• 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.

Nvidia DX12 driver holding back Ryzen

Panos said:
If you bothered to watch the video, you will see the Intel with RX480 was better also on DX12.
Same applies to the overclocked TXP against the the RX480s CF @ 1300, the video shows near the end. The 480s shouldn't be nowhere near the TXP, however they beat it with a big margin.
And where you see TXP, you can put the GTX1080Ti also. And 2 480s cost £400 these days, while the TXP £1180 and 1080Ti £700+
Click to expand...

1) i did watch the video.
2) im not talking about the CPU here, im saying that we all know AMD has tuned better for DX12 for ages, My point is that AMD could have tuned their AMD GPU driver to Work better with Ryzen thus increasing the perf further.

im not defending NV their driver for DX12 sucks. im saying AMD+AMD tuning could also be reason for a huge jump
 
smogsy said:
1) i did watch the video.
2) im not talking about the CPU here, im saying that we all know AMD has tuned better for DX12 for ages, My point is that AMD could have tuned their AMD GPU driver to Work better with Ryzen thus increasing the perf further.

im not defending NV their driver for DX12 sucks. im saying AMD+AMD tuning could also be reason for a huge jump
Click to expand...

Is a Gameworks game though that came out last year, and no patch exists to make it better on AMD CPUs.
Also you do understand DX12 supposed to reduce the dependency on the CPU not tax it. Which works on the Intel CPU, not as good as when the AMD GPUs were used, but especially on Ryzen is a huge hit. That doesn't materialise either on DX11.
 
and no patch exists to make it better on AMD CPUs.
Click to expand...

i'm talking about the AMD GPU driver eg crimeson NOT game patches

this is my point
if AMD knows for example switching CCX's on AMD CPU requires 140ns (which it does) they could code in to their AMD GPU driver to use instruct the threads on Ryzen to use the the first 4 cores more effectively (increasing their lead) over Nvidia on AMD cpu system.

a good/interesting test would be

Test 1: Ryzen + RX480 + Orignal Tomb Raider driver
Test 2: Ryzen + RX480 + last updated driver

how big is the difference?

once AMD optimized Their GPU driver for Tomb Raider, they would generally optimize further.

so how big is the gap between Test 1 & 2.
if the gap is large then AMD GPU drivers are helping out the tuning of AMD CPU's

thus increasing their lead over running AMD+Nvidia.

that not to say this is wrong, or NVdia/intel drivers are amazing

the point is: would an AMD+AMD combo be much more effective
(if AMD are actually tuning CPU instructions in their GPU driver)

Nvidia did it with DX11 + intel, so why not AMD+AMD
 
Some more evidence showing an AMD Furyx achieving good results compared to a TitanXP with Ryzen.

https://www.reddit.com/r/Amd/comments/62s11m/german_tech_magazine_ct_did_a_test_with_a_fury_x/

OjSIcgM.jpg


The following shows the 720P results comparing the Ryzen with TXP and FuryX:

Hh02eaV.png


We have to consider that the FuryX is roughly half the speed of a TitanXP so the results are significant. The single FuryX is less than 10% slower than a TitanXP with Ryzen which means either it's a cpu bottleneck like people are claiming or the Nvidia driver is holding back performance on Ryzen, even in DX11. Using dual FuryX would confirm this either way.

I suspect though that If they had used mGPU FuryX, the Ryzen+AMD gpu combo would probably be faster than the Intel+TXP in all the games tested.
 
Last edited:
I don't think this is showing what you/people think it is showing - dual FuryX might help to confirm though.

People need to pull out stuff like process explorer and look at what is going on.
 
fs123 said:
We have to consider that the FuryX is roughly half the speed of a TitanXP so the results are significant. The single FuryX is less than 10% slower than a TitanXP with Ryzen which means either it's a cpu bottleneck (unlikely at 720P) like people are claiming or the Nvidia driver is holding back performance on Ryzen, even in DX11. Using dual FuryX would confirm this either way.

I suspect though that If they had used mGPU FuryX, the Ryzen+AMD gpu combo would probably be faster than the Intel+TXP in all the games tested.
Click to expand...

Bloody hell! This is mental.

I hope a lot more publications actually look into all of this.
 
Yeah Nvidia are holding back Ryzen performance, as they want to penalise themselves, make everyone buying a Ryzen, pair it with an AMD card as well, instead of one of theirs :p

PapaLazaru said:
They had to show it off the the best GPU of the moment, it would have looked a bit stupid showing it off with a 480.
Click to expand...

No, they need to show it off running at its absolute best, if its absolute best would have been it being used with an AMD card, then thats what they should have used, what another massive **** up from AMD :D
 
Last edited:
LoadsaMoney said:
Yeah Nvidia are holding back Ryzen performance, as they want to penalise themselves, make everyone buying a Ryzen, pair it with an AMD card as well, instead of one of theirs :p

No, they need to show it off running at its absolute best, if its absolute best would have been it being used with an AMD card, then thats what they should have shown.
Click to expand...

Or more likely, their DX12 driver is not actually doing much hence why the Intel cpu is also gaining fps with a 480 mgpu setup.

Regarding the use of TitanXP, I guess even AMD may not have realised that the Nvidia driver may have issues with Ryzen.
 
LoadsaMoney said:
Yeah Nvidia are holding back Ryzen performance, as they want to penalise themselves, make everyone buying a Ryzen, pair it with an AMD card as well, instead of one of theirs :p
Click to expand...

Like apart from yourself, no one's blaming Nv for Ryzen anything, only that they are crap at DX12.:p
 
nothing to do with amd backing all tombraider games then....also having a very deep involvement with them :p
 
They must have known, are are they too dumb to run their own CPUs with their own cards as well.:p

Dx12s crap anyway, massive fail as well :D


Theres not long to go now, before we see why they've only been showing Vega running DOOM and BF1 :p
 
Last edited:
~>Dg<~ said:
nothing to do with amd backing all tombraider games then....also having a very deep involvement with them :p
Click to expand...

ROTTR is an Nvidia sponsored GW's title, the cynic in me doesn't wonder if AMD were involved, as they had most to gain, it would probably have launched with Async DX12.

LoadsaMoney said:
They must have known, are are they too dumb to run their own CPUs with their own cards as well. :p
Click to expand...

Just too dumb, if they new 480 CrossFire performance was there they'd have been all over a £400 v £1200 comparison.
 
~>Dg<~ said:
nothing to do with amd backing all tombraider games then....also having a very deep involvement with them :p
Click to expand...

AMd only got involved in Tombraider 2013 and only provided TressFX. ROTTR is a gameworks game and uses Purehair which is derived from TressFX but never credited to AMD because AMD open sourced it without any conditions. There are no AMD logos in the game.
 
tommybhoy said:
Like apart from yourself, no one's blaming Nv for Ryzen anything, only that they are crap at DX12.:p
Click to expand...

Not a big deal though is it, with Dx12 being a massive fail, all we've got is a couple of Dx11 games patched with it laters, and hardly any using Vulkan.

I don't care that Nvidia are crap in a couple of games compared to AMD, as i know that when it gets here proper, and in a good number of games, when it matters, they won't be crap, just like the tessellation.

All them years AMD had a tessellator on their cards while Nvidia didn't, and then when the games started coming, Nvidia done a one, and BANG!, AMD were smashed, one hit!.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom