can indeed have good flat screen and VR version. Standing first person games however cannot. The difference between a keyboard/mouse/gamepad and a pair of knuckles/Vive/touch is too great to compensate for, in either direction.

> Alyx. Sounds the same, but this spelling has the right gender.

Our female friend Alexandra introduces herself as Alex.

Oops, now I remember that one of my classmates back in middle school had the exact same name.

I missed an opportunity to shut up, there.

I agree, Valve needs to better support hardware to build trust. But I think some perspective is worth mentioning.

Every Valve hardware device has had a longer maintenance and support lifecycle than any Google Nexus/Pixel device ever sold. Valve has consistently been better than Google about their hardware + software lifecycles.

It's totally fair to compare the two, but that doesn't necessarily make them equivalent. Valve's track record isn't good, but it's not bad either -- it's not like Google's.

That's silly. The point is that the series of products has been discontinued; if Valve had a v2 version of the Steam Link or Steam Controller, nobody would be talking about this.

A company is allowed to iterate on products, and yes that sometimes includes discontinuing a specific product, but that's not necessarily the same as abandoning a market.

Nintendo has discontinued the GameBoy / DS line. Because they now sell the Switch Lite, which is their new way to solve that problem. You can argue "Nintendo sucks, they discontinued their handhelds, we can't trust them on hardware", and they sort-of have. But they gave that product line 10 years of support, and they haven't abandoned those folks, the Switch Lite exists for them now to bridge over to the new mobile gaming market. Not all "discontinuings" are of equivilant weight as Google's product-killings.

Steam doesn't sell a Steam Controller, true. Because much of the Steam Controller stuff now supports nearly any controller (Switch Pro / DualShock 4 / Xbox One XINPUT, including the full accessibility controller / etc). Yeah, it's not quite the same as the Steam Controller (no touchpad joysticks), but they didn't abandon that market. And every Steam Controller ever sold is still fully supported in software, even in new games made this year (6 years after launch).

Steam doesn't sell a Steam Link. But they didn't abandon it, they upgraded it. Steam Link now works on Windows, Mac, Linux, Apple iPhone, iPad, Apple TV, Android TV + Android Phones, Samsung Smart TVs, and such. They don't sell you their own box anymore, because that would be super redundant, most of the TV boxes people already own can run Steam Link already, and they continue to support Steam Link on many different clients to this day. Last time I checked, even original v1 Steam Links are still fully supported in all software + service (6+ years after launch).

---

In comparison, Google often just hard up kills stuff. Google Reader died, replaced by nothing. Android Things died, replaced by nothing. Google Cloud Print died, replaced by nothing. There's an entire website dedicated to just listing random stuff Google's killed - https://killedbygoogle.com/ . At least half of those products or services have no graceful fallover or upgrade path to anything else.

I get that it's a fair comparison. But I strongly believe it's not an equivalent one. As someone who has bought plenty of Google and Valve hardware over the years, I'd be a lot happier with Google as a company if they put even just half the effort on the support and maintenance lifecycle for their products, that Valve already does for nearly everything they touch.

Going by the dates on Wikipedia, the 2015 Chromebook Pixel had 6 years and 3 months of updates. The Pixelbook will have 6 years and 6 months, the Pixel Slate 7 years and 9 months, and the Pixelbook Go 6 years and 9 months.

1st gen Chromecast stopped getting feature updates after 6 years, but as far as I can tell is still in scope for security fixes. 2nd gen Chromecast seems to still be fully supported 5.5 years after release with no EOL announced yet.

These all seem like very reasonable amounts of time. How long do you think support should actually last?

Really, an x86 laptop should get software updates until the wheels fall off, or something actually needs a CPU extension that's not present.

If you want to be petty, don't add new quality of life features to devices past the sell by date, like say no Android apps on older Chromebooks, even if they've got the hardware virtualization support needed. But when you kill the existing printer service, it's kind of rude to not push the replacement local printer daemon, or leave a half-ass version that doesn't really work either.

"Consistently positive reviews from users" are not enough.

Steam Link became obsolete, and the controller just didn't sell (despite being excellent IMO!)

This could see a similar fate, or it could be the breakout hit made possible by their learning from past hardware

Steam controller suffered from doing too much, I think. People want plug-in and go from their couch games, having a fully programmable all-singing all-dance super controller turned out to be more effort than most wanted to deal with.

Steams built in controller customization features are worth it alone - for every controller. No digi pad makes them bad for retro games but I have others for that.

