• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

Vulkan 1.0 API specification complete, undergoing final review & polish

I didn't say for games. I'm saying Vulkan will have an great influence outside of games and it's landing around the same time as Wayland.

I agree that Vulkan might be genuine game changer but Wayland has been in the process of landing for approaching 5 years now though. It's been a very slow transition, and right now, the benefits are limited (excepting the double buffering issue raised earlier in the topic).

I would be more optimistic about Wayland if nVidia shipped a driver that supported non X11 window systems (and all that that implies: using the existing kernel mode setting infrastructure in Linux, providing an EGL implementation on the desktop and everything in between). I believe they are working on one internally, but no release so far.
 
Going forward: All Valve games is a guarantee.

I've been saying for a while Gabe has an agenda to fuck M$ and take over PC gaming by making it open source.

I expect L4D3, HL3 and all future Valve games to be SteamOS (Linux) exclusive.
 
I've been saying for a while Gabe has an agenda to fuck M$ and take over PC gaming by making it open source.

I expect L4D3, HL3 and all future Valve games to be SteamOS (Linux) exclusive.

First off, no that isn't happening, they have made that abundantly clear and that is indeed the point of Vulkan and their work on Source 2 as they have already shown.

He doesn't have that "agenda" against "M$". The issue is a very real one. MS is their main platform and where they do their work and innovation. Their success comes from the open nature of the platform and what that allows to happen for the market and for Valve, which no other platform has other than Linux. If MS close that, as their Windows 8 direction was suggesting along with the rest of the market in general, Valve would start losing the ability to do what they do and what gets them money and growth.
 
I would like to see some hard numbers on the relative performance of the Linux and Windows networking stacks. You claim Linux outperforms Windows here. Can you(or anyone else) provide evidence, hopefully with some citations?

It's one of the reasons why Linux is so widely used for networking (routers, proxies, DNS servers). Even Microsoft have created their own Linux distribution for networking in Azure.

Here's one example, you could probably find a few more by searching:

http://www.slideshare.net/Principle...rise-linux-6-vs-microsoft-windows-server-2012

I don't think it's anything controversial, just one of the key things Linux was made and optimized for.
 
been living under a rock I presume?

No. I just don't think that Valve can do it. They aren't a company like Apple. They might have quite a few guys like Wozniak. But what they need is guys like Steve Jobs. They also don't understand the market. People don't care about Linux like they do. People care about comfort. They care about the latest tech. They want all their games to run. Windows already does that. Linux is a solution to a problem which gamers simply don't have. This problem only exists in the heads of a few idealists.

Besides that, D3D has always been proprietary and has always introduced a dependency on Windows. I would think that is common sense that the games industry would want to avoid any dependency on a proprietary system, maintained by a company that has been investigated by the US DOJ for anticompetitive behaviour and has consistently demonstrated incompetence in both management and software development

The game industry does not care about this "dependancy on a proprietary system", because it does not affect them negatively. Again, this is a problem that only exists in the heads of some idealists. This is not a real problem for developers. Companies like EA, Ubisoft or Activision don't have any problems with Windows being what it is.

Gemüsepizza, you really are something else, huh!

Guess what: DX12 and Windows 10 is NOT the only thing that matters.

I did not say that it is the only thing that matters. It is the only thing that matters for AAA development in the future.

Let me tell you why.
First: Vulkan IS compatible with Win10.
So you have two choices, basically.

One: develop for Win10 and Win10 only

Two: develop for everything, targeting Vulkan - which includes Win10 as well.

Companies that develop AAA titles will offer a DX11 and DX12 rendering path for a transitional period. After that, they will switch to DX12. Their engines already support DX11. This is not a problem.

At that point, it does not matter where the AAA devs flock, because there *IS* a market for non-aa products, indie games and a tons more in between.

I never said there wasn't.

Therefore, it makes MORE sense to support Vulkan in the long term than to lock yourself to a single api.

Like I said above, the big companies do not have a problem with locking themselves to a single API for their AAA titles. They have been doing this for over a decade.

