Ryzen's poor performance with Nvidia GPU's. Foul play? Did Nvidia know?

[IllogicalGlory]

Except they try to ensure that tomorrow doesn't come. It's much more profitable to sell you the same gpu over and over.

Only if you're stupid enough to let them. NV's GPU release schedule has otherwise been quite reliable with a consistent 70% uplift in performance in the same power envelope in each bracket ever since Kepler.

 

[amenx]

Nvidia knows that anything on their part that may hamper its performance under ANY CPU is bad for them. I think they are aware of whatever shortcomings they may have with high multi-cores and are probably working frantically to address them. They would be stupid not to from a competitive/business sense.

 

[amenx]

This wouldn't be the first time that NVidia cards weren't cards that would last a long time. Use it for a year. Discard. Buy another NVidia card.

And there are certainly enough NVidia customers to keep doing this on a yearly or twice yearly basis.

Ah, Kepler... the one arch that couldnt keep up with the times now represents all Nvidia products and short longevity as a strict marketing policy to sell more cards to Nvidiots. That must be why so many are holding on to their 290s,390s as a year went but with clearly superior 1070/1080 as non-viable upgrade paths... still waiting for the elusive Vega.

 

[beginner99]

Maybe the devs don't know how to code dx12 yet. They need some time to learn the new technology and sharpen their skills.

That maybe but much more importantly what is missing is the entire toolset and a PURE DX12/Vulkan engine.

All DX12/Vulkan we see now is just "tacked on" more or less. And engine fully designed to take advantage of DX12/Vulkan and their multi threading capabilities is incompatible with dx11. This is still years out.

But this current issue is again solely due to the fact that NVs uArch is unsuited for DX12 and they are emulating it in drivers.

 

[3DVagabond]

Nvidia knows that anything on their part that may hamper its performance under ANY CPU is bad for them. I think they are aware of whatever shortcomings they may have with high multi-cores and are probably working frantically to address them. They would be stupid not to from a competitive/business sense.

I agree. I really doubt AMD sent them a kit to optimize their hardware for. Too likely they'd look for bugs and shortcomings to leak.

Besides, I see this as a win/win for AMD their cards and CPU's together give you better performance than nVidia with Intel. At least if the example shown is a trend and not an outlier.

 

[w3rd]

Nvidia knows that anything on their part that may hamper its performance under ANY CPU is bad for them. I think they are aware of whatever shortcomings they may have with high multi-cores and are probably working frantically to address them. They would be stupid not to from a competitive/business sense.

As of now, I believe you are correct.

In that Nvidia is realizing that what they have known for a while, is now coming back to haunt them. But I think this is much more of a bigger issue, than many may think. This is not a glitch, or an error. It is because of FERMI and CUDA and Nvidia resting on their laurels and being caught by AMD's uber development of their 64bit unified driver and Crimson suite. (Nvidia is going to have to write all new driver modules when Volta spins anyways, from what it sounds.)

AMD's Software advancements, are in part because of their HSA development. Which is a bit of an irony here. Because 4 years or so ago, ATI/AMD (Radeon) was getting hammered by fans & gamers a like, about their drivers, glitches, jitter, or IQ in games, etc. Heck, even the load time for the graphic catalyst suite was an issue. But that was before the gamer, Dr Su took the helm..! (I think a meme of Jen-Hsun Huang with his pants down, would be a satirical insert here.)


Again, I do not think Nvidia's driver can be rewritten easily. Perhaps someone else who is more knowledgeable can chime in, but I believe a complete rewrite of NV's unified suite, needs to be redone. Anyone of knowledge know as to the exact difficulty in rewriting their unified drivers..?

 

[scannall]

So, how many of these Ryzen reviews are going to be redone with AMD GPU's instead? Probably very few, since they already made all their launch noise. And, it's a problem the review sites should have already known about. But that doesn't make grist for the mill as it were.

 

[Shivansps]

This is stupid, nvidia has nothing to gain from this, i could make the same argument in reverse "there is a foul play on AMD drivers making Intel look bad" and it would make more sence.

Now the issue here is very simple, either nvidia drivers are not CCX aware or it just scales badly over 4 cores. Also i thnk they are still using SSE2?

 

[unseenmorbidity]

This is stupid, nvidia has nothing to gain from this, i could make the same argument in reverse "there is a foul play on AMD drivers making Intel look bad" and it would make more sence.

Now the issue here is very simple, either nvidia drivers are not CCX aware or it just scales badly over 4 cores. Also i thnk they are still using SSE2?

Idk why people keep saying this, because it's not true. Of course there were potential gains from both ends.

