Valve throws it's weight behind Vulkan.

[cageymaru]

Link to article on WCCFTECH.

Quote from the article and Valve's Dan Ginsburg.

Since hosting the first Vulkan face-to-face meeting last year, we’ve been really pleased with the progress of the API and we think it’s the right way forward for powering the next generation of high performance games.

"Here’s why we think Vulkan is the future. Unless you are aggressive enough to be shipping a DX12 game this year, I would argue that there is really not much reason to ever create a DX12 back end for your game. And the reason for that is that Vulkan will cover you on Windows 10 on the same class of hardware and so much more from all these other platforms and IHVs that we’ve heard from.

Metal is single platform, single vendor, and Vulkan; we are gonna have support for not only Windows 10 but Windows 7, Windows 8, we’re gonna have it on Android and all of the IHVs are making great progress on drivers, I think we’re going to see super rapid adoption. If you’re developing a game for next generation APIs, I think it’s clear that Vulkan is the best choice and we’re very pleased with the progress and the state of the API. We think it’s gonna power the next generation of games for years to come."


Shots fired. You know he does have a very good point about covering all the OS's with one API. And it can even be ported to the PS4 it seems according to the article.


Discuss.

 

[Ruoh]

Sounds good to me. I'm all behind anything that can help Linux gain traction as a gaming platform.

 

[Tup3x]

Sounds good to me too. I'm not a fan of closed APIs that are often locked to single OS or specific hardware.

 

[Darakian]

Valve is also making a linux based game console OS. Their preference for linux compatible APIs shouldn't be surprising.

 

[Semantics]

Valve is also making a linux based game console OS. Their preference for linux compatible APIs shouldn't be surprising.

Also showing how little steam OS has done pretty sure this just is an extension of how little this means.

 

[twonunpackmule]

Valve is also making a linux based game console OS. Their preference for linux compatible APIs shouldn't be surprising.

They've always favored open ended development though. Even if they weren't developing an OS.

 

[cageymaru]

Also showing how little steam OS has done pretty sure this just is an extension of how little this means.

Well it is going to be available for Windows 10 also. So if you're a developer you'll have that platform covered also with one API. Which should save time and money in the long run. Which we know are the only thing that corporations are worried about.

 

[DPI]

Also showing how little steam OS has done pretty sure this just is an extension of how little this means.

Pretty sure SteamOS hasn't launched to retail yet.

November we'll see the Vulkan based games Valve has been working on, along with a bunch of other launch titles.

Vulkan = modern API that benefits everyone on all platforms
DX12 = modern API that benefits Microsoft on one version of one platform

Not a difficult concept to appreciate why Vulkan might be beneficial and attractive to developers longterm, especially to reduce porting barriers as mobile processors only get more powerful.

 

[tybert7]

Talk is cheap Valve. If you want to help increase the adoption of vulkan, make it worth developers wilds. Instead of taking a 30% cut of game sales, take a 15% cut if they offer a vulkan api that performs on par or better than the equivalent dx12 version (if it exists at all). That will give a nice financial incentive to have game developers put in the extra work to craft not just a dx12 version, but a vulkan version as well.

 

[DejaWiz]

Talk is cheap Valve. If you want to help increase the adoption of vulkan, make it worth developers wilds. Instead of taking a 30% cut of game sales, take a 15% cut if they offer a vulkan api that performs on par or better than the equivalent dx12 version (if it exists at all). That will give a nice financial incentive to have game developers put in the extra work to craft not just a dx12 version, but a vulkan version as well.

If a game dev utilized Vulkan to make their products usable on multiple platforms, that's a huge incentive because of the increased audiences they are catering to. More platforms usually means more sales because of that.

I infer that Nintendo would be rolling in a lot more moolah if they had developed their cornerstone games for cross-platform availability...if the Mario games and Amiibo system were available for my kid's XB1, I'd have bought them (as one example).

 

[dragonstongue]

if they can make the functionality, user freindly that windows usually is without all the crap that 10 has brought forth from disabling older games, out of date games/apps, forced updates that can and have made a boot loop scenario, a better default no wasteful UI, not have the grab a whole wack of things to enable functionality or getting rid of things just to have them still there in the background even when implicitly told GET OUTTA HERE, then yes I think many will be happy to move from MSFT.