At this point, this is like arguing for focusing on Swift because iOS is what matters only in mobile gaming (right? right? eh), so who cares whether other platforms are able to run your code or not...

This is a silly comparison. I was arguing that Windows/DX12 will be the future of AAA development. Swift is neither the present or the future of all mobile gaming, because it's only on iOS, and the market is split between iOS and Android. There is no such split in the PC market. Windows dominates everything. Linux doesn't have anywhere near the relevance for PC gamers like Android has for mobile gamers.

Maintaining DX11 and DX12 means maintaining two rendering paths which are fundamentally different in their approach.

Instead, simply maintaining one Vulkan path for compatibility with all relevant versions of WIndows and alternative OSes is much easier and less costly.

Really, did Vulkan piss on your pizza?

They only have to do this for a transitional period for select titles, and all major graphic engines already support DX11. They only have to add DX12 support, so the costs won't be high. I also don't think game devs want to trade in the comfort of DirectX regarding things like documentation. And like I said, alternative OSes are irrelevant when it comes to AAA gaming.

Really, did Vulkan piss on your pizza?

/* microsoft faboy gibberish */
same rock I presume?

Grow up guys.

So the usual "nothing will change, why bother" horseshit? Which has already been proven false by Microsoft's reaction to Mantle/Vulkan. Just the threat of a viable competitor has given us a DX12 comprised of largely the same features in short order. Any work done to compete with entrenched proprietary API via open source should be encouraged, it benefits literally every developer and end user. There is plenty of room for Linux/Vulkan to grow in the entertainment/living room PC space, since Microsoft have botched those initiatives at every turn, which drove them to Xbox in the first place.

And devs will welcome DX12 with open arms.

But please, give us some good reasons for Khronos group members to just bend over and say "yeah, we should probably just keep tying our business to the whims of an anticompetitive company that is willing to lock their proprietary API to whichever operating system they're trying to push". As if that's good for them or their customers.

Again, show me how this affects companies like EA, Ubisoft, ... negatively. It doesn't. They don't care about that stuff,

Unknown Soldier: what you do not seem to realize is that you will NOT be able to distinguish a Vulkan-ran game on Win10 and a DX12-ran game on Win10.

Therefore, the fact that linux is now a possibility without extra effort is just a bonus (although rationally thinking, even if a > (b+c+d), a+b+c+d is STILL > a. That is just rationality.
Supporting Vulkan does not mean mandatory linux installations. It means linux support becomes "default" alongside Win, which is INSANELY good.

But it doesn't. Just because a game will use Vulkan, doesn't mean it will run automatically on Linux "without extra effort".

If you buy a pre-built SteamOS box, then connecting it to your TV and playing supported games on it is as easy as doing the same thing on a console. And with Vulkan, you should be able to actually enjoy competitive performance in large-scale games.

Right, "supported games". With Windows you do not have to care about "supported games". Everything just works. And you get the best performance. There is no reason for consumers to switch to SteamOS. Like I said, SteamOS is a solution to a problem which only exists in the head of some nerdy, idealistic engineers. Those guys might be willing to put up with the limitations of SteamOS. Gamers aren't.

But more importantly, as a developer you can target all versions of Windows with Vulkan, rather than just Windows 10.

Developers can already do that with their existing engines which support DX9/DX11.

Actually, yes. Go download the latest Ubuntu. A few clicks and you have a full operating system with a functional "app store" anyone can search directly to add games to their system.

Hell, you don't even need to actually install Ubuntu to try it anymore. You just liveboot right off the USB stick, and when you're ready to make the install permanent, you can just install as-is onto the hard drive.

What an absurd statement. Ubuntu is certainly *not* easier to install, configure and play games on than Windows. Windows is a vastly superior platform which is easy to use, mature and stable, offers high performance and support for every imaginable software.

Yup. Linux has come a long way. For a typical user it's actually simpler than windows because it's "closer" to smartphones with appstores (in distros like Ubuntu). Linux gets complicated if something goes wrong and you have to deal with the terminal. But the community and troubleshooting is the best out there. So even that is a lot easier than in the past.