The main value-prop for the steam controller was its touch pads, which allowed you to play mouse-based games from the couch (in an era when controller support for PC games was spotty). They were a cool idea, and they worked, but what ended up happening is all the games just added support for regular controllers, and that was just a better experience.

> the controller just didn't sell

From my perspective, the controller was never sold. I could not buy it anywhere, I am assuming they were only made in a relatively small batch and sold mostly in mainland US.

I had been looking since it was released for a place to buy it at a reasonable price, but could only find exorbitantly priced resellers that would ship from USA.

They sold at least 1.3 million steam controllers [1]. That's relatively small compared to the 3 major console's controllers, but it's still a lot. IIRC the only place Valve officially sold it was on Steam itself. Based on forum posts from the time and Valve's Index shipping FAQ, they could ship it to the USA and Europe. While you can't buy it there anymore, its store page is still up [2]

[1] I can't link it directly, so look for "Steam Controllers are played with a diverse set of games" https://steamcommunity.com/games/593110/announcements/detail...

[2] https://store.steampowered.com/app/353370/Steam_Controller/

>IIRC the only place Valve officially sold it was on Steam itself.

As far as I know, that's correct, but it was region locked. You could not officially buy it where I live.

Valve has the same problem Google has: no new venture will ever have the same margins as their primary cash cow and therefore won't be worth maintaining.

And this is why you separate your business units so that your hardware peeps can manage their own P/L.

On the other hand, they need to protect themselves from threats to their cash cow. Hardware and operating systems play a part in that, which is why Google needs Android to ensure access to mobile eyeballs.

Valve created Steam Machines as a hedge against the Windows Store becoming like the App Store (walled garden) but neither took off. Whether VR will become a major part of gaming is yet to be seen, although Facebook has invested heavily in their Oculus Quest ecosystem which does not require Steam or Windows.

All of those are still supported. Pulling a Google would require them to stop supporting them.

Everybody stops building things that don't sell well. Very few continue supporting them.

I would not hesitate to buy hardware from Valve.

Differentiation might include some sort of game+device compatability badge or certification, built in steam controller and all that customization, and game discounts/rental program when played on the device.

I own a Steam Controller, Steam Link, and a Valve Index. They're all excellent devices and I really wish I had purchased a second Steam Controller and Steam Link when they were clearing their inventory. Even after discontinuing the hardware, the devices have continued to function as well as they ever have (actually even better thanks to continued updates).

I'm not surprised Valve has trouble continuing to produce hardware with low sales volumes. That said, they have certainly proven themselves to me with the quality of their devices so this one naturally has my attention. I don't buy products until reviews are out, but this one sounds pretty good on paper. The GPD Win 3, Aya Neo, and One Xplayer have all already proved the concept is viable and I'm excited to see Valve's take on it.

Also, to be fair the whole reason they stopped making Steam Links is because there are many other products out there now which can run the Steam Link software. Their hardware offering simply became redundant. There's no shame in discontinuing a product which no longer makes sense for the overwhelming majority of potential buyers.

Valve's efforts in hardware (and Linux support) are motivated by getting leverage in negotiations with Microsoft.

After all, Microsoft could pull an iOS and insist on a cut of all sales on Windows - or make the microsoft store the only way to distribute software.

Like a pilot with a parachute, they'll dutifully maintain it - but they hope not to need it.

steam controller is amazing if only due to the backside buttons. these are apparently patented by a patent troll.

Why did Valve get out of the AAA game development business? It seems like hardware has largely been a mistake for them. Sure, I can have 20 different variants of an x86 computer and various HMD interfaces, but if all the games suck ass what's the point?

Valve is still surfing on the inertia of their initial software-based innovations, and is only continuing to cruise along by siphoning off others in their (lightly) curated market.

How much money is sitting in the coffers at Valve Software? What practical force prevents leadership from allocating just 1% of that pile of cash to something like TF3? It almost feels like we are being intentionally suffocated by our entertainment industry these days. It's the exact same story at Blizzard/Activision, EA, etc., but at least in those cases you can blame the publicly-traded aspect of the organizations.

Artifact was a good card game only killed by really dumb marketing and monetization decisions.

Half-Life Alyx is almost universally hailed as the best VR game and as good if not better than the previous Half-Life games.

That being said, you're right about them neglecting a lot of their IP. Specifically a TF3 might be really cool to see, along with more love for Counter-Strike.

Half-Life: Alyx was definitely nowhere near as good as previous Half-Life games. Graphics and VR don’t determine a good game.