Keeping in mind that steam essentially was made up of/from the same group that made windows the gaming platform it has become over the years. While DX12 is great it is up the the vendors to ensure they have the proper hardware/software switches to get the most out of it, so in this regard it is free and open to all though is only via win10, Vulkan might be a tad more open, but does it not also rely on certain software/hardware limitations having being Linux/HSA founded in the first place?

We all want valid alternatives, if MSFT would have not been tools with the last couple of releases I do not believe Valve or anyone would have been as focused on seeing them done away with. Vulkan would be great if it was the OS that made the systems i.e consoles ability to do true cross-platform play regardless of any other actual hardware difference and licencing rights, however I believe it will be a while before we see that truly.

 

[5150Joker]

If Vulkan takes off that would be nice but then we've had APIs like OpenGL forever and they never went anywhere so I don't hold much hope for Vulkan.

 

[Pieter3dnow]

The idea is the same as with OpenGL nothing is new. Can you list the OpenGL based games (on windows) past year or two?

The problem lies not so much with what Vulkan does but how Windows is still dictating the developers path . As soon as we get game engines using Vulkan then were another step closer to a place where developers can finally shed windows as a sole target for their product.

.

 

[Pieter3dnow]

If Vulkan takes off that would be nice but then we've had APIs like OpenGL forever and they never went anywhere so I don't hold much hope for Vulkan.

That is because you do not understand what Vulkan is for you it is just a name change, for others it is an API which allows "close to metal" performance. Developers can get away with developing just for Vulkan and support "all" operating systems......

 

[DPI]

If Vulkan takes off that would be nice but then we've had APIs like OpenGL forever and they never went anywhere so I don't hold much hope for Vulkan.

When Vulkan succeeds it will be for all the reasons OpenGL didnt: the latter was a terrifying jigsaw puzzle of two decades of bloated and appended code that was too much trouble for most devs and IHV's to deal with. Too. much. baggage.

Vulkan dumps all that OpenGL gunk entirely and starts clean slate. So it abandons OGL backward compat but that's a good thing longterm.

 

[Michaelius]

They need to lead by example and show some solid AAA Vulcan games themselves. Microsoft is already doing it for DX12 and windows store.

 

[Semantics]

Pretty sure SteamOS hasn't launched to retail yet.

November we'll see the Vulkan based games Valve has been working on, along with a bunch of other launch titles.

Vulkan = modern API that benefits everyone on all platforms
DX12 = modern API that benefits Microsoft on one version of one platform

Not a difficult concept to appreciate why Vulkan might be beneficial and attractive to developers longterm, especially to reduce porting barriers as mobile processors only get more powerful.

I don't care if they call it "Beta" it's been public for nearly 2 years now and has done nothing...

if they can make the functionality, user freindly that windows usually is without all the crap that 10 has brought forth from disabling older games, out of date games/apps, forced updates that can and have made a boot loop scenario, a better default no wasteful UI, not have the grab a whole wack of things to enable functionality or getting rid of things just to have them still there in the background even when implicitly told GET OUTTA HERE, then yes I think many will be happy to move from MSFT.

Keeping in mind that steam essentially was made up of/from the same group that made windows the gaming platform it has become over the years. While DX12 is great it is up the the vendors to ensure they have the proper hardware/software switches to get the most out of it, so in this regard it is free and open to all though is only via win10, Vulkan might be a tad more open, but does it not also rely on certain software/hardware limitations having being Linux/HSA founded in the first place?

We all want valid alternatives, if MSFT would have not been tools with the last couple of releases I do not believe Valve or anyone would have been as focused on seeing them done away with. Vulkan would be great if it was the OS that made the systems i.e consoles ability to do true cross-platform play regardless of any other actual hardware difference and licencing rights, however I believe it will be a while before we see that truly.

Unified systems will never occur on consoles, Console makers would refuse the lack of control.
-----------------------------------------
DirectX is not only a collection of APIs but a collection of APIs designed to interact with each other and toolsets to help developers reach their goals.
Vulkan is just the equivalent of D3D12 again repeating the failures of OpenGL, it's far from a suite for designing games, general software is a different deal but games? Fat chance.

 

[M76]

To me valve has absolutely no weight as a game developer, as they don't develop games, they refuse to even do HL2 EP3 so they can shove their opinion where the sun don't shine as far as I'm concerned.

 

[twonunpackmule]

To me valve has absolutely no weight as a game developer, as they don't develop games, they refuse to even do HL2 EP3 so they can shove their opinion where the sun don't shine as far as I'm concerned.

