Not sure why this made me think of Battletoads..... Back when troubleshooting consisted of pulling out a game cartridge and blowing it.... Must be getting old
.
.
Not sure how much pressure CDPR will get from MS though. MS won't he be happy if CDPR will use Vulkan. The more studios break their lock-in, the more pressure there will be in the opposite direction to support Vulkan on Xbox. And that will make cross platform development cheaper.
Last edited:
Doom was developed by id Software, who traditionally used OpenGL in their games instead of Direct3D, so the move to Vulkan made some sense. But Skyrim (SE) did indeed have to be ported from D3D to Vulkan to run on the Switch, and there are currently jobs like this one available at BGS Maryland for engine programmers with Vulkan or D3D12 experience. Maybe Starfield will support one or both of those APIs, although that does not necessarily imply a Linux release (it can still be an advantage when using Wine).
However, I recall from a conference earlier this year that the CDPR CEO said they have no plans to release their games on the Nintendo Switch, so that means less incentive to invest in Vulkan support, and I do not remember any jobs that require knowledge of the API either (the requirements are always Visual Studio, .NET, DirectX, HLSL, etc., all Microsoft technology), although I do not check them often or extensively. To be honest, I expect CP2077 to be similar to TW3 in terms of technology used, Direct3D (11 may or may not be supported, not sure how many people will still use Windows 7 in 1.5-2 years on PCs that are fast enough to run Cyberpunk), Nvidia GameWorks, and a comparable set of third party software to previous RED engine titles, with some evolutionary changes. In other words, I think the Linux port (if any) would be outsourced to a company like Feral or VP.
Interview with Peter Mulholland who worked on the Linux port of The Witcher 2.
Some key points that he made about why CDPR could drop the Linux port of TW3
1. The resulting performance wasn't adequate (Vulkan wasn't available).
2. TW2 for Linux was profitable, but not super profitable, so they had concerns about expenses.
3. Negative reactions on initial poor performing release of TW2.
Translating this into the context of CP2077, Vulkan today removes issue #1. And lower costs of porting plus grown Linux gaming market should help with #2. #3 is somewhat strange as a reason, since as said above, CDPR were working on TW3 way after that criticism calmed down, and fixed updated versions of TW2 were well accepted as good quality Linux release.
Some key points that he made about why CDPR could drop the Linux port of TW3
1. The resulting performance wasn't adequate (Vulkan wasn't available).
2. TW2 for Linux was profitable, but not super profitable, so they had concerns about expenses.
3. Negative reactions on initial poor performing release of TW2.
Translating this into the context of CP2077, Vulkan today removes issue #1. And lower costs of porting plus grown Linux gaming market should help with #2. #3 is somewhat strange as a reason, since as said above, CDPR were working on TW3 way after that criticism calmed down, and fixed updated versions of TW2 were well accepted as good quality Linux release.
This is a big reason I'm hoping for more Vulkan development by big titles.
The PS4 & Xbox are fully capable of handling Vulkan as they both run on AMD hardware, they just don't support it via the OS, which is crazy because just like Nintendo, Sony & Microsoft are Khronos members, but Nintendo is the only one acting like it.
Honestly, we have the tools to significantly reduce the technical challenges of development, every major platform is capable of running Vulkan, Windows, Linux, MacOS, Switch, iOS, & Android have active support, PS & Xbox are able to they just don't. Cross development would become significantly less expensive & more resources could be spent on the actual game and not just how it renders. It would change the OS landscape though, if it is a significantly smaller challenge to port to other OS' how many publishers will agree to platform exclusive titles and how many users will stick with Windows if other platforms are just as good at gaming?
Nothing is going to change without at least a few bigger titles coming out that just flat refuse to support D3Dx, & it won't be the major AAA's who do this, but once enough do everyone would reap the benefit. Right now if you don't want to be a platform exclusive you are developing against a minimum of 2 APIs, but there are only synthetic barriers preventing it from being 1. Sadly it will take an independent but influential enough company like Ninja Theory or CDPR pushing the agenda, and I just don't see anyone prepping for a battle over that hill.