Half-Life 2 and associated technologies put Valve in the position they are today.

I would say Half-Life: Alyx is better. Immersion is much better, gameplay is better. I don't see any reason to prefer HL2 over Alyx. User scores at least indicate they are very, very close to eachother. On steam Alyx has a higher score (98%) then HL2 (96%).

> Half-Life Alyx is almost universally hailed as the best VR game

MS Flight Simulator 2020 and Hellblade: Senua's Sacrifice might be other "best VR game" contenders.

I haven't played Hellblade but FS 2020 can't be even really be called a VR ready game. Even with a 3080 you struggle to reach 90 fps. It's a vomit machine.

Try to increase cache from default 8GB to 64GB, that helped a lot. I play it on a 10980XE/128GB RAM/RTX3090 and HP Reverb G2 at high details and flying over Paris with the latest update is now very smooth.

That’s some billionaire’s PC in most of the world.

It's slower than some $300 Ryzen (5600X), just more threads and memory. It's for VR development.

> Hellblade: Senua's Sacrifice

How so? I wasn't able to play past the intro when they literally letterboxed me in a cutscene.

You probably hit some bug, I was able to finish it completely without any major issues (I think it crashed like twice). Alyx, while cool, is too static, Hellblade features some fast combat movements and both are atmospheric in their own way.

Actually I "ragequitted" (well, not rage, just disappointment) when it seemed to be an adaptation of the 3rd-person perspective.

Might have been infantile on my part, I'll give it a chance again once I repair my rig.

I thought the UI for MSFS in VR was awful.

Though I think some of that may be because of the game controller I use. It's an Astro C40, which due to driver issues, can't decide to tell Windows if it's an XBox 360 controller or an actual Astro C40. If it identifies as a C40, then games don't know what it is and don't have any default bindings and I have to configure it all, and with a game like MSFS 2020, there's a lot to bind. But also, lately, I've been having an issue with it deciding that the controller only has two buttons and nothing else, not even analog sticks.

But I digress...

I think MSFS in VR would be a LOT better if it gets support for controller-free hand tracking, like the Leap Motion. I'd still use a game controller for the main inputs (Flight stick, brakes, rudder, probably throttle as well), but I'd love to be able to just reach and flip light switches, turn dials, etc.

Being the best VR game isn't a very high bar. From what I read from people who play a lot of VR games, the amount of polish was the only thing to make it stand out, and without the Half-Life name, it would just be a slow hallway shooter that didn't do anything innovative like the previous Half-Life games did.

> Half-Life Alyx is almost universally hailed as the best VR game and as good if not better than the previous Half-Life games.

Yet weirdly, Valve refuses to sell their Index VR system to people in Australia.

As I discovered recently, when looking to get one, here in Australia. :/

There are people selling some international version via Amazon, however apparently if any kind of warranty issue happens then you're screwed. So, "no" to that.

Probably related to

https://www.accc.gov.au/media-release/high-court-dismisses-v...

Hardware devices have an even higher cost burden for warranty handling then software products, So given they're not really wanting to do proper warranty on software I don't see hardware being better.

Interestnig. Valve doesn't honour that agreement anyway.

I bought a game a while back that turned out not to work on my system (very clearly), and Value rejected the (several) refund requests I created for it.

So it became a matter of "do I want to do a chargeback on this (and lose my entire Steam account), or not?". Decided to just put up with it, as it wasn't much money.

That being said, I rarely buy via Steam any more, and never full price.

From memory they give refunds based on some time threshold of hours played. So if you’ve played more than say 2 hours, it’s much harder to get a refund

Yes, that's well known. The game in question had about 15 mins of total time, as it _wouldn't run_ on my system.

So, not even close to the threshold. Still, they just kept rejecting it with some form-letter-like excuse (eg no actual conversation), and that's the end of the matter each time. No appeal, etc.

The most recent update for Counter-Strike is honestly really great. It's a difficult game to touch as the community is in a pretty good spot meta game wise. I'd love for them to shake things up faster and more frequently, but as a spectator it's a super competitive game and I love watching it.

Oh, I feel you! CS is one of those games where, at least with regards to weapon balance, it's in such a good spot that the community actively doesn't want many changes. Most people are also content with the same 7 maps and don't play new ones when they're added. Still, I think seeing a proper map rotation with competitive seasons would be really cool, especially since the community makes so many high quality maps! A bit more could be done to battle cheaters though, as those have been getting very bad in the last few months once again.

Definite agree on the watchability, CS is the only e-sport I really watch the big tournaments for, they just nailed the aspect of it being easy to understand and creating a lot of high-adrenaline moments.

Alyx is fun to play but the jokes and dialog are really bad. Previous half life games did a lot more with a lot less.

Artifact was just straight up a trash game.

I hate this new idea that every failed game is just "bad marketing"

I would argue that games lasting 30-60min was an issue as well (for a 1v1 card game)

Valve have too much employee direction to participate in the race to the bottom that is modern video games.

But, to your point-- everyone got out of the AAA game development business. We've had one Elder Scrolls and one GTA for ten years now. All of those insane articles dramatizing Bioshock: Infinite as the last AAA title were true. It's like the Waterworld of games, the last big effects movie before everything became cheap CGI.

Games are lootboxes, middling cartoony style, addiction mechanics with some social stuff lightly sprinkled on top now.

Sidenote: Making TF3 would be especially weird because Valve basically popularized all of those trends.

That's just the format now. Even CDPR and other companies clinging to the old formula are just coasting on past momentum, with the cracks starting to show.

> Making TF3 would be especially weird because Valve basically popularized all of those trends.

> That's just the format now.

It simply doesn't need to be this way. It would take 1 executive decision to change all of this.

Sony manages to pull this off still, if only to sell hardware.

The Last of Us pt. II is an incredible, highly detailed, long single player game with no microtransactions of any kind. Most modern Sony titles are like that.

Yeah the consoles are little islands of genuine content too (Although, a new Zelda could come out a bit faster...). But it's almost like it only works there, as a loss leader.

> Why did Valve get out of the AAA game development business?

Dota 2 and CS:GO are both AAA games that are still under development. Valve never stopped developing.

The more accurate question would be why did Valve get out of the single-player AAA game development business? The answer is risk and profit. The same reason why Disney keeps making Marvel and Star Wars content. Established properties are good earners with very low risk - and there's no lack of rabid fans who love the companies for it.

Or 1% of that cash into improving Steam?

I wonder who will be the target for these devices, because a great part of portable consoles strategy was: to be a gateway for kids to buy into the "big-boy" consoles down the line.

Basically win their hearts and trust early, so they can become loyal throughout their life.

So these devices had to check some boxes, one of the main factors being the price.

These devices had to be cheaper then regular consoles (among other things) - though they shared the same kind of price strategy between each other: sell almost as cost (or at loss in case of the PS3), to make up later with game sales.

Nintendo blurred that line into one device with the Switch, but they still check the box of the device being affordable.

With that said, there were already some approaches to Nintendo Switch like hardware for the PC market, some with waaaay better performance then the switch (with both intel or AMD socs), yet they are failing in what I believe is crucial for these devices: price.

They are expensive.

I'm not saying they aren't wonderful pieces of engineering, but clearly they aren't meant to even compete on the same level. They can't sell at cost/loss, because they can't make it up with licensing or their own IP, so they have to make money on the hardware itself.

Therefore I always assume these devices are for a very particular small set of gaming enthusiasts who are found of the idea of playing PC games on the go (even with limited battery life), and are willing to pay more then a full fledged console for this experience.

I'm curious to see how Valve will tackle this, or if they are fine with that (which they seemed to be in other hardware releases).

Slap steam on it and take the Steam cut?

That's the tricky part, if these devices aren't locked to a proprietary OS/marketplace, users can shop around and you're basically helping your competitors while taking all the losses.

If it's running SteamOS (and it will be, even if there's Windows compatibility Valve probably doesn't want to pay for licenses), it might as well be locked to Steam for 90% of people. Epic's launcher doesn't have a Linux version (since they've claimed as have many other developers that the Linux community is 1% of users and 90% of complaints due to distro differences), nor does EA, Ubisoft, etc. They could port it, but by time they had something user friendly and working (which may be difficult since you probably won't get more than Steam Big Picture which doesn't have support for installing non-Steam applications at the moment) Valve would've sold a boatload of these and a bunch of Steam games.

Edit: not to mention anti-cheat, which is a whole other kettle of fish when you have to integrate into a entirely other kernel with different plumbing. Valve's anti-cheat doesn't rely on kernel modules or a ton of client side software at all, so they have a cross-platform advantage there too.

Yeah this is where valve's investment into Linux would potentially pay off. Any platform they make could run Linux and work really well while basically all of steam's competitors either haven't done anything for Linux or have actively avoided/worked against it.