Right, because they don't still support DOTA2 or Team Fortress. Source 2.0 is not an updated engine. Gotcha.

 

[n=1]

When Vulkan succeeds it will be for all the reasons OpenGL didnt: the latter was a terrifying jigsaw puzzle of two decades of bloated and appended code that was too much trouble for most devs and IHV's to deal with. Too. much. baggage.

Vulkan dumps all that OpenGL gunk entirely and starts clean slate. So it abandons OGL backward compat but that's a good thing longterm.

On the history (and decline) of OpenGL, this guy wrote a gigantic wall of text, but I thought it was an interesting read.

 

[M76]

Right, because they don't still support DOTA2 or Team Fortress. Source 2.0 is not an updated engine. Gotcha.

Ah, multiplayer doesn't count for me, gotcha! It's the easy way to make a quick buck. Throw out a few multi games, without any substance. As in real campaign, real story, real narrative, real characters, and real missions, and real ai.

 

[5150Joker]

When Vulkan succeeds it will be for all the reasons OpenGL didnt: the latter was a terrifying jigsaw puzzle of two decades of bloated and appended code that was too much trouble for most devs and IHV's to deal with. Too. much. baggage.

Vulkan dumps all that OpenGL gunk entirely and starts clean slate. So it abandons OGL backward compat but that's a good thing longterm.

The cynic in me thinks MS has too much weight for big publishers to really get behind Vulkan. Valve isn't enough despite how popular DOTA and CS:Go are. I guess we'll see in the coming year how many titles have Vulkan support vs DX 11/12.

 

[david_]

Sweet, for all those games Valve is making now.

 

[m2h]

The cynic in me thinks MS has too much weight for big publishers to really get behind Vulkan. Valve isn't enough despite how popular DOTA and CS:Go are. I guess we'll see in the coming year how many titles have Vulkan support vs DX 11/12.

I would say MS has money and support to provide for DX12 to developers. Two very successful platforms Windows and Xbox One.

The good thing about Vulkan is that its platform open.

So it really comes down to what target audience the developer / publisher is aiming for, and what long term goals they have. A lot of developers would be safe only developing for MS as we already know.

 

[5150Joker]

I would say MS has money and support to provide for DX12 to developers. Two very successful platforms Windows and Xbox One.

The good thing about Vulkan is that its platform open.

So it really comes down to what target audience the developer / publisher is aiming for, and what long term goals they have. A lot of developers would be safe only developing for MS as we already know.

I think support is the key word you noted and with MS actively trying to unify their platforms (Xbox, PC, tablet) I can see them pushing DX 12 pretty hard. That's why I'm doubtful about how successful Vulkan will be unless major publishers + NVIDIA/AMD get fully behind it.

 

[DPI]

The cynic in me thinks MS has too much weight for big publishers to really get behind Vulkan. Valve isn't enough despite how popular DOTA and CS:Go are. I guess we'll see in the coming year how many titles have Vulkan support vs DX 11/12.

And its not just Valve behind Vulkan. Example the Frostbite lead at DICE/EA has stated Vulkan/DX12 support in the Holiday 2016 Battlefield title (though he doesn't outright say Battlefield obviously but not hard to read between lines). Nvidia also showed off their beta Vulkan driver at a conference a few months ago, so they've got a team dedicated to it.

Really the best thing MS and it's water carriers can do for Vulkan is to keep laughing it off as a nonfactor and ignoring it. That always works out well for MS.

 

[Cobalt35]

I just wish they'd stop with the hype talk and show us the products. Almost 100% of the time that a company hypes things up like they're doing with this, the end product fails to even come close to the hype. I'm 100% all for it if it works, I'm just not holding my breath for it to do so based on hype.

 

[DPI]

I don't care if they call it "Beta" it's been public for nearly 2 years now and has done nothing...

Not a lot it could "do" before the defacto API was finalized, and the partners developing for it aren't press releasing the fact every day. Least not the way MS partners do, because there's big marketing dollars in a hypermarketed OS called Windows 10 that the platform owner is trying to lock everyone into.

Also remember that Xbone and PS4 titles didn't exist and those platforms had zero marketshare until the platforms actually launched. Devs therefore needed to have access earlier, before those platforms had done anything.

I definitely encourage naysayers to write Vulkan and by extension SteamOS off though, so that this will be comedy to revisit later on.

 

[Ashbringer]

The big thing about Vulkan is SPIR-V. SPIR-V can take any shader code and make it work on Vulkan.