Interesting. Nice to finally get some details, even if it isn't "official" official.
Regardless, Vulkan should solve both 1 and 3. 2... I think you're going to find some debate there, but maybe the market truly is bigger now as you've said before.
There are no easy measurable public metrics for the market size, but here is an interesting trend for the amount of Linux games on Steam at least:
https://steamdb.info/linux/
From the personal observations, I constantly see posts about former Windows users switching to Linux for gaming, so there is a gradual influx of actual gaming users.
Big distributors like GOG and Steam would have user numbers, but they are not public.
Last edited:
Steam is apparently looking to release some sort of alternative to Steam Spy, so we might see that sort of information soon(ish).
First of all, a lot of people are already using Linux for gaming (feel free to switch). So this is great news. But it's also important for professional applications that rely on graphics, not just games. Many of them are used on Linux.
These benchmarks however are interesting even if you aren't using Linux. Since Nvidia is using basically the same driver on Windows and Linux with just a system shim over it, and assuming claims that Nvidia driver is well optimized and mostly rarely blocks hardware potential are true, the above benchmarks mean that AMD hardware is indeed competitive (in addition to Mesa and Linux kernel driver for AMD being now well optimized).
So this shows that in hardware, Nvidia doesn't have an edge.
Last edited:
Steam doesn't publish precentages of all their users. Survey is not something that's used for estimating the market. Here is more relevant data:
https://steamdb.info/linux/
It's far from "nobody uses".
It's surely a lot smaller than Windows one, but it's growing. Anyway, that's besides the point. The good thing is that AMD hardware is indeed mostly competitive, unlike some who claim that Nvidia hardware is much better. But it took benchmarks on Linux to demonstrate it
The only thing that Nvidia does better in hardware now is power consumption. That's indeed better in their GPUs.
The only thing that Nvidia does better in hardware now is power consumption. That's indeed better in their GPUs.
It's growing? Surely if it were it would affect those Steam survey numbers some way but as far as I can remember it's been more or less 0.5% for a while now with only minor changes from time to time.. And I can't come up with a SINGLE recent AAA game that supports the OS.
Linux gaming, apart from niche indie games, is more or less dead. And building an expensive gaming PC to game on Linux is a colossal waste of money.
Linux gaming, apart from niche indie games, is more or less dead. And building an expensive gaming PC to game on Linux is a colossal waste of money.
Last edited:
No, one of the users. But if you need to contact the owner it should be easy, he is responsive.
Nah, I was just wondering. I've been reading it the past couple weeks, I like it a lot. Friendly community.
Sure, both in amount of users and in amount of games. Again, Steam survey numbers aren't really showing market potential. But even if you assume they are accurate, they show the growth of Linux market as well. I.e. if you use them to estimate total number of Linux users on Steam (I wouldn't use it for that), you'll see that it's growing every year. Percentages themselves aren't relevant to market potential. Total amount of users is. That's what can translate into potential sales.
Hah. I can say same thing about Windows. Linux actually lets you use your hardware how you want, unlike Windows, so I see no reason why building high end PC to use with Linux is not a good idea. Many Linux gamers I know use high end components, to be able to play demanding games.
I'm not really keeping track of what's called AAA or not these days, besides I'm only buying DRM-free games, so that's usually orthogonal to legacy publishers stuff commonly called AAA. But it doesn't mean there aren't any recent demanding games for Linux.
Besides, a lot of Windows games are perfectly playable on Linux in Wine and good hardware is as useful as always to get better performance. In fact, higher end hardware is actually more important in such case, since Wine introduces some performance hit that needs compensation. It's either related to not most optimal translation of APIs, or working with DirectX bytecode (DXBC) to translate it to SPIR-V rather than working with source shaders. That loses optimization potential, that's why Wine performance is expected to be somewhat lower than on Windows or than good ports that use shader sources.
Last edited: