OpenGL 4.5 Released With New Features

[Kraut]

From the sounds of it at the last time I had a briefing from them (June), didn't sound like their unified plan was still going through.... Or at least didn't make any worthwhile progress where they wanted/could share an update with me during a conf call.

Wasn't that about a kernel interface to load a blob? Not going to happen with the kernel folks!

[zanny]

Mesa will likely support many of these sooner rather than later. A lot of them are pretty easy to implement.

You might also notice that Brian Paul (the Mesa project founder) is listed as the author on many of the extensions

Thank you for your hard work over there!

[_ONH_]

Wasn't that about a kernel interface to load a blob? Not going to happen with the kernel folks!

Well that was about adapting the kernel interface off the oss and modify it to make it work with both driver hat might work with the linux kernel, since they could tell the interface is for the oss driver and hook the closed into it. An interface just for the bloob woldn't have a chance.

[efikkan]

This is yet another minor update, which is to be expected since there is no new generation hardware available to expose a next generation API.

Direct state access is useful, but it's nothing new.

What we really need is a low level shading language allowing us to control the pipeline without a bunch of API calls to the driver. Nvidia GPUs are already capable to control threads from the GPU. This would be a compact language to implement, and any legacy features can be implemented on top of that.

[carewolf]

Why AMD, why not anybody other that Nvidia release driver on specs release day . I expect this from nvidia they do it like this traditionaly, and i did not expect that from ATi/AMD they never do it like this, only after some time .

Traditionally? Did you miss the part about NVidia taking two years to implement 4.4. They probably had all the new required extensions already, the new OpenGL standards are usually formed out of extensions almost everybody already support.

[blackout23]

Traditionally? Did you miss the part about NVidia taking two years to implement 4.4.

It took them 1 day to ship 4.4... Tiny difference.

http://www.phoronix.com/scan.php?page=news_item&px=MTQxODI

[Marc Driftmeyer]

Michael,

You list extensions for the highlights when these are the guts of change:

• Direct State Access (DSA) ? object accessors enable state to be queried and modified without binding objects to contexts, for increased application and middleware efficiency and flexibility;
• Flush Control - applications can control flushing of pending commands before context switching ? enabling high-performance multithreaded applications;
• Robustness - providing a secure platform for applications such as WebGL browsers, including preventing a GPU reset affecting any other running applications;
• OpenGL ES 3.1 API and shader compatibility ? to enable the easy development and execution of the latest OpenGL ES applications on desktop systems;
• DX11 emulation features ? for easier porting of applications between OpenGL and Direct3D.

AMD was championing the first two and forced the hand by releasing Mantle.

[liam]

Broadcom: 1 [ARB_clip_control]
Qualcomm: 1 [ARB_derivative_control]
Nvidia: 7 [ARB_clip_control, ARB_cull_distance, ARB_direct_state_access, ARB_get_texture_sub_image, KHR_robustness, ARB_shader_texture_image_samples, ARB_texture_barrier]
AMD: 4 [ARB_cull_distance, ARB_conditional_render_inverted, KHR_context_flush_control, ARB_direct_state_access]
Intel:
Unity: [ARB_direct_state_access]
CodeWeavers: 1 [ARB_clip_control]
VMWare: 4 [ARB_clip_control, ARB_cull_distance, ARB_conditional_render_inverted, ARB_get_texture_sub_image]
Valve: 1 [ARB_clip_control]
HI Corporation: 1 [ARB_clip_control]
Blizzard: 1 [ARB_clip_control]
Transgaming: 1 [KHR_robustness]



Big GPU vendor:
Nvidia

Big missing GPU vendor:
Intel

Big virtual GPU vendor:
VMWare

Big middleware vendor:
Unity

Big game dev:
Blizzard



And where is Google?

Why would you expect google? These companies either produce gpus, or software that requires gpus. Google just consumes apis, but don't need to push for anything in particular in order for their product to work well.

[dungeon]

Traditionally? Did you miss the part about NVidia taking two years to implement 4.4.

How that can be .

Anholt today rendered first texture on Pi, out of nothing in about month vc4 is now in working state

[zanny]

What we really need is a low level shading language allowing us to control the pipeline without a bunch of API calls to the driver. Nvidia GPUs are already capable to control threads from the GPU. This would be a compact language to implement, and any legacy features can be implemented on top of that.

You want GPU assembly. I do too. We already have it, though - Intel and AMD publish their ISA's. But Nvidia does not, and obfuscates their microcode, and makes it a pain in the ass to really know how the GPU even works. And all the ARM vendors are even worse.

It is insane to think you have computers, which modern GPUs pretty much are - self contained computers - where the instruction set is not public.

Because GPU's should really be behaving like CPUs now. Instruction sets with extensions (preferably a shared common ISA, which each vendor creating and proposing extensions... like how OpenGL is run) that the community and companies can just build compilers for. And then we could write whatever higher level languages we want on top of the GPU, and let those compete on their own.

The problem right now really is a lack of competition. If nobody can see the ISA, nobody can emit shaders besides the proprietary driver, and thus nobody can create alternative implementations.