Nvidia's biggest worry is an AMD with money in the bank. AMD's Radeon Division has been keeping the freeloading CPU division afloat for years now. Now the freeloading faildozer is gone, and the rich billionaire Ryzen moved in. Ryzen looking bad, is good for Nvidia, because it means Ryzen sales will be lower, which helps Nvidia two fold. Less money for AMD, and the issues with their drivers on high core count CPUs is less mainstream.

However, it does make their GPUs look worse in some titles, when compared to AMD. However, since most reviewers test with Nvidia, this isn't a huge issue. Most people would see such discrepancies and AMD bias, or AMD optimized titles, even if they aren't.

On the other hand, Nvidia's drivers don't work well with Ryzen sometimes, which in turn makes AMD's GPUs look better. This could potentially boost sales for AMD GPUs, especially for Ryzen builders.

 

[imported_jjj]

Nvidia can't stop the tide and Intel is moving to more cores too.
Plus the cat is out of the bag and it's not going back, even reviewers will be aware of the issue by the time Vega arrives.

For now it's really manageable as one can just use DX11 instead but it would become a disaster if Volta has the same issues.

 

[crashtech]

It will be interesting to see if Nvidia fixes this before Vega drops.

 

[Topweasel]

It will be interesting to see if Nvidia fixes this before Vega drops.

They kind of have to, I am sure with the Titan Xp now and the Titan XP prior being more than 1k that a large portion of their users (not the largest but a significant portion) are Intel HDET owners, so 6850, 6900, and a handful of 6950 guys. Knowing that the cards are handicapped on their systems isn't going to help Nvidia in the long run (why spend a grand in the future on cards from a manufacturer that snubbed their superior CPU choice (i imagine a lot of these owners the Beemer or Merc equivalent of PC enthusiasts)). Eventually these guys will get past the "it's just a Ryzen optimization issue".

Then if Vega is even 75% of what people think it's going to be the scaling in DX12 will be marvelous and its going to put pressure on Nvidia to fix DX12 scaling even if it means they can't hide the performance penalty that AMD doesn't have with DX12. This issue all starts from there, Nvidia did everything they could to hide the fact that there is a penalty not seen on competitors in DX12.

 

[sontin]

Best thing about this thread:
A GTX1060 and Intel combination is much faster than Ryzen and RX 480 and yet people are blaming nVidia for AMD's new medicore processor.

 

[IEC]

Yep, that 1060 + Intel sure looks "much faster"

Just a few examples of "much faster":

Spoiler

Civ VI GTX1060

Civ VI RX480

Rocket League GTX1060

Rocket League RX480

 

[sontin]

If you only play two these three games - sure. If you use other games not so: http://www.techspot.com/article/1374-amd-ryzen-with-amd-gpu/

In 3 of 4 DX12 games the combination of GTX1060 and i7-7700 delivers better performance than the RX 480 and 1800X.

And dont let us start with pre-Low-Level-APIs...

 

[krumme]

Best thing about this thread:
A GTX1060 and Intel combination is much faster than Ryzen and RX 480 and yet people are blaming nVidia for AMD's new medicore processor.

Yes in old dx9 and dx9 derived games.
Enjoy your 4 thread cpu here.

AT bm suite isnt excactly bleeding edge.
If we take the only remotely new engine that tries to get away from dx11 drawcall problem - civ - its a huge win for the r5 480 combi.

But hey we have reciewers showing a 7600 is faster than a r5 1600x in bf1. How they have moral to publish such destructive and wrong bm is beyond me but i understand your confusion then.

 

[imported_jjj]

Best thing about this thread:
A GTX1060 and Intel combination is much faster than Ryzen and RX 480 and yet people are blaming nVidia for AMD's new medicore processor.

Nvidia has the same problem with Intel's products when those products have more than 4 cores.

 

[w3rd]

Again, I don't think this is anything nefarious by Nvidia.

And Nvidia is not doing this on purpose to Ryzen. They are doing it to themselves, on any 6+ core machine, no matter what. In which Ryzen happens to illustrate Nvidia shortcomings. All this really means is that Nvidia's software is not as advanced or as robust as AMD's.

And that Radeon cards have a distinct advantage when that threshold is met.

 

[sontin]

Nvidia has the same problem with Intel's products when those products have more than 4 cores.

There doesnt exist a problem with more cores. You guys need to understand what DX12 is. Right now i come to the conclusion that you and others believe that DX12 is just DX11 with a few new extentsion. Multi-threaded rendering with DX12 is in the hand of the developer. Is DX11 faster than the developer has not optimized the game. Dont blame nVidia for something out of their hands.

BTW: It is ironic that most problems happen in Gaming Evolved games like Warhammer, CIV6, Deus Ex and Battlefield 1. Think about it.

 

[krumme]

There doesnt exist a problem with more cores.

