Qualcomm’s new Snapdragon G series processors are designed for handheld game consoles like the upcoming AYA Neo Pocket S, which is expected to ship with an Android-based operating system.

But in an interview with The Verge, Qualcomm’s senior gaming director Mithun Chandrasekhar says the chip could eventually power devices running other operating systems as well.

When Qualcomm launched the Snapdragon G3x Gen 1 processor a few years ago, it was basically a modified version of a Snapdragon 888 smartphone processor. But the new Snapdragon 3Gx Gen 2 is not designed for phones.

It delivers up to twice the graphics performance of its predecessor, up to 30% faster CPU performance, and has peak power consumption of around 15 to 18 watts, putting it in the same category as entry-level laptop processors or the custom AMD chip that powers Valve’s Steam Deck.

But Android is kind of a strange platform to target when launching a chip with that level of horsepower. Most Android games are designed for phones, so they run just fine on devices with lower-power, lower-performance chips. And streaming games over the internet or a local network doesn’t really require a high-performance GPU either.

One area where this sort of chip does come in handy is emulation: there are Android emulator apps that let you run everything from Game Boy to Nintendo Switch titles on Android. But while that’s a feature that Chinese companies like Anbernic, Powkiddy, and AYA might play up, emulation is sort of a legal grey area (the emulation apps may be legal, but where exactly are you getting your games from?) and it’s unlikely we’ll see Qualcomm acknowledge this use anytime soon.

But one area where we’re seeing a lot of growth in the mobile gaming space? Handhelds with x86 processors that either run Windows (which most PC games target) or Linux (which Valve has made into a viable gaming platform with its SteamOS operating system and Proton software that allows thousands of Windows games to run on Linux).

And it looks like Qualcomm may want to play in that same space. You can already find a handful of Windows laptops and tablets with Qualcomm processors, and it’s possible that we could eventually see gaming handhelds that pair Snapdragon G series processors with Windows or other operating systems… and Chandrasekhar emphasizes in his interview with The Verge that the company is experimenting with operating systems other than Android or Windows.

One of the things I’ve read in nearly every review of the Asus ROG Ally handheld gaming PC is that one of its key selling points is also one of its key weaknesses: it ships with Windows. That means that most PC games are supported out of the box without the need for any sort of compatibility layer. But it also means that the underlying operating system is designed for keyboard and mouse input rather than for small handheld devices with touchscreen displays and game controllers.

Chandrasekhar agrees that the Windows ecosystem isn’t really built for handheld game systems (at least not yet), and so the company is testing Proton-like software that could allow users to play Windows games on non-Windows (or Android) operating systems.

So it’s possible that we could see a Steam Deck rival in the next few years with a Linux-based operating system, support for Windows games, and an ARM processor rather than an x86 chip.

But it’s also possible that we won’t. It’s still early days for Qualcomm’s new handheld gaming plans, and it’s not like Intel and AMD are going to sit around and wait for Qualcomm to catch up. And while Qualcomm offers reference designs to hardware partners, even if the company does make competitive chips, it will still be up to device makers to decide whether to use them or stick with AMD (which seems to dominate the space at the moment).

By the way, if you want a closer look at Qualcomm’s new Snapdragon G3x Gen 2 handheld gaming reference design, I recommend checking out The Verge’s article, which has a number of real-world pictures of the device.

Support Liliputing

Liliputing's primary sources of revenue are advertising and affiliate links (if you click the "Shop" button at the top of the page and buy something on Amazon, for example, we'll get a small commission).

But there are several ways you can support the site directly even if you're using an ad blocker* and hate online shopping.

Contribute to our Patreon campaign


Contribute via PayPal

* If you are using an ad blocker like uBlock Origin and seeing a pop-up message at the bottom of the screen, we have a guide that may help you disable it.

Subscribe to Liliputing via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 9,446 other subscribers

Join the Conversation


Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

  1. There’s no legal grey area in emulation. The bios is from my machine, the games are ripped from my own media. Companies who try to fuck with emulator creators can screw themselves

    1. It’s absolutely a grey area, not just in terms of laws, but lawsuits. The deniability is easy for you as the user, but not so easy for a company like Qualcomm.

      The emulators themselves are legal, but the resources needed to run the emulators are not legal to distribute (BIOS, games, etc). Even if Qualcomm kept their hands clean of that part, there’s not adequate assurance that they would be immune to accusations of implying the need for it on the part of their customers.

      The reason it’s a grey area is because, while there aren’t sufficient laws to protect video game companies from emulators being distributed, there are equally insufficient laws or precedent to protect people who develop the methods and tools needed to support these emulators. There have been many successful lawsuits against the teams who make hardware and software needed to extract data, or circumvent security that protects this data.

      Also, considering the fact that emulation development requires the deniability of “clean room engineering”, there’s also the potential that Qualcomm’s history with various other technology partners could further expose them to legal issues.

      This might be a little far fetched of an example, but Qualcomm and Hitachi have had a long standing partnership. Hitachi has made CPUs and other hardware for video game consoles (like the Dreamcast). If Qualcomm uses some technology in their Snapdragon products that was acquired through that partnership, it could give Sega an avenue to sue Qualcomm if they marketed (or implied) a product as being suitable for Dreamcast emulation.

      This scenario seems even more possible when you consider that Hitachi developed the SH4 CPU to work closely with Windows and DirectX on the Dreamcast, which could easily conflict with the work Qualcomm would need to put into their “Proton-like software” being discussed in this article (it would be emulating Windows and DirectX components on ARM.