Troubleshooting is a pain in the ass with Linux because of the incredibly fragmented market, the poor support from GPU vendors, game companies and the not user friendly design of virtually all distros. It is just a fact that Windows is superior in almost every way. This is not fanboy-ish. It's simply reality. Windows is the best option for gaming right now.

Stockholm syndrome, or capture-bonding, is a psychological phenomenon in which hostages express empathy and sympathy and have positive feelings toward their captors, sometimes to the point of defending and identifying with the captors. These feelings are generally considered irrational in light of the danger or risk endured by the victims, who essentially mistake a lack of abuse from their captors for an act of kindness.
https://en.wikipedia.org/wiki/Stockholm_syndrome

First you insult people as Microsoft fanboys, now this. I am not a Microsoft fan. I am not defending Microsoft. I am just describing the reality. What some are doing here is wishful thinking. And if you point that out to them, the become aggressive and insult you.

That is a picture of the Khronos group, which EA has been a part of, and the companies supporting Vulkan. EA's logo is right there.

Just because they are a member of the Khronos group, which many have been for years because of OpenGL, doesn't mean that all their future products will use Vulkan.

That makes no sense. Windows is compatible with older version of DX, but DX is not compatible with older versions of Windows. I can run D3D9 on Windows 10, but I can't run D3D12 on Windows Vista.

People who still use Windows Vista aren't interested in products which use DX12.

Between a Linux distro, creating hardware for the first time, a new graphics API... yeah I'm pretty sure that all counts as "serious efforts"

You need to calm down

Everyone can "create" a Linux distro. They did not "create" hardware. They did not "create" a new graphics API. Valve's problem is that they are viewing this too much from a nerd perspective. Normal users just aren't interested in Steam machines, why should they? To save a few bucks? They buy Windows machines instead. Steam machines only sound cool to nerds.

I think you are very clearly wrong when you're saying that developers will not support Linux. There is a lot of evidence out there that suggests otherwise. Right now Linux is used by so ething like 1% of Steam users, right? Valve released SteamOS, they are working on a new engine and they are pushing a new cross-platform API, but all of the above aren't anything more than baby steps. Right?

So, the fact that as of right now there are 1727 Linux games on Steam is amazing. With very little effort so far Valve has managed to convince tons of developers to port their games to Linux, surprisngly including some very big names. Dying Light, Alien Isolation, Shadows of Mordor, Borderlands series, Grid Autosport, Civilization V, Bioshock Infinite, XCom, Witcher 2, The Metro series, Dead Island. More are coming, Capcom just recently announced that Street Fighter V will be coming to Linux.

Those are old ass games. Nobody cares about those, that's not how you convince gamers. You need same day releases with the Windows version at the very least so that people would even think about considering them. But you won't get those, because devs do not have a problem with Windows. And you won't get gamers to choose the Linux version of a game, because they don't have a problem with Windows. Again, the only people who have a problem with Windows and DX12 are some idealistic nerds. But that's not enough for Linux and Vulkan to succed in the AAA gaming market.

It's one of the reasons why Linux is so widely used for networking (routers, proxies, DNS servers). Even Microsoft have created their own Linux distribution for networking in Azure.

Here's one example, you could probably find a few more by searching:

http://www.slideshare.net/Principle...rise-linux-6-vs-microsoft-windows-server-2012

I don't think it's anything controversial, just one of the key things Linux was made and optimized for.

And this is relevant for gaming how?
 
Apple seems to not want much gaming on osx. They are much more interested in gaming on ios. Apple seems to think it's a good time to make a proprietary api for the mobile market where they are strong.

It doesn't only have an impact on games though. Did you see the iPad Pro presentation where they showed Autocad, and said that workload wouldn't run at 60FPS on any PC?

Well that's because it was using Metal. Using DX12 or Vulkan for the renderer would have the same effect, Vulkan would just allow the developer to cross port the app natively across all major PC OSs far more easily rather than having another team who are set to master Metal.

