Vulkan/OpenGL Next Generation Initiative: unified API for mobile and non-mobile devices.

[Deleted member 13524]

From Anandtech:

For example OpenGL NG will be a single API for desktop and mobile devices alike – there will not be an ES version of OpenGL NG, the desktop and mobile will be unified. As mobile GPUs have nearly caught up in functionality with desktop GPUs and OpenGL NG is a clean break, there is no need to offer separate APIs on the basis of legacy support or hardware feature gaps. There will be just one modern OpenGL: OpenGL NG.

I guess it's true after all that it doesn't make much sense to keep supporting a castrated API for mobile devices.

 

[Grall]

The first season will be a little rocky, second will improve. By the third, things really take off.

 

[DavidGraham]

That if they can get to reach consensus, I still vote each major party will try to cater for it's own gains as much as possible, just like last time.

 

[Ailuros]

That if they can get to reach consensus, I still vote each major party will try to cater for it's own gains as much as possible, just like last time.

The way I see it mobile GPU (hw) developers are damn close to desktop APIs if not for some on the exact same level; how far can the road to DX12 theoretically really be in the next 3-4 years from now and how high are the chances that we will see any major D3D update in the meantime?

 

[Deleted member 13524]

The way I see it mobile GPU (hw) developers are damn close to desktop APIs if not for some on the exact same level; how far can the road to DX12 theoretically really be in the next 3-4 years from now and how high are the chances that we will see any major D3D update in the meantime?

I don't really think DX12 will be in the way of OpenGL. They both have their own places. DX12 will rule the PC and xbone. OpenGL is very well rooted in mobile devices (where DX12 is practically non-existant).

What is probably more worrying to Khronos are the deviations being taken from google (android extension pack) and apple (metal).
Clearly, they exist because OpenGL ES hasn't evolved the way they wanted.

 

[sebbbi]

This is good news. Better driver certification process and API refactoring were both very much needed. Also it will be easier for developers to support Mac and Linux in the future with the same rendering code as the mobile devices. This means (slightly) more games for these platforms.

NVIDIA was bringing desktop OpenGL 4.4 to mobiles (with Tegra K1). I wonder what does this announcement mean for these plans.

 

[Helmore]

NVIDIA was bringing desktop OpenGL 4.4 to mobiles (with Tegra K1). I wonder what does this announcement mean for these plans.

Nothing much probably. Next gen OpenGL is still years away from being finalized.

 

[Ailuros]

I don't really think DX12 will be in the way of OpenGL. They both have their own places. DX12 will rule the PC and xbone. OpenGL is very well rooted in mobile devices (where DX12 is practically non-existant).

What is probably more worrying to Khronos are the deviations being taken from google (android extension pack) and apple (metal).
Clearly, they exist because OpenGL ES hasn't evolved the way they wanted.

I didn't mean or imply that Dx12 is in its way. My point just was that the gap between ULP mobile and desktop is small and will equal zero sooner than later; definitely not worth keeping OGL & OGL_ES.
Other than that every DX needs its Mantle, as much as every ES needs its Metal and yes its just an euphemism.

One DX is enough for future windows & windows phone and one OGL for ULP mobile & desktop irrelevant of market penetration for each case.

 

[xpea]

if OpenGL NG doesn't become another Long Peak scenario, its a good initiative.
This OGL ES non sens must die. Google AEP and Apple mantle are big signs of ES failure.
Now let's see the evolution and most important the timing. hopefully, Khronos internal politics won't cause too much trouble...

 

[Helmore]

Anyone willing to guess at a timeframe?

 

[willardjuice]

Par for the course with this group. I lol'd at the notion that Long Peaks failed because Khronos wasn't being inclusive enough. While I'm not against a democracy per say, Khronos's problem is not a lack of participants. In addition, the fact they're not even sure they can achieve a clean break is a sad joke. I don't even know how that's still a discussion. If that can't even agree on that, what will happen with a "real" api issue/disagreement?