      I think the bigger risk for Qualcomm here is burning their bridges with video game companies. Their competitor, Nvidia, is doing well with Nintendo right now. A similar deal for Qualcomm with a competitor would be ideal for them, but getting too friendly with the emulator crowd could sour a deal like that.

  2. If Qualcomm wants to jump into the high-end gaming space, they need to port SteamOS and Steam client to ARM and invest some money into quick development of Box86/64 to a stable state. If they commit to supporting Linux and open source like Valve does with Wine and bundle all these software components into a working system image for handhelds running their Snapdragon, then there would immediate excitement and they would be able to compete right away.

    1. It’s sort of up to Valve to port SteamOS to ARM due to copyright and stuff, but even if they did they’d still have to come up with a way to emulate x86/64 or automatically re-compile binaries during installation or something, that always works despite the complexity, since the games were all compiled for x86/64.

      1. Actually SteamOS is open source so anyone can fork their own distribution – that is how HoloISO came about


        Also since there exists archlinuxarm.org with many development packages that can be used to bootstrap from, it would not take much effort for Qualcomm to bring up an optimized SteamOS to their chips if they wanted to. However in the case of Qualcomm since they and Valve are both companies, Qualcomm should partner with Valve instead so they can get direct access to Valve’s private repo and offer to support and do all the development work behind the scenes so Valve would not have to invest their own resources. But this will allow Qualcomm to provide an iso of SteamOS Snapdragon Edition from their site that would be close as possible to SteamOS for x86_64 but for native ARM.

        As for x86_64 games, the answer is box:


        Instead of emulating or recompiling, the software translates existing x86_64 to ARM. It is exactly like Wine and Proton, therefore is not legally ambiguous. If Qualcomm were smart and reinventing the wheel, they would take a playbook directly from Valve and pay box developers to polish the code for Snapdragon and speed up the release.

        It will be a win for Valve as well because of course expanding Steam games onto more platforms. Heck if the performance results prove successful Qualcomm should offer to partner with Valve further on releasing a cheaper $200 handheld to supplement the current Steam Decks. It would be smaller and fanless using Snapdragon but still maintain the draw of current Steam Deck handhelds like keeping an SSD slot to allow user upgrades. This Steam Cloud device would clearly be marketed towards cloud gaming, casual gamers, kids. But with the box software being developed they can start sneaking in some older Windows games as well. So all the necessary pieces are currently there already with a path forward for Qualcomm to take with minimal cost on their part.

  3. At least for my gaming preferences, iOS and Android gaming never took off.

    Outside possibly getting Nintendo or another gaming company to use this chip for their next handheld, I’m skeptical this chip will be used with a Linux platform to run PC games. Or has Qualcomm and their partners been supporting the various x86/ARM translation project (they seem very far away until they’d be ready for a production/commercial device)? Even less likely they’ll get game devs to port to Linux on ARM.

    1. Yeah. They either get a console maker to use the chip for a console or get it in a handheld that can somehow play PC games with good performance (translation in Windows/Linux or devs port to ARM).

      Neither of these sound likely before this chip becomes outdated.

      1. Not gonna happen.
        If Qualcomm is ever lucky to be in a position to make the next Nintendo 4DS, Super Switch, PSP v3, or Xbag…. well those device companies will not want this, nor would they want what they have currently. They are going to want something custom.

        What this is, this is Qualcomm penny pinching. They’re making so much money, and with the scraps of less-great chips they want to dump them into UMPC sector. That might not be a bad thing, depending on price.

        Here’s why it’s not a good thing. They have the software stack for AndroidOS. They are not going to spend the resources to properly port Linux Distro unto this. They are merely saying “hey remember when we didn’t want you to do that, well here, we won’t stop you”. But it’s basically upto the likes of GPD, Aya Neo, Anbernic to do the heavy lifting and spend millions of dollars to build the software. Those companies could afford it, but they are petty and cheap, so they won’t.

        Valve potentially could, but they should not. Let Qualcomm stew in their own greed. Valve is already super busy trying to cram AAA-Title games down from a 150W Level to a 15W level, and do a Windows-Linux conversion on-top of that (DX-OGVLK). If Valve would ever go towards ARM they would want something great, and not this mediocre stuff Qualcomm is pushing.

        With that all said, at least the Snapdragon Software Stack itself is solid. The MediaTek Dimensity still has its issues. Whilst Exynos, Google Silicon, HiSilicon are all slacking behind. Then you have other SoC Providers and it’s a minefield of weird software from MediaTek Helio, Unisoc Tiger, RockChip, AMLogic, Allwinner, etc etc.

        Between all these SoC’s, I would much prefer the QC 8g2 from the AYN Odin2. Or perhaps the QC 8CX g3 which is supported by Android/ChromeOS/Windows 11S…. BUT instead of all of those I would instead prefer an AMD r7-7840u inside my portable handheld/tablet/ultrabook/etc etc.

        If I were to choose ARM, it would be with the caveat that I’m giving up access to legacy software, backwards compatibility, and AAA-Titles… so in return the device must inhabit a different niche (longer battery life, passive cooled, pocketable).

    2. Too bad iOS/Android games usually target the lowest common denominator phone without a controller accessory + touch controls didn’t really take off much in creating compelling (to me) games and your fingers always covering the screen is not a good experience. Meanwhile, PC games usually don’t target potato PCs.

      As for Qualcomm finding a partner to create a handheld + platform to play PC games. I agree that it’s going to be difficult for Qualcomm to do that and I doubt they’d make their own.

      Also, Qualcomm isn’t exactly known for supporting open source much (very much the opposite) and coupled with fairly short-term proprietary driver support/updates so I doubt a Linux handheld playing Windows PC games has a high chance of happening.