In that case since they already did the iOS one it may not be so bad, but for things that are on PC OS's and not mobile, that's where Vulkan would be a big draw.
 
Gemüsepizza;189900686 said:
What an absurd statement. Ubuntu is certainly *not* easier to install, configure and play games on than Windows. Windows is a vastly superior platform which is easy to use, mature and stable, offers high performance and support for every imaginable software.

Well, you just outed yourself as not even having booted up Ubuntu.

It's seriously put in USB or CD ->

What language do you want ->

Give us an username and password you'd like ->

We're gonna do this to your hard drive, if you really want to you can customize the partitions but we got you ->

wait a few minutes

100% working OS with giant "Internet" and "Get games and apps" buttons
 
Well, you just outed yourself as not even having booted up Ubuntu.

It's seriously put in USB or CD ->

What language do you want ->

Give us an username and password you'd like ->

We're gonna do this to your hard drive, if you really want to you can customize the partitions but we got you ->

wait a few minutes

100% working OS with giant "Internet" and "Get games and apps" buttons

It's almost the same with Windows. What you are omitting is the countless problems one can experience when using Linux for gaming. And then there is the extremely limited library...

I was simply answering a question. I'm well aware that networking is irrelevant for gaming. Who'd want to play games over a network?

Networking isn't irrelevant for gaming. That's why I am asking. How does this help a gamer?
 
Gemüsepizza;189901280 said:
It's almost the same with Windows. What you are omitting is the countless problems one can experience when using Linux for gaming.

"Hey, you're going to need official, non open source drivers to play this game. Click this button to install them"

*click, download, install, just like "FIND HARDWARE" bubble on Windows*

*game works*


Hell, you need DOSBox to play DOS games on both platforms anyway, DOS isn't supported on 64 bit Windows. And Wine can take care of most non-Linux native stuff, and Ubuntu even auto-recognizes Windows binaries and has Wine ready to go to run them.

It even supports my esoteric USB wireless adapter out of the box! I still need to keep the install CD around when I reinstall Windows.
 
Gemüsepizza;189900686 said:
What an absurd statement. Ubuntu is certainly *not* easier to install, configure and play games on than Windows. Windows is a vastly superior platform which is easy to use, mature and stable, offers high performance and support for every imaginable software.

This reads like marketing spiel...
 
I've been saying for a while Gabe has an agenda to fuck M$ and take over PC gaming by making it open source.

I expect L4D3, HL3 and all future Valve games to be SteamOS (Linux) exclusive.

They don't have to take over PC gaming because they already did with a windows software called steam.

And a pair of shooters wouldn't make people make the jump to SteamOS.
 
"Hey, you're going to need official, non open source drivers to play this game. Click this button to install them"

*click, download, install, just like "FIND HARDWARE" bubble on Windows*

*game works*

Hell, you need DOSBox to play DOS games on both platforms anyway, DOS isn't supported on 64 bit Windows. And Wine can take care of most non-Linux native stuff, and Ubuntu even auto-recognizes Windows binaries and has Wine ready to go to run them.

Hey, let's install "Alien Isolation" on my new shiny Steam machine yay.
...
...

Wait, what does that mean "No support for AMD/Intel GPUs"?
 
I really hope Apple decides to hop onboard at some point. Metal is great and all, and I understand why they needed it. But Vulkan is a far more attractive option for those who wish to code for platforms outside OS X and iOS.
 
Gemüsepizza;189900686 said:
Developers can already do that with their existing engines which support DX9/DX11.
Not nearly as conveniently, no.

I don't know if you realize this, but the entire paradigm of how you interact with an API differs significantly between Vulkan/DX12 (and current console APIs) and DirectX11 (or worse 9). If you introduce a new feature which depends on (or is merely simpler to implement using) the Vulkan/DX12 paradigm, then also having to support it in your "legacy" branch for DX11/9 is a significant effort.