Let's assume the SteamOS would make this device 100% locked to Valve's ecosystem, which would give Valve the ability to sell these devices at cost (by charging fee's on games sold). Let's set the price on the same ballpark of the current consoles prices.

What would make people choose this platform over Windows, PlayStation, Xbox, or Nintendo?

This is where IP comes to play, and why Valve shouldn't have dropped the ball with their game development investments, and they would have exclusive games to get the ball rolling.

What percentage of the PC gaming market does steam have? Even if competitors like Gog and Epic get in the console, it won't change the percentages that much.

They could still do the "game sales will offset hardware" idea, right? They could offer a platform locked down to Steam. Sure, it will be hacked at some point, but it's not a comfortable general-use device, so they can expect most people to actually buy games to play on it.

The way I'd see this working is if it's not for playing games locally at all. Just boot straight into Big Picture, and allow for streaming games either from your local network via Steam Link or from a Steam cloud streaming service that gets announced together with this.

I think you could hit a really attractive price point for a a device like that, and it seems like a compelling device for Steam's existing user base. At least for me, I think it'd mean tossing the Switch into the junk drawer.

Targeting it for playing games locally just makes no sense. It's either going to be too expensive to have any real market, or too slow to run the stuff modern PC gamers want to run. Either way, it'll be a niche device that Valve shouldn't waste their time with (PC manufacturers will do it anyway). But nobody is making a dedicated portable streaming device as far as I know, and it seems like something that'd fit really well with a streaming service.

I think I agree with you pretty much on everything.

>But nobody is making a dedicated portable streaming device as far as I know, and it seems like something that'd fit really well with a streaming service.

NVIDIA kind of did it with the NVIDIA Shield some years ago, and I think they're still supporting it.

My recollection is that the original Shield was marketed as much for playing (Android) games locally as for streaming, and had a much more powerful GPU and CPU than contemporary mobile phones did. That's kind of losing the advantage of it being dedicated to streaming, which is that you could really cut down on the BOM.

Streaming is still in a awkward position, both user experience wise, the cost of the streaming games (with data caps), and the competitive aspect of some games.

People get blown away by the amount of data used by NVIDIA solution, Google Stadia is pretty much dead...

It's hard to say if streaming would be the win condition.

>NVIDIA kind of did it with the NVIDIA Shield some years ago, and I think they're still supporting it.

Shield no longer receives updates. GeForce NOW still runs on it, though. https://www.youtube.com/watch?v=pSWCJxS0-HA

Shield TV, the Android TV-powered settop box that most people nowadays call "Nvidia Shield", is still supported going back to the first version in 2015.

Why not just buy a solid controller for your phone or tablet?

How could this hypothetical device possibly be better than an iPhone 12 Pro Max, Note, Galaxy S, etc?

by having access to the steam pc game library, which is huge

Both major mobile OS platforms support steam link.

The games have to be run on some machine though

Just to add to your comment, you can already turn any phone to a "streaming Switch" for $30, there are dozens of controllers like this:

https://smile.amazon.co.uk/dp/B08BRX3RT3/

the Steam Link android app lets you stream any of your games to your phone like this already. With a little more setup you can stream Wii games to your phone through Moonlight from CEMU and run them at 1440p60, while the Switch can barely do 720p30.

I could certainly see this being the flagship for the launch of a steam cloud streaming service a la Stadia, Luna, geforce now, etc. That does beg the question of how they would price it. I think they would have to price the hardware very aggressively or sell at a loss -- the cost of entry on these other streaming services is what, $50-60 for a controller? Although if gamers with big steam libraries could bring them with them to a streaming platform rather than re-buying then that would be a major differentiator.

On the local streaming front, I've struggled to do in-home streaming in a comfortable way with anything but a hardwired network between the PCs streaming. Maybe this will be different if they optimize it for wireless streaming, but on laptop wifi -> desktop ethernet I found it to be fairly high latency and lossy.

Mine is probably a niche use case, but I will absolutely jump on one of these simply due to my lack of storage space.

I live in a 450 square foot loft. I don’t have space for a tv here, so consoles were out of the question until the Switch came along. There is no room in my life — literally — for a gaming PC, but if they can cram that into another Switch? Sign me up.

It’ll be a big draw for university students in dorms as well.

Take a look at the GPD devices in that case, they might suit your exact need if you're ok with the price they're asking.

Also, I don't feel people talk about this much but gaming desktops tend to be noisy, ugly, and/or gaudy on top of taking up lots room. I'd love to just get rid of this box but I don't care enough about games to buy a gaming laptop, which just only slightly solves the ugly/gaudy problem for me. I'd love to just stream going forward and never have to put together another PC or buy another expensive video card again and having the ability to do this on a handheld sounds very tempting!

If you take some time to plan it out and pay attention to thermals, you can absolutely make a relatively unassuming Micro-ATX build nowadays, and just have a sleek small(er) black box sitting there.

https://www.reddit.com/r/sffpc/

Noise is never an issue when you dont use stock coolers. Undervolting current gen gpus reduces a lot of noise and heat. A ps5 takes up more space than a PC.

No need for 2.5/3.5" drives gives you a lot of room to get rid of ;)

> yet they are failing in what I believe is crucial for these devices: price.

IMO Before coming to price, none of the Switch-likes are good products for the broader market. They lack the UX, polish, brand, games, and pretty much everything that makes Nintendo consoles such a big success.

I wouldn't compare the experience of playing Super Mario Odyssey on Switch and some old game on a janky emulator or some AA game on 20fps and broken UI.

Even if they had the same price tag, I wouldn't expect something like Aya Neo to eat into Nintendo's market share. Only a fully developed platform can do that.

Actually I wasn't mentioning the android devices, or the ones dedicated to be emulation machines, but kind of a "newish" win10 machines where you can install pretty much what ever you want (from PC games to emulators), but come at a high price tag (600€++).

Everyone has to start somewhere, and history has shown that pricing was a major factor for the success of consoles (besides ofc IP, distribution, branding).

You have great examples, like Gameboy VS Gamegear or Playstation VS Sega Saturn.

Thing is whilst the device might be comparatively expensive to the Switch, the games will be able to be played for the entire lifetime of the Steam online store and on multiple platforms (SteamPal itself, laptops, desktops, any device with SteamLink, cloud gaming like GeForce NOW, any future devices etc).

> ...to be a gateway for kids to buy into the "big-boy" consoles

Game consoles are like vinyl record players. Despite the huge nostalgia market in a demographic with lots of disposable income and lots of potential free time due to a looming retirement, to "kids these days" it's already a legacy format that's dead from any practical perspective.

Honestly at this point I kinda feel like Valve is to the Steam store what Google is to search/ads. I do not trust either of them to follow through with any amount of support for any project they announce outside of the massive money spigot that is their core business.

AMD recently patented their DLSS competitor (albeit with a really dumb name, "Radeon Gaming Super Resolution").[1]

If this launches on RDNA2 as rumored, it could be very interesting to see it maybe having a 1080p screen and just upscaling a 720p rendered image to that native resolution to save on computing power.

[1]https://www.techradar.com/news/amd-patents-gaming-super-reso...

I am surprised nobody mentioned GPD yet

https://www.youtube.com/watch?v=YNfbfPhOiE0

There is also the Aya Neo[0], the One Xplayer[1], and even the (now quite old) Razer Edge[2].

[0] https://www.ayaneo.com/aya-neo

[1] https://www.indiegogo.com/projects/onexplayer-best-performin...

[2] https://www2.razer.com/gaming-systems/razer-edge

Seeing how there is now the Aya Neo and the Onenotebook Onexplayer, this might actually work. Have a handheld with a powerful enough laptop chip and a Thunderbolt, when at home it sits in an eGPU dock and use it like a traditional desktop, and on the go either connect the game controllers or a keyboard to use as a laptop.

I worry that it would get abandoned rather quickly given how Valve has released hardware in the past.

What about the HTC Vive and the Valve Index?

I suppose their VR hardware would be the exception.

SteamOS, Steam Machines and their controller all got abandoned even though they all seemed like pretty good ideas.

Those are still getting software updates even though they aren’t on the market anymore. Additionally, they spawned the “Big Picture” mode, an absurdly robust controller mapping feature, and a Wine variant that benefits all Linux gamers.

Every time Valve makes a hardware product, Steam gets some new killer feature.

A lot of consoles are sold at a loss, where costs are made up by people buying games through the console on the makers store, I wonder how valve will go about this because I don't see a lot of people getting this that didn't already have steam on another device.

It also really doesn't seem like they have anyone with the skills in their wheelhouse to do good by this.

A big part of differentiating a console from a PC is building an operating system that prioritizes the game process uber alles. It's not just a fancy window manager slapped onto a commodity PC. Even the original Xbox had a lot of custom OS work done to it, stuff that would eventually serve as the test bed for developments in DirectX.

Does Valve even have any operating system developers? I thought their core dev body was mainly focused on screwing modders and getting kids addicted to slot machines.

