• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Unigine Heaven 4 benchmark

Caporegime
Joined
12 Jul 2007
Posts
40,436
Location
United Kingdom
6900xt@3040/2120
[email protected]/4.0
32GB 4800mhz DDR5

21.11.3 Driver

7472

Great score and miles ahead in 1st place, well done. :cool:

Book me in for 4th place please @Kaapstad

5600x @ PBO @ 4900mhz
32GB @ 3900mhz CAS14
3090FE @ Core 2010 Mem 5251




  • Where are you at @LtMatt where is that AMD 6900XT beast lets see her run!!! We need some AMD results its all green atm!!

Getting Forza Horizon 4 bench thread flashbacks again when you got all cocky, only to then get beaten. You are going to need a 3090 TI to compete with the little old 6900 XT mate. :p

6900 XT 2900/2124Mhz
21.12.1
1080P
HO7JWCx.png

Merry Christmas. :cry:

I'll run this again when I'm rocking a Ryzen with VCache as I'm a bit CPU limited in this bench, seems to prefer Intel CPUs looking at the results.
 
Caporegime
Joined
12 Jul 2007
Posts
40,436
Location
United Kingdom
Great scores from the 6900XTs :cool: on a cold morning I reckon I could get a few more mhz out of mine.
Nice scores from the red team there, I didn't realise AMD would be as strong as this in Heaven 4.0. Will do a run later.
It's not had any optimisations since forever as its such an old benchmark.

Honestly surprised to see the 6900 XT sitting in No1 spot given how much this bench has always been dominated by Nvidia.
 
Soldato
Joined
6 May 2009
Posts
19,885
Interesting, with the 497.29 driver and 5800X and 3080FE@stock i'm getting lower scores than some on here with similar hardware but minimum fps is quite a bit higher.

RAM is overclocked (3600MHz, CL16 @ 3800MHz, CL14)

Some people hitting 20/30fps lows with high specced cards, something is wrong there and the card is probably throttling which will make gameplay terrible (if running those same clock speeds when gaming)

Surely 60fps as a minimum at 1080p should be the benchmark for current gen cards?
I did used to think it was all about max and averages but when gaming you only notice the minimums as the game runs like a dog if it suddenly drops.
Max fps is meaningless if you are capping fps to monitor refresh. A 10 / 20 / 30 fps miniumum will cause problems though.

6rYfYQv.png

There's a challenge. The highest minimum fps!
65.8 for me so far
 
Last edited:
Soldato
Joined
25 Nov 2020
Posts
2,506
Interesting, with the 497.29 driver and 5800X and 3080FE@stock i'm getting lower scores than some on here with similar hardware but minimum fps is quite a bit higher.

RAM is overclocked (3600MHz, CL16 @ 3800MHz, CL14)

Some people hitting 20/30fps lows with high specced cards, something is wrong there and the card is probably throttling which will make gameplay terrible (if running those same clock speeds when gaming)

Surely 60fps as a minimum at 1080p should be the benchmark for current gen cards?
I did used to think it was all about max and averages but when gaming you only notice the minimums as the game runs like a dog if it suddenly drops.
Max fps is meaningless if you are capping fps to monitor refresh. A 10 / 20 / 30 fps miniumum will cause problems though.

6rYfYQv.png

There's a challenge. The highest minimum fps!
65.8 for me so far

I'll try to remember to run another Benchmark on that driver to compare with yours. I've got the same CPU and GPU. I've not overclocked my RAM though. Only if it's of any interest like.

Are you running the GeForce/Ryzen auto OC tools?
 
Soldato
Joined
6 May 2009
Posts
19,885
I'll try to remember to run another Benchmark on that driver to compare with yours. I've got the same CPU and GPU. I've not overclocked my RAM though. Only if it's of any interest like.

Are you running the GeForce/Ryzen auto OC tools?

I don't even think its the driver, minimums have always been 55fps+ even on older (460.89) drivers.
RAM overclocking has probably helped minimums somewhat, maybe 10-15fps increase, so around 20-25% on default RAM clocks and timings.
Having a quick scan over this thread, I don't see any 1080p minimums over 60fps :confused:
Some are total spuds, like 10fps :D That'll be lovely when gaming!!

Nope, none of the other tripe like sound HD drivers, Geforce experience or anything.
Just GPU driver and MSI afterburner.