Conversely, with Vulkan you can have a single API and paradigm target a huge variety of devices -- crucially, all relevant versions of Windows, and not just the one MS wants to push at a given point in time.
 
It's one of the reasons why Linux is so widely used for networking (routers, proxies, DNS servers). Even Microsoft have created their own Linux distribution for networking in Azure.

Here's one example, you could probably find a few more by searching:

http://www.slideshare.net/Principle...rise-linux-6-vs-microsoft-windows-server-2012

I don't think it's anything controversial, just one of the key things Linux was made and optimized for.

Yeah I'm not arguing that Linux hasn't got a strong network stack for servers. Although the presentation you cite was created by Red Hat, so I would doubt it's reliability. They are also comparing RHEL servers to Windows Server with typical server workloads, measuring throughput, not latency. I was more talking about the typical throughput of a Windows workstation vs a Linux workstation, in other words a typical PC gaming machine.

For desktop stuff I would imagine that latency in the network stack is dominated by the NIC driver and the scheduler, but that's just a guess. I'm looking to see if there are any publicly available independent papers that give measurements on this workload.

I fear this tangent is also not massively relevant to game performance that much so I regret drawing the discussion in this direction. Suffice to say, game performance is dominated by GPU costs. That is why, at least right now, Windows performance is stronger than Linux for most games.
 
Gemüsepizza;189900686 said:
Companies that develop AAA titles will offer a DX11 and DX12 rendering path for a transitional period. After that, they will switch to DX12. Their engines already support DX11. This is not a problem.

Yes it is, because Windows is constantly in a transitional period, and considering more than half the people using Windows are using Windows 7, we may be talking about DX13 by the time developers feel comfortable not supporting DX11 anymore.

Here comes Vulkan, one set of APIs and logic.

  • Less APIs mean less lines of code
  • Less code means less time to develop, which means less cost
  • Less code means less bugs
  • Less bugs in development means less QA testing, which means less cost.
  • Less bugs in development means less bugs on release.
It is a win for consumers (find fewer defects), developers (use less time), and publishers (spend less money).

Have they talked about performance yet? SteamOS is currently a negative hit to most games.

If you're referring to that ArsTechnica article, Vulkan is the solution to that problem. IIRC they used a Pentium chip on their testing computer, for some odd reason, and that highlights the massive amount of CPU Overhead OpenGL has.

Better CPU utilization is one of the main benefit of low-level graphics APIs. Here's a demo.
 
Vulkan will pretty much be the standard Linux API thanks to Google and Valve
Developing for DX seem like a waste of resources given how limited the reach is for that API

Nvidia and Intel need to provide the driver support, AMD is already in bed
 
Gemüsepizza said:
They only have to do this for a transitional period for select titles, and all major graphic engines already support DX11.
Majority of active PC userbase (north of 200M) is still primarily served on DX9, because a significant portion of that userbase is there. Sure, this still excludes a large portion of steam users(there is only some overlap), but we're talking about a market-segment that has multiple titles with ~1B/year in revenue, among others. Most of them don't offer DX10/11 code-paths/or only as no-frills alternates - and it's unlikely they'll offer DX12 anytime soon either.
We've been in "DX9 transitional period for select titles" for roughly a decade now, and it's still there, holding back the market.

Developers can already do that with their existing engines which support DX9/DX11.
At an increased cost (to the developer) and reduced quality(to the consumer). Or cut-the cost and just offer universally reduced quality with DX9, which continues to work on all versions of Windows.
Yes that doesn't apply to "AAA" console ports, but their $ market-share is tiny (relative to console and PC overall market).

the only people who have a problem with Windows and DX12 are some idealistic nerds.
Every PC-release I work or have worked on, we have to make market-exclusion/cost/quality compromise decisions that I mention above. DX12 changes nothing here, Vulcan potentially could.
 
That a good, cross-platform, low level API is close to being available is exciting. Hoping to see wide support.

Someone mentioned there were rumors of a console supporting it as well? Can I get a source?
 
That a good, cross-platform, low level API is close to being available is exciting. Hoping to see wide support.