They are doing a lot of work on Wine as part of Steam Proton (the Wine distribution built into the Linux version of Steam), so I'd imagine they have at least some of this type of developer on staff.

> A lot of consoles are sold at a loss

I don't believe that's the case anymore. Newer console generations have moved on from exotic ISAs and the like towards using a lot of off-the-shelf parts for smartphones or PCs. The Switch, for instance, has been sold at a profit from day one [1].

[1] https://web.archive.org/web/20170131084504/https://www.bloom...

The Switch also has very little power compared to gaming consoles. It's the same price as an Xbox One S, but with far less processing power.

Not to split hairs but it's the same retail price as an Xbox _series_ S, which is a generation newer. Due to supply issues you are mostly correct, in practical terms.

It's portable though. A mobile phone has far less processing power than a workstation or laptop (and some are nearly as expensive) but it's also a fraction of the size / weight / power consumption / heat output.

I don't believe it's ever been the case. On some days (especially early on) the console may sell at a loss, but over the lifetime of the console, the hardware is almost certainly net profitable.

I personally really like the idea. I'd love to be able to casually play PC games on the road without having to drag a laptop around. Of course unless it has something particularly appealing, it will have to be priced around the switch/lite level.

That's very interesting to hear. Not least because Steam hosts your multiplayer sessions for any game now, so it could potentially be very seamless. There is so much potential here! I hope they work hard on their voice chat feature too, and get everything right from the start. And finally, Linux, using Vulcan I assume!

This is very complicated and interconnected market segment.

You cant just play games on this device out of the box (some games yes) you need some game devs to adopt it and have releases on this yet another platform.

Valve would have to have a dedicated support team and long term strategy for it to become a switch rival.

And with Valve track record it will be an abandonware from the get go.

What support is needed outside of controller support and whatever OS it runs (I'm assuming linux)?

And Steam have been collecting that data for a while now - each game is tagged with no/partial/full controller support. They know what will work OOTB.

They need to secure games on it. Even if this console ends up being PC in small form factor.

I dont have switch but if i were to buy a console it would be switch, not steam console. I can play my games on PC, if i get console I'd rather have games made specifically for that form factor, plus Nintendo has amazing catalogue of exclusives.

If you judge Valves actins in response to Epic agresively taking their already entrenched market - basically nothing. What chance do they have to compete with Nintendo, or mobiles?

I might come across negative, but Valve doesn't make it easier.

I understand getting a Switch for the exclusives, but outside of that I wouldn't get a portable console for access to games I couldn't play otherwise, but access to games outside of the normal environment (my desktop PC).

If you look at the metacritic list of best games for Switch:

https://www.metacritic.com/browse/games/score/metascore/all/...

Most games are just ported PC games, which would be playable on a Steam portable console (assuming they're available on linux).

Most games are made for consoles first, rarely other way around.

Anyhow, you can play on your screen with docked switch. You can play on your couch with steam link, or pc hooked up to tv.

I am just not seeing a competitive edge for this device.

> Most games are made for consoles first, rarely other way around.

Except in the actual case I mentioned with the list, where almost all of the cross-platform ones are made for PC first, from Divinity to Hades to Undertale to Inside, Steamworld, Shovel Knight, Hollow Knight, Ori, Bastion, Dead Cells, Into the Breach, Fez, etc etc etc.

Seeing how basically all games that aren't first party or specific exclusives are lighter (hardware wise) PC games actually speaks to a huge market for those games on a portable device.

> Anyhow, you can play on your screen with docked switch. You can play on your couch with steam link, or pc hooked up to tv.

Playing on the couch would be competing with Xbox and Playstation, not Switch, which as you said is already a solved problem.

This would be for anywhere you play your Switch undocked, and you don't want a Switch for whatever reason (likely more indies, emulators, "apps").

> This would be for anywhere you play your Switch undocked, and you don't want a Switch for whatever reason (likely more indies, emulators, "apps").

And now we came the full circle. Back to the original point. I don't believe Valve is capable of competition at this point.

I would require serious people and money effort to create this 'whatever reason'. Because I gave you a big one for Switch and I don't see one for Valve (obscure small indie games that don't get ported to Switch is hardly a selling point).

I think you missed for the third time the point that almost all Switch games that aren't Mario, Zelda and the likes are specifically "obscure indie games". And there are multitudes more that aren't ported to Switch.

And even the ones that are, would be available here as well, along with potential other features that Switch doesn't have (over the air streaming games), etc.