I suspect the status quo will continue. They'll come up with a plan and by the time every party vetoes parts that don't run/fit well with their platform it'll be shit (i.e. late and still behind other apis). Nothing from that article remotely comes close to convincing me otherwise.

Can't wait for 2016 when they roll out DX12 lite.

 

[rpg.314]

Par for the course with this group. I lol'd at the notion that Long Peaks failed because Khronos wasn't being inclusive enough. While I'm not against a democracy per say, Khronos's problem is not a lack of participants. In addition, the fact they're not even sure they can achieve a clean break is a sad joke. I don't even know how that's still a discussion. If that can't even agree on that, what will happen with a "real" api issue/disagreement?

I suspect the status quo will continue. They'll come up with a plan and by the time every party vetoes parts that don't run/fit well with their platform it'll be shit (i.e. late and still behind other apis). Nothing from that article remotely comes close to convincing me otherwise.

Can't wait for 2016 when they roll out DX12 lite.

The press release seems pretty clear that it will be clean break.

Not sure why you think otherwise.

 

[willardjuice]

Going by the article:

At this point the single hardest sell for Khronos and the members backing the initiative will be the clean break. This is a large part of what doomed Longs Peak, and Khronos admits that even now this isn’t going to be easy; even a year ago they may not have been able to get consensus. However as Mantle, Metal, and Direct3D 12 have made their own cases for new APIs and/or clean breaks, Khronos tells us that they believe the time is finally right for a clean break for OpenGL. They believe there will be consensus on the clean break, that there must be genuine consensus on the clean break, and have been passionately making their case to the consortium members.

Perhaps that info is incorrect. And to be clear, I think it's likely that it'll happen, but it sounds like that is not set in stone yet. It's just a bit concerning.

 

[rpg.314]

Going by the article:



Perhaps that info is incorrect. And to be clear, I think it's likely that it'll happen, but it sounds like that is not set in stone yet. It's just a bit concerning.

I think Khronos's own statements carry more weight than what the press reports. In my experience the press generally just regurgitates the slide decks, losing meaning and precision in the process.

Also note that the members in the group are all hw or platform or game companies, and no companies with legacy software.

 

[Ryan Smith]

Going by the article:



Perhaps that info is incorrect. And to be clear, I think it's likely that it'll happen, but it sounds like that is not set in stone yet. It's just a bit concerning.

The idea is not that Khronos needs to sell the members on a clean break if OpenGL NG is to become a reality. Without a clean break OpenGL NG cannot come to fruition. In other words, it isn't about whether OpenGL NG will be a clean break (it will), it's about whether the members will agree to a clean break so that OpenGL NG development can move forward.

 

[willardjuice]

The idea is not that Khronos needs to sell the members on a clean break if OpenGL NG is to become a reality. Without a clean break OpenGL NG cannot come to fruition. In other words, it isn't about whether OpenGL NG will be a clean break (it will), it's about whether the members will agree to a clean break so that OpenGL NG development can move forward.

Right, which is exactly the problem! OpenGL NG (and a clean break) needs to happen if OpenGL wishes to remain remotely competitive. The fact that this isn't settled yet at this stage is concerning. The "clean break" should be a given, not something up for discussion. There are harder questions to answer...

 

[willardjuice]

Also note that the members in the group are all hw or platform or game companies, and no companies with legacy software.

I think you'd be surprised who pushes for legacy software.

 

[Rodéric]

So will it be OpenGL 2.0 "Lean & Mean" or will it be OpenGL 3.0 "Long Peaks" this time ?
Oh... that wouldn't make any difference !


They failed too many times for me to care.

 

[Rodéric]

I think you'd be surprised who pushes for legacy software.

nVidia

 

[Dave Baumann]

http://techreport.com/news/26922/amd-hopes-to-put-a-little-mantle-in-opengl-next