Someone mentioned there were rumors of a console supporting it as well? Can I get a source?
Conjecture from the recent news nintendo is now a Khronos member.
 
That a good, cross-platform, low level API is close to being available is exciting. Hoping to see wide support.

Someone mentioned there were rumors of a console supporting it as well? Can I get a source?

Nothing concrete, but
1) It is known for certain that Sony has contributed to Vulkan's development (and I don't know what else they would use it for?)
2) Since announcement, Nintendo has joined the Khronos Group, and they do have a console coming along the way.
 
That a good, cross-platform, low level API is close to being available is exciting. Hoping to see wide support.

Someone mentioned there were rumors of a console supporting it as well? Can I get a source?

Might have come from this.


http://gamingbolt.com/ps4-should-su...etely-native-for-current-gen-yet-brad-wardell

PS4 Should Support Vulkan, PS4’s API Not Completely Native For Current Gen Yet: Brad Wardell

Brad Wardell on how Vulkan can probably make easier for developers to make games on PS4.



Earlier this year, Brad Wardell stated that just like DirectX 12 on Xbox One, PlayStation 4 fans will also have something to look forward to. At that time, he did not revealed what exactly it was, but in an exclusive interview with GamingBolt’s Kurtis Simpson he revealed that he was referring to Vulkan. For those of you who are not aware about what Vulkan is, it’s a low level API derived from AMD’s Mantle and is platform agnostic. Under development at Khronos Group, the new API provides more control over the GPU and reduces CPU overhead.

According to Wardell, Sony’s current API is much low level compared to Mantle and even Vulkan but they should look into adding Vulkan support for the console as it will reduce a lot of developer overhead for cross platform development.

“What I was referencing at the time was Vulkan. We’re part of the Khronos Group and now it depends who you talk to at Sony and this gets in to a debate. Sony has a very low-level API already for the PlayStation 4. The problem I have with it is that if you want to make use for it you’re writing some very specific code just for the PlayStation 4. And in the real world people don’t do that right. I write code generally to be as cross-platform as I can.”

“Now maybe in Unity or Unreal, one of the other guys will write their engines in such a way so that they make the most use of it, but that’s going to take time. Whereas if they use something like Vulkan, it’s not as low-level as their API, but Vulkan has the advantage that it’s really easy to write for it. So you’re more likely to get developers to code to that and get more games on to Sony then you would otherwise.”

With Microsoft launching DirectX 12 for the Xbox One, is there is a chance that Sony might actually completely support Mantle going forward? Wardell disagrees.

“No, because their low-level API is still lower level than Mantle and Vulkan. So what I’m hoping is that they will support Vulkan.”

“Let’s say I write a game for the Steam Box and the PlayStation 4 supports Vulkan, the Steam Box supports Vulkan. It wouldn’t be that much more work for me to have my game work on the PlayStation 4. Whereas right now if I want to develop the game for the PlayStation 4, I have to learn their special custom API, that has shader languages that are different than what I’m used to, and I’m pretty sure that I have to send stuff in text instead of binary form.”

“I hate OpenGL (laughs). They’re old, their current one is just archaic. I don’t want to have to learn that, my brain is already full of OS2 and Linux crap, I don’t want to learn yet another short-term API. If I can just learn Vulkan then I can get to a lot of platforms, I don’t want to have to learn Sony’s special API, even if I would gain a few frames-per-second in doing so.”

He also revealed to GamingBolt that the current API for the PS4 isn’t completely native yet and that it has been updated from last gen. Regardless, we are not seeing the full potential of either consoles.

“With the PlayStation 4 and the Xbox One they’re not even remotely scratching the surface for what people can do and there’s still…I mean on the PlayStation 4 and their low-level API, they’re all still very…they’re like written for last-gen but updated for this gen. I wouldn’t say they’re completely native yet, I mean they are native but you know these words all get misused, but this gen’s graphics are still very far behind where they’re going to be.”

