links are a bit broken but can be clicked at least.
anyway yeah.. it shows that CPU doesnt matter in this bench.

CPU, RAM or HDD/SSD I must add.
My version of GTX 1660Ti seems to be sh*ttier… Maybe because it is (I have the cheapest one from ASUS with only a single fan) and it goes to throttling pretty fast. In fact, both times I measured my second run because I wanted the card to pre-heat first. :)

my 6900 xt

Impressive! :D

that's sick

!link

edit:
my 6900 xt

damn !

Really need to update this I guess .

Here's my 6800XT

!http://downloads.unrealnetworks.co.uk/benchmarks/supersition/SP-6800XT.png

I've built a second workstation using Ryzen 9 5950X and ASUS TUF Gaming 3080Ti OC.

With no overclocking, it produces this when running Ubuntu MATE 21.04


I had a tinker and applied a modest system-wide overclock.

Ryzen 9 [email protected] base all core and DDR4@3600Mhz CL16 with tuned sub memory timings

• +120 Mhz GPU Clock Offset (1935 Mhz - 1950 Mhz)
• +1000 Mhz Memory Transfer Rate Offset (10500 Mhz or 21000 Mhz @ double data rate)
• +25W Power Limit (375W target)
• 100% Fan duty (all GPU fans)


NVIDIA Settings tweaks

• Force Composition Pipeline - Disabled
• OpenGL Image Settings - High Performance
• PowerMizer Preferred Mode - Prefer Maximum Performance

Last edited by Wimpress on 1 Aug 2021 at 11:33 am UTC

So what am I supposed to think of those results:

!2K Extreme

!4K optimized

Is running the benchmark in 2K bad because it's not the native resolution?
Or is 2K-Extreme more demanding than 4K?
And if it's about GPU performance, shouldn't those results be close to each other?

Oh, just seen, that I still have v1.0 while you guys are using v1.1 of the benchmark......... Why is the download so slow? And when will they add vulkan support?

Or is 2K-Extreme more demanding than 4K?

The 2K extreme profile is pretty much one of the most GPU heavy things available. You might be able to set up a custom 4k on extreme for nice single digit fps.

Or is 2K-Extreme more demanding than 4K?

The 2K extreme profile is pretty much one of the most GPU heavy things available. You might be able to set up a custom 4k on extreme for nice single digit fps.

Well, not really. If these would be the settings:

!Settings

Then it's pretty much a solid 11!

!Results

Updated Google sheet too. Hopefully i got everything right.

Updated the the OP top 50 pic too!

I am now listed twice on positions 36 and 37 in the pic.

hello all here is a comparison of my 6900 xt on linux vs windows i did:
!link

very competitive. if my math is correct, a 7.13% difference in score. i wonder how the results would be if it was a clean vulkan vs vulkan.

9.33% difference with my math. To get percentages, you divide the smaller number by the larger and then move the decimal.

It's nice to see the comparison side by side like that. Thanks for sharing.

The smaller number is around 93% of the bigger one, so the difference is around 7%.

It’s also different depending on whether you’re counting the increase relative to the smaller number, or the decrease relative to the bigger one.

Those on OSS drivers might want to do a rerun that aint on the dev branch, just tried 21.3!

!link

A couple of points more than my last result but nothing extraordinary.
!link
BIOS-switch is set to "normal" mode AFAIR.

Last edited by crt0mega on 4 Jan 2022 at 5:14 pm UTC

This is my 6700XT, a good little performer.
!link

RTX 3080, first random run without any tweaks:

!https://i.imgur.com/vDtvHjZ.png

First run with new graphics card (RX 6900 XT) Not planning to tweak it before new PSU arrives, my current 550W is very underpowered for that card.
!https://i.imgur.com/CD5nlx4.png

550W

Thats risky !

i just need to edit the recording I did and uploaded it on Youtube but you can see all my other Benchmarks here.
https://www.youtube.com/playlist?list=PLnIYWF-c4CDZvfWizXwZ_33nXc16sMgj5