I am using a custom fan curve though, temps when benching are never over 70c. Gaming usually up to 60c max. Idle temps usually around 25c in a 20c room

UGPJd3W.png
 
Last edited:
Associate
Joined
4 Dec 2018
Posts
86
Great score and miles ahead in 1st place, well done. :cool:



Getting Forza Horizon 4 bench thread flashbacks again when you got all cocky, only to then get beaten. You are going to need a 3090 TI to compete with the little old 6900 XT mate. :p

6900 XT 2900/2124Mhz
21.12.1
1080P
HO7JWCx.png

Merry Christmas. :cry:

I'll run this again when I'm rocking a Ryzen with VCache as I'm a bit CPU limited in this bench, seems to prefer Intel CPUs looking at the results.
Thanks man. Those above 3ghz clocks arent the easiest to run either. But saved by the 12900k here i think.
 
Soldato
Joined
6 May 2009
Posts
19,885
Why are peoples minimums still quite low on the top spec machines? 5950X, 6900XT overclocked I would have thought something like a minimum of 70 or 80fps

A new higher minimum, 66.6fps

hlnEuXB.png
 
Caporegime
Joined
12 Jul 2007
Posts
40,436
Location
United Kingdom
Why are peoples minimums still quite low on the top spec machines? 5950X, 6900XT overclocked I would have thought something like a minimum of 70 or 80fps

A new higher minimum, 66.6fps

hlnEuXB.png
The benchmark is 13 years old in a few days.

It is likely something to do with lots of CPU cores and a very old, single threaded application.

It's not related to the GPU at all.
 
Soldato
Joined
6 May 2009
Posts
19,885
The benchmark is 13 years old in a few days.

It is likely something to do with lots of CPU cores and a very old, single threaded application.

It's not related to the GPU at all.
Yeh, I get that it's an old benchmark but why should my stock 5800X outperform others and higher clocked CPUs at minimum framerates?
Like you say it probably means nothing in current gen games but would still be interested to know why 60fps+ minimum in this aging benchmark is not achieveable many others. Is there a ways of increasing the minimum?
Also, a low fps minimum would indicate GPU throttling and probably would be great even in current gen games... Game working nice with fps locked at 120 / 144 or something, then fps takes a nosedive to 20
 
Associate
Joined
20 Nov 2020
Posts
1,120
Yeh, I get that it's an old benchmark but why should my stock 5800X outperform others and higher clocked CPUs at minimum framerates?
Like you say it probably means nothing in current gen games but would still be interested to know why 60fps+ minimum in this aging benchmark is not achieveable many others. Is there a ways of increasing the minimum?
Also, a low fps minimum would indicate GPU throttling and probably would be great even in current gen games... Game working nice with fps locked at 120 / 144 or something, then fps takes a nosedive to 20
This is the nature of such a benchmark. There are even scores with 10 FPS min. If all that matters is the final score, then you don't care about the min FPS. you tune your system to get the biggest score. -> more FPS most of the time instead of less but more constant FPS.
 
Soldato
Joined
6 May 2009
Posts
19,885
This is the nature of such a benchmark. There are even scores with 10 FPS min. If all that matters is the final score, then you don't care about the min FPS. you tune your system to get the biggest score. -> more FPS most of the time instead of less but more constant FPS.
Its not the nature of the benchmark. 10fps minimum would imply a system or hardware error, probably something throttling. I used to think that the final score mattered the most, sure, it does when going for highest score but if you game and get 800fps max and 1fps minimum but lock fps to 120 then the 800fps is meaningless. The 1fps low however would make the game unplayable. Best having 120fps constant with no drop below 120, surely
 
Associate
Joined
4 Dec 2018
Posts
86
Why are peoples minimums still quite low on the top spec machines? 5950X, 6900XT overclocked I would have thought something like a minimum of 70 or 80fps

A new higher minimum, 66.6fps

hlnEuXB.png

Cause of stutters. The bench almost stoppes UP during runs on most setups. Do it has a moment of almost 0 fps during the run.
 
Associate
Joined
20 Nov 2020
Posts
1,120
Its not the nature of the benchmark. 10fps minimum would imply a system or hardware error, probably something throttling. I used to think that the final score mattered the most, sure, it does when going for highest score but if you game and get 800fps max and 1fps minimum but lock fps to 120 then the 800fps is meaningless. The 1fps low however would make the game unplayable. Best having 120fps constant with no drop below 120, surely
But it is not a game that's the problem. Such a benchmark is a competitive tool where all that matters is the score and not the min FPS above a certain value. If you get the best score by having a min FPS of 10, then your card is the best.
Plus, the min FPS can be a single dip in the benchmark which happens for some reason on some systems, so the FPS will be better than yours 99.99% of the time.
 
Back
Top Bottom