Stay tuned for more coverage from our interview with Brad Wardell for more stuff including new information on Ashes of Singularity, more on Vulkan and Mantle in the coming days.
 
Gemüsepizza;189900686 said:
Lots of replies

Dude... that was one really long reductio ad absurdum.

Your entire argument is on the behalf of people you you have not demonstrated to anyone in your argument that you know, game developers and publishers. You seem to think because Ubuntu doesn't have easy to install AAA titles it cannot flourish if it did have them. That developers won't accept Vulkan with open arms, even when many are currently a part of or using tools by those in the Khronos Group working on Vulcan.

Really, what did Linux and OpenGL ever do to you personally?

Why is it hard to look at the way apps and mobile devices have changed the way developers of all types of software are looking to find success in cross platform development tools?

Why can't DX12 and Vulkan exist together if they are not too dissimilar? It is not like Windows games are going away soon. In fact, there is a promise of more opportunities for more windows games on older versions to have parity with newer windows versions. Everybody wins.
 
I look forward to seeing what the Dolphin devs can do using Vulkan.

Not nearly as conveniently, no.

I don't know if you realize this, but the entire paradigm of how you interact with an API differs significantly between Vulkan/DX12 (and current console APIs) and DirectX11 (or worse 9). If you introduce a new feature which depends on (or is merely simpler to implement using) the Vulkan/DX12 paradigm, then also having to support it in your "legacy" branch for DX11/9 is a significant effort.

Conversely, with Vulkan you can have a single API and paradigm target a huge variety of devices -- crucially, all relevant versions of Windows, and not just the one MS wants to push at a given point in time.

To be fair this is more of an explanation of why DX11/9 support will eventually be dropped by AAA console game devs rather than an explanation of why they would support Vulkan. Vulkan would probably have a lot more appeal to them if the xbox one wasn't also DX12. I'm not sure what the incentive is for AAA game devs already making an xbox one version to use anything other than DX12 for the PC version going forward if DX 11/9 are off the table.
 
I look forward to seeing what the Dolphin devs can do using Vulkan.
Neobrain doesn't sound too positive about the whole DX12/Vulkan story:
neobrain said:
Well, technically all of what you wrote is true. However, the presentation of the media along with echos like yours lead to people overhyping technologies like D3D12 and Vulkan, which is a worrying effect because the low level APIs will not magically yield on-average improvements of utopic numbers like 20%. Furthermore, as a developer myself I find the repeated requests for the new APIs to be quite annoying, first because they are not actually the silver bullet people think it is, and second because the API isn't even targeted at hobbyist developers - it's designed for AAA engine authors, who have enough manpower and knowledge to actually put it to good use.

Sure, expert graphics developers could apply D3D12 to emulation in an efficient way - but expert graphics developers could also just optimize existing code bases today. The reason why it hasn't happened yet (to the extend it would be possible) is not a lack of tools, it's an issue of lack of manpower and knowledge.

Either way, I hope my initial reply didn't come across as offensive or something. I hope this explanation clears up my point a bit.
https://www.reddit.com/r/emulation/comments/3xcd6m/unofficial_and_experimental_build_of_dolphin_with/cy3yh14


https://forums.dolphin-emu.org/Thread-unofficial-dolphin-dx12-backend?pid=394177#pid394177
Super interesting thread about an unofficial DX12-Backend.
Reading right now.
 

That's just because he's grumpy and dislikes people popping up asking the same thing over and over. Some of the devs have said there is some interesting stuff they can potentially use it for. It's just not going to be a magic bullet that lets people using AMD cpus suddenly run games like xenoblade or rogue squadron 2/3 at a reasonable speed. Also the biggest benefit to using Vulkan is probably for those using the emulator on anything but windows.
 
Does anyone expect Google to be able to reign in the fragmentation for this somehow? For instance if we assume it will come with N then for anyone getting phones in the US before the fall like the next Galaxy it probably won't be until February-March 2017 before that gets rolled out to those devices on official channels.
 
I look forward to seeing what the Dolphin devs can do using Vulkan.