 

[razor1]

well initially its only GLSL, later on yeah probably.

 

[EODetroit]

Its really about speed. The main future game I'm worried about right now is Star Citizen. If I can run Star Citizen using my GTX980 on Windows 7 with the same or better performance because of Vulkan than I could using DX12 on Windows 10, then I'll stay on Windows 7. If Star Citizen performs better on DX12, then I'll update to Windows 10. I don't give a shit about marketing, I care about benchmarks for the games I want to play.

 

[Ashbringer]

Its really about speed. The main future game I'm worried about right now is Star Citizen. If I can run Star Citizen using my GTX980 on Windows 7 with the same or better performance because of Vulkan than I could using DX12 on Windows 10, then I'll stay on Windows 7. If Star Citizen performs better on DX12, then I'll update to Windows 10. I don't give a shit about marketing, I care about benchmarks for the games I want to play.

You should be more concerned about Nvidia's lack of Async Compute hardware. Without it the gains on Vulkan and DX12 is very little compared to what AMD hardware receives.

 

[socK]

i feel like vulkan is dragging its feet pretty hard given that i made a working DX12 application 4 months ago.

guess it doesn't matter too much, given how straightforward it should be to port - but as a mere mortal i'd like to play with it sometime this century.

 

[crusty_juggler]

A virtual hat vendor and niche hardware creator has thrown it's "weight" behind Vulkan?

Meh.

 

[Anarchist4000]

You should be more concerned about Nvidia's lack of Async Compute hardware. Without it the gains on Vulkan and DX12 is very little compared to what AMD hardware receives.

Nvidia doesn't necessarily need async compute though. The goal of Async was to get more efficient scheduling of the hardware. If they were highly efficient synchronously, as evidenced by their DX11 performance, then Async wouldn't add that much. AMD on the other hand appears to have had a lot of untapped potential in their GPUs.

i feel like vulkan is dragging its feet pretty hard given that i made a working DX12 application 4 months ago.

Maybe, but they might also be tying it into the Steam Box launch November 10th. If Valve launches with the current OpenGL support and without a next gen API on linux it will be rather underwhelming. Intel and AMD have been putting a lot of work into open linux drivers lately, but they have a ways to go to be finished before November 10th. Vulkan, being rather light on driver work, they likely have working already.

With Nintendo(upcoming console/handhelds) joining Khronos and Android adopting Vulkan for graphics I'd say the API is getting good attention so far. Not to mention that Vulkan was designed by a handful of AAA developers.

 

[Tup3x]

You should be more concerned about Nvidia's lack of Async Compute hardware. Without it the gains on Vulkan and DX12 is very little compared to what AMD hardware receives.

Bullshit. Especially in Star Citizen the biggest bottleneck are draw calls.

 

[Filiprino]

I've not tested SteamOS, but I play computer games on GNU/Linux, as that is my main OS.

The only problem I see is that current ports are more like MacOS X ports instead of Windows games: reduced graphical options or subpart performance due to the use of underperformant translation layers to do a quick job of having as many games as possible in catalog.
Users have complained about some games. I hope Alien Isolation (available next week for SteamOS / GNU/Linux) is a better port. I want to buy it.

As for Vulkan, really, it is THE API. Multiplatform and implementing the model Mantle did introduce. Small driver, control on user space. Nice shader front end (SPIR-V) to program shaders against in the language you want to if it has an available intermediate representation (IR) compiler.

Recently Nintendo did join the Khronos Group. That must mean something. At least that they want early access to standards developed by Khronos.

Really, GNU/Linux is a valid platform today. I've been using it for everyday usage since 10 years ago. And now you can play games on it! For any particular application you can always use a virtual machine. Current processors have incorporated many hardware virtualization features to have VMs transparently running, like a native machine.

Maybe Valve could attract more pseudo power users by creating an XML-based configuration system, equivalent to the control panel. But that requires also an effort from the community which has already adopted things like Systemd, a clusterf**k of binary, monolithic processes.

 

[Stoly]

I can't believe it took 2 pages for Half-Life 3 CONFIRMED!!!

 

[Cobalt35]

Also, the only thing Valve are good for any more in Steam. As has been said, they can butt the fuck out of everything else.

 

[tybert7]

Actually, far more important than valve supporting Vulkan in my mind is google supporting it in android, because that will be the go to standard for any and everyone trying to make higher performing games on the android platform, which still dwarfs iphones in terms of total sales if not total profits.