Personally, if I were to buy a Switch, I would have to buy all the games I wanted to play on their for, I assume, $50 each. With something like this I own half of the games on the Switch top 100 games already, on Steam. If it could access Humble Bundle games etc too, would be a plus.

But the price point will probably play a big part.

People buy consoles because they expect a perfect out of the box experience. If you try to play a GPU intensive 3D shooter on the handheld the performance or graphical quality will be terrible. The game must support full screen and the UI shouldn't be too small to become unusable on a small screen.

Seems like I finally found some use for my CS:GO junk and selling them...

Might as well pick up reasonable device for some mobile use. It probably supports bluetooth stuff and steam controller, so shouldn't be too horrible experience.

What IP does Valve own that could run on a portable device? Do they think people want to play Portal 2 on the go? As if indie developers want to put resources into porting their content on some niche handheld. Valve has really lost the plot.

They don't have any assurance in hardware. Steam, while successful, isn't the only contender on PC gaming. It's not 2007 anymore, there's Epic Games, Origin, and I'm sure a bunch of other platforms all hosting a small set of popular titles.

They're out of touch, which is sort of ironic given the company was founded by ex-Microsoft employees in the 90's. They went from a Game Development Studio into one of many Game distributors. They're better off sticking with that, or at the most, pushing their platform onto the Switch. Get with the times.

Depends on their model... but a very large selection of PC games already ship with gamepad support (I mean, a huge selection of PC games already run on other consoles for example). I mean, that's really it. If you have gamepad support and can run acceptably on the hardware, then you're good to go - basically no extra dev required.

The device will run on Linux, depending on the existing title, that might not be a breeze. Not to mention the optimizations developers will have to make to scale their games down to run on this thing.

They have tried to make a tabletop console thing for l4d where 4 players play l4d as usual and the 5th player would spawn zombies or items on them. I suppose this console is related to that

So... basically GPD Win?

https://en.wikipedia.org/wiki/GPD_Win_3

A portable gaming PC means optimization - in terms of performance and interface - is going to be critically important.

But knowing Valve, they will keep the same old exorbitant 30% commission which makes it uneconomical to actually launch great titles targeted at this device, same as for the Index VR device.

By the way, on Switch the effective commission is 25%, given that customers are rebated 5% of their eshop spend. Nintendo also bothers to apply curation and quality control to their store, which Steam does not.

Good analysis, but you harm your credibility by saying Nintendo curates eshop titles.

This is only technically true, and no layperson would be able to tell. 99% of the eshop is ports of shovelware android ‘games’ that consist of little more than cover art.

I have two titles on Switch and Steam. What you are saying is incorrect. The process to get approved and launch on Switch is substantially more difficult than Steam, and the quality of titles is as a result much higher.

Because of the lack of curation and quality control, I make a lot more revenue for the same product on Switch than on Steam.

Just picked up the new macbook air m1 and I am staggered to find that I can play starcraft 2 and bioshock decently on a laptop that draws 30watts and is capacitivily cooled... Really just unbelievable.

So if valve could someone get ahold of some cutting edge arm processors in the next few years then this could be very viable(though I doubt they will get it out before the next version of the nintendo switch).

Valve has money, but not Apple or Microsoft money. I doubt they'd be able to source a high-power arm chip + the necessary hard/software features to make something like the rosetta compatibility mode possible.

Not necessarily right now but a few years down the road when it actually releases(though no one knows when they'll actually release knowing valve's track record).

Not sure why people are downvoting me....

According to the article they are shooting for a late 2021 release date if nothing goes wrong. Fingers crossed they'll say more at E3, which they're attending this year.

Not sure why people are downvoting you either though.

Why not making a phone and enter mobile market?

Phone manufacturers are all just selling tweaked Qualcomm reference designs. Unless you want to go the route Pine64 or purism are going or have the resources to build your own SoC like Apple or Nvidia did you will have no control over the device. This is why even high end Android phones have such short support periods.

Sounds like the Aya Neo...

If only it also had a Thunderbolt for an eGPU.

Making a PC like console might backfire. PC games are full of botters and cheaters (not saying you can't cheat on console, but I see 1-2 cheaters in a year on console compared to 50% of PC games population). It's the main reason I play games on console only.

Not everyone is into multiplayer games. There's a lot of single player market available.

And non competitive multiplayer. And multiplayer within a group of friends.

“50% of the population is cheating in PC multiplayer games” is wild. Can you cite a source for this absurd claim?

Reply on news.ycombinator.com