To be fair this is more of an explanation of why DX11/9 support will eventually be dropped by AAA console game devs rather than an explanation of why they would support Vulkan. Vulkan would probably have a lot more appeal to them if the xbox one wasn't also DX12. I'm not sure what the incentive is for AAA game devs already making an xbox one version to use anything other than DX12 for the PC version going forward if DX 11/9 are off the table.

for starters they'd miss out on customers who haven't upgraded to windows 10. that's a pretty huge deal.

if vulkan gets proper driver support(which it looks like it might), then there's few reasons not to use it.
 
for starters they'd miss out on customers who haven't upgraded to windows 10. that's a pretty huge deal.

if vulkan gets proper driver support(which it looks like it might), then there's few reasons not to use it.

Microsoft is giving a free upgrade to everyone that has 7/8. I realize there are people that won't make the move, but I still think most AAA console publishers are going to go DX12 with their big titles and not look back. I think a lot of it depends on how much AMD cares about Vulkan. If AMD isn't going to step up and push Vulkan by co-marketing some high profile games, then I'm not sure who will.
 
Microsoft is giving a free upgrade to everyone that has 7/8. I realize there are people that won't make the move, but I still think most AAA console publishers are going to go DX12 with their big titles and not look back. I think a lot of it depends on how much AMD cares about Vulkan. If AMD isn't going to step up and push Vulkan by co-marketing some high profile games, then I'm not sure who will.

AMD is perfectly happy to market dx12 too. A lot of the dx12 literature for the layman seems to come from AMD.
 
Microsoft is giving a free upgrade to everyone that has 7/8. I realize there are people that won't make the move, but I still think most AAA console publishers are going to go DX12 with their big titles and not look back.

Not with Windows 7 still having >50% marketshare among all desktop Operating Systems they wont.
 
Ubuntu is fairly easy till you hit some wifi adapter or printer or another accessory that doesn't have obvious driver beside a tarball. Then it's terminal time. The issue is of course with hardware manufacturers and not Ubuntu but let's not pretend there is same amount of support vs Windows.

That said Vulcan is a good thing, will be interesting to see what level of support game developers will give it (and DX12) since it does require more from a Dev vs say DX11.

It's funny you mention that, because I tried my live boot USB stick of Ubuntu and realized it didn't have any drivers for my Broadcom WiFi adapter.

A quick Google gave me this:
http://askubuntu.com/questions/55868/installing-broadcom-wireless-drivers

I'm not really sure how far I want to go with this just to muck around with a USB-based install of Linux.
 
No doubt Vulkan will be a great success on Linux, but what about Windows ? Does it stand a chance against DX12 ?

Depends on the engine and how the team wants to best utilize the tech available. Just because its on Windows doesn't mean the game has to be DX12 only. ;)
 
I still remember having games were I could pick the API and I never knew which option was the best.

Will this be the same in the future? Do I have to read DF face offs before I play PC games that support both APIs?
 
I still remember having games were I could pick the API and I never knew which option was the best.

Will this be the same in the future? Do I have to read DF face offs before I play PC games that support both APIs?

I want to go back to the days of original Unreal Tournament when I could pick between 3Dfx GLide, OpenGL, Direct3D, and software rendering.
 
It even supports my esoteric USB wireless adapter out of the box! I still need to keep the install CD around when I reinstall Windows.

You do have to admit that's a much more recent thing though. Linux support for wifi cards was a god damn nightmare for a very long time. It's still not anywhere near great. I'm hopeful for Vulkan's success, the market is made better through competition. I'm still very far from thinking Linux is going to be taking over any significant market share. I've been hearing that every year since RPM first appeared.
 
Is there any new mechanism that prevents nvidia, amd, google etc from introducing proprietary extensions? I.e. gameworks at api level.
From what I understand old OpenGL became inconsistent quasi hardware specific because of this (and possibly bad management of the core api that required those extensions to compete) and that's what killed it.

"Android UE4 Vulkan" sounded disconcerting.
 
Top Bottom