Radeon RX Vega 56

Hi guys,

So after using MSI Afterburner and running Heaven Benchmark 4.0 the results are thus:

GPU operating more or less at 99% of the time

CPU: Core 1 average about 50% usage... the rest of the other cores are more or less 10% or so.


What do you reckon this says?
 
GPU is definitely not bottlenecked and most likely will work at its full potential except in special cases you shouldn't bother with right now. You're okay in the GPU field and getting what you paid for.


Your CPU is acting up, tho. Mine works with all cores the same in Heaven. Could be an optimization thing with Heaven. Try with actual games. But either way, if the GPU still works at ~99% there too, you need not worry about it. Consider that only if it causes a bottleneck or if the same abnormality hinders CPU operations like compression or decoding.
 
Last edited:
So why is it my performances on the majority of tests are very noticeably lower than averages? A dud GPU? Or perhaps my CPU is just on its arse, not able to communicate well with the GPU? There must be something. Maybe even something as small as just being DDR3 ram, which I know shouldn't have that much of an impact.

Edit: Cheers, will give it a whirl with actual games and see what happens.
 
Last edited:
So why is it my performances on the majority of tests are very noticeably lower than averages? A dud GPU? Or perhaps my CPU is just on its arse, not able to communicate well with the GPU? There must be something. Maybe even something as small as just being DDR3 ram, which I know shouldn't have that much of an impact.

Edit: Cheers, will give it a whirl with actual games and see what happens.

99% GPU is not a poor performance video card wise. If the performance it gives, software/game wise, is low, then the video card is just weak, and Vega 56 is not a weak video card. 99% GPU utilization is a typical full non-bottlenecked utilization.

As for the DDR3, I have those at 1600Mhz and they work very well. Heavy games like Destiny 2 don't have problems with it playing at 60fps and more. My computer is a 2012 spec. comp. all around even video cards (I'm on SLI GTX 680 4GB). My case should be similar to yours... unless your 8-core CPU is much weaker than my 4-core CPU?

What majority of tests do you mean? You can run the same test with them.

Why do you think that?

Because it's managing to run at 99% utilization. That's what it means to not have a bottlenecked GPU. I thought you knew that!
 
Last edited:
Because it's managing to run at 99% utilization. That's what it means to not have a bottlenecked GPU. I thought you knew that!

AMD_man said:
I mean, if it is at 100%, and the card isn't, then yes, but that's not the only way, since it may happen that the game can't use 100% of the processor.

Benchmarks stress components individually to get accurate information about their performance. Of course you are gonna get 99% load, how else would the benchmark test your components?

The only way to know for sure is by testing multiple games.

Also, did you see this?
 
We're using Unigine Heaven that simulates gaming benchmarking. That's a combination of everything used for games, including GPU, CPU, VRAM and RAM. And GPU can never really be tested without having CPU involved.


Different games have different GPU and CPU utilization, of course. I suggested testing in general and not Heaven specifically. The test with Heaven shows that there is no bottleneck harming the GPU in Heaven. 99% GPU utilization means it's not facing a bottleneck in Heaven. That's something known in GPU benchmarking communities.


No bottleneck indications and targets in gaming:
1- GPU gets to ~99%.
2- CPU and everything else is used in lowest possible percentage. Yet still if they reach 100% without holding back the GPU from reaching 99% they are not a bother.



Bottleneck is having a component(s) holding back the main component. In game graphics main component is GPU. In file compression main component is CPU. In internal file transfer main component is storage. In network file transfer network controller is the main component. Any component holding those back in their department from reaching ~100 % is a bottleneck.


The topic is about bottlenecking the GPU, right?


Benchmarks stress components individually to get accurate information about their performance. Of course you are gonna get 99% load, how else would the benchmark test your components?


This is a stress/performance test to know the capability of the component, not a bottleneck test. It could be used to discover bottlenecks, but if it does not reach ~99% in the target component, it is not a good stress/performance test. Again, becnhmarking is not our goal here. It is finding bottlenecks thru them. We're trying to find a bottleneck here, not stress/performance test.



The only way to know for sure is by testing multiple games.

Also, did you see this?


You means it's a powerful GPU? Yes, I'm aware of that. Is it related to bottlenecks?
 
