Zink Is Ending 2021 In Fantastic Shape For OpenGL Over Vulkan

[b8e5n]

That is why i am looking forward for 4k testings, as overhead should be less limiting a d revealing more about raw graphics power.

[aufkrawall]

There is also such a thing as GPU render time overhead.

[rmfx]

Great stuff !! We get closer and closer to Zink being able to replace the army of gl native drivers.
Can’t wait to see the upcoming Copper Zink version benchmarks against RadeonSI now.

[STiAT]

Seeing all this work I wished there was a good vulkan driver for nvidia cards.

The OGL one is in a pretty bad shape still, Vulkan does not exist at all to my knowledge.

But this way we would not need a OGL and a Vulkan driver but just a good Vulkan driver.

I have my reasons to have a NVidia card (easier to passively cool only with heatpipes), but I'd like to get rid of the proprietary driver one day.

Or have AMD actually produce cards not getting that hot :-).

[Quackdoc]

Holy Moly, I knew it was good, but I didn't think it was THIS good. looks like zink will be in great shape for VMs when venus gets upstreamed. infact judging by that preformance, Im almost tempted to override my default opengl to zink for a little while to see how it fairs.

There are so many cool operating systems like dahlia OS that i want to try using for a "Main" OS, but cannot due to the risk of it impeding my workflow, so I was waiting tor venus + zink to come along so I can use a VM as a daily driver.

we are way closer than i thought we were.

Now we just need venus and zink for win10+11.

[You-]

I have my reasons to have a NVidia card (easier to passively cool only with heatpipes), but I'd like to get rid of the proprietary driver one day.

Or have AMD actually produce cards not getting that hot :-).

It is hard to get new cards these days but I think you are out of the loop here.

AFAIK the RTX30 series are space heaters and dissipate much more power than the Radeon alternatives.

[darkbasic]

Zink has 87% of radeonsi performance. Keep optimizing until it gets the other way around

[lumks]

Zink working on Windows will certainly make AMD Windows users happy

Why would it? Do you forget that this exists:

https://www.phoronix.com/scan.php?page=news_item&px=Microsoft-Collabora-DirectX

- https://devblogs.microsoft.com/direc...rs-to-directx/
- https://www.collabora.com/news-and-blog/news-and-events/introducing-opencl-and-opengl-on-directx.html

[oiaohm]

Why would it? Do you forget that this exists:

https://www.phoronix.com/scan.php?page=news_item&px=Microsoft-Collabora-DirectX

- https://devblogs.microsoft.com/direc...rs-to-directx/
- https://www.collabora.com/news-and-blog/news-and-events/introducing-opencl-and-opengl-on-directx.html

Microsoft development on opengl support on dx12 is major stalled.

https://mesamatrix.net/

Zink is miles ahead of Microsoft funded dx12 attempt to have opengl on top of it.

Really by the mesamatrix progress on dx12 opengl backend it does not look like microsoft is putting the serous funding it. Please note we are not talking massive serous funding here. Zink is one valve paid full time developer at core. Either Collabora is pocketing a lot of Microsoft cash or Microsoft is simply not paying to have a decent open-gl implementation on top of DX12.

The reality is Zink being fund to support windows and provide a decent opengl implementation at this stage looks more likely than the dx12 mesa opengl back-end being highly useful. Of course Microsoft does have the money to throw at this problem if they want to.

[user1]

Why would it? Do you forget that this exists:

https://www.phoronix.com/scan.php?page=news_item&px=Microsoft-Collabora-DirectX

- https://devblogs.microsoft.com/direc...rs-to-directx/
- https://www.collabora.com/news-and-blog/news-and-events/introducing-opencl-and-opengl-on-directx.html

Isn't that exclusively for Windows subsystem for Linux, or did I miss something?