Last edited:
Different games have different GPU and CPU utilization, of course. I suggested testing in general and not Heaven specifically. The test with Heaven shows that there is no bottleneck harming the GPU in Heaven. 99% GPU utilization means it's not facing a bottleneck in Heaven. That's something known in GPU benchmarking communities.
No, it doesn't. It just means the benchmark can use 100% of the GPU. And this is a GPU stress test. Of course it's reaching 100%, what did you expect? All of the tasks are designed to use a high percentage of the card and not the other stuff, because the other stuff doesn't matter. That's why the processor load was so low.
No bottleneck indications and targets in gaming:
1- GPU gets to ~99%.
2- CPU and everything else is used in lowest possible percentage. Yet still if they reach 100% without holding back the GPU from reaching 99% they are not a bother.
I agree with number 1, in gaming.
I don't understand what you meant in number 2. Why does it have to be the lowest possible percentage?
This is a stress/performance test to know the capability of the component, not a bottleneck test. It could be used to discover bottlenecks, but if it does not reach ~99% in the target component, it is not a good stress/performance test. Again, becnhmarking is not our goal here. It is finding bottlenecks thru them. We're trying to find a bottleneck here, not stress/performance test.
Exactly, the card is supposed to reach 100%. This only proves the benchmark is doing its job.
You means it's a powerful GPU? Yes, I'm aware of that. Is it related to bottlenecks?
Yes, as the OP said in his post, there are other benchmarks of systems using the same card and a different processor, and the results are higher.
test.png
 
No, it doesn't. It just means the benchmark can use 100% of the GPU. And this is a GPU stress test. Of course it's reaching 100%, what did you expect? All of the tasks are designed to use a high percentage of the card and not the other stuff, because the other stuff doesn't matter. That's why the processor load was so low.


I repeat: the topic is about bottlenecks, not stress tests. Whither the benchmark reaches 100% or 1% GPU utilization has nothing to do with bottlenecks in a stress test in the first place. CPU utilization should be monitored too. But still, all it does it finding what a build with that GPU is capable of GPU wise. Reaching 100% is not an of course thing here. Try a single core Pentium 4 or even a Core 2 Duo with RTX 2080 Ti and no stress test will ever get the GPU to ~99%. Everyone concerned with bottlenecks know that. Look it up on Youtube. Many testers did it with games on unlocked FPS and very high settings, which is a stress test too.


A stress test does not necessarily mean reaching 100% GPU. Just Google Unigine Heaven Bottleneck and you will see. Many will show held-back GPU usage. Heaven is GPU intensive alright, but no matter what, games still use CPU and need some power from it.


I agree with number 1, in gaming.
I don't understand what you meant in number 2. Why does it have to be the lowest possible percentage?


To be clear, I meant "the 'No-bottleneck' indications and targets in gaming are."


The thread is about RX Vega 56, a GPU. So I believe the whole thing is about gaming. But let's see the OP. Not sure if he had a different initial thought at hand when this thread was started.


In no. 2 it doesn't have to be the lowest possible usage. It just has to be low enough to not bottleneck the GPU. I mean that those concerned with bottlenecks prefer lower usage in them and the lowest the better because this mean the CPU will have more life in it.


Exactly, the card is supposed to reach 100%. This only proves the benchmark is doing its job.


The benchmark is not what decides a build can reach 100% in GPU. The specifications of the build are what do. Only if the build is capable of 100% GPU on a benchmark, that benchmark can reach it. I referred to this above.


Yes, as the OP said in his post, there are other benchmarks of systems using the same card and a different processor, and the results are higher.
test.png


This only means that in this specific setup; i.e.. the VRMARK and its settings, a bottleneck is causing those two different results. Only percentage monitoring here confirms that, and there's non. So this is not a good bottleneck test even tho is does find a bottleneck. It could be an optimization issue if no records of CPU and GPU percentage usage is there. This is a VR test. It's known for being demanding.



Now, if heaven is tested with the CPU that makes the GPU give better performance in the above linked test, the performance will be the same as what the OP got since he already reached 99% (full performance actually, but GPU's don't like to go over 99%). No CPU can make a GPU go more than what it can on it's fixed clock speeds.


My recommendation for the OP is to not bother with bottlenecks right of the bat, but to play normally and monitor the practical frames-per-second count (or performance in general) for any frame drops at the preferred game settings. Only if bothering drops were noticed, one should look for bottlenecks if they want to clear that performance drop.
 
Last edited:
Would it be at least fair to say that the pictured test results provide at least a strong indication, maybe not guaranteed, that the CPU is holding things back? Is it possible that even the GPU running at 99% is just not being met sufficiently with the 8350s speed/architecture, etc. regardless of its utilisation? The 8350 is a bit of a grandpa now in respect of the vega 56 in all fairness!
 
Back
Top Bottom