Depends on game engine. In bf1 mp 64 dx11 on amiens and some other maps a 6900 and 1800x consistently have frametimes over the equivalent of 60fps while a 7700k dips into the 40fps. A 7600 far worse.

So yes. Sometimes its better with more cores.
But i wouldnt buy a i5 today unless you tollerate dips to the 30 fps in the newer mp games. I changed mine and its nice to get rid of that slideshow in bf1.

 

[imported_jjj]

There doesnt exist a problem with more cores. You guys need to understand what DX12 is. Right now i come to the conclusion that you and others believe that DX12 is just DX11 with a few new extentsion. Multi-threaded rendering with DX12 is in the hand of the developer. Is DX11 faster than the developer has not optimized the game. Dont blame nVidia for something out of their hands.

BTW: It is ironic that most problems happen in Gaming Evolved games like Warhammer, CIV6, Deus Ex and Battlefield 1. Think about it.

You are refusing reality in your irrational need to support a corporation.
If you would stop to look at data and get informed ,you would look less clueless and trollish. Your initial comment was pure trolling to begin with, completely offtopic and aimed only at getting a reaction - and for that i'm ignoring you from this point on.

Nvidia has issues with more cores under DX12 and it's not a DX12 issue as Radeon GPUs don't have a similar problem.

 

[InfoFront]

Neither AMD nor Nvidia have anything to gain from GeForce cards performing poorly with Ryzen. There's no foul play here.

-AMD still prefer if you pair your GeForce with a Ryzen rather than an Intel CPU.
-Nvidia will gladly sell you a GTX 10x0 for your new Ryzen build, they're not going to deliberately sabotage performance and steer you away from GeForce just because AMD also happen to make GPUs.

As usual with Ryzen, things tend to get very emotional and overblown. A few games, mostly DX9, seem to not perform optimally with Nvidia cards at the moment. Doesn't mean there's a big controversy or something is "broken" and needs "fixing". It just means Nvidia, like everyone else, will have to profile and optimize for this brand new architecture. I'd expect these few performance anomalies to get fixed and general performance to improve in the next few driver updates.

Nvidia has something to gain.

They have a monopoly on high and upper-midrange GPUs. If you want a high end GPU (GTX 1070+) you have to buy Nvidia. Gimping the performance of their GPUs with AMD CPUs won't hurt their own sales at all, since consumers have no alternatve. It will only hurt AMD, which is definitely beneficial to Nvidia.

Gimping the performance of their midrange (1050-1060) GPUs slightly won't effect sales of those significantly. Most of the sheeple in the market for a mid-range or lower GPU simply extrapolate down the monstrous performance of the high end GPUs. There's some logical gymnastics that take place in consumers' minds. Something like: 1080Ti > All, therefore Nvidia > All. I digress, but the point is that Nvidia can gimp the performance of their mid range GPUs and it won't matter. Case in point: the RX480 is a better value than the GTX 1060, but the 1060 will outsell the hell out of the RX480 any day.

 

[sontin]

You are refusing reality in your irrational need to support a corporation.
If you would stop to look at data and get informed ,you would look less clueless and trollish. Your initial comment was pure trolling to begin with, completely offtopic and aimed only at getting a reaction - and for that i'm ignoring you from this point on.

I like your arguments here. :up:

Nvidia has issues with more cores under DX12 and it's not a DX12 issue as Radeon GPUs don't have a similar problem.

Did you even read what i wrote? The DX12 driver is not allowed to intercept with the application. Multi-threaded rendering is a job of the application. That's the reason why you see this huge jump in the 3DMarkt API test with DX12/Vulkan: http://www.guru3d.com/news-story/qu...v2-3-3663-vulkan-api-overhead-benchmarks.html

But i guess it is easier to "ignore" someone when you cant deal with facts. BTW you can clearly see that Tomb Raider is scaling well from DX11 to DX12 on a 8 Core/16T 1800X processor with a GTX1070:
DX11: https://youtu.be/nLRCK7RfbUg?t=405
DX12: https://youtu.be/nLRCK7RfbUg?t=351

You guys just need to chill. There is nothing wrong with Ryzen and nVidia. It is just a matter of early adapter DX12 games. And like someone said: These "problems" exist on Intel, too.

 

[Leyawiin]

Its an oversight, not a conspiracy.

 

[w3rd]

I like your arguments here. :up:



Did you even read what i wrote? The DX12 driver is not allowed to intercept with the application. Multi-threaded rendering is a job of the application. That's the reason why you see this huge jump in the 3DMarkt API test with DX12/Vulkan: http://www.guru3d.com/news-story/qu...v2-3-3663-vulkan-api-overhead-benchmarks.html

But i guess it is easier to "ignore" someone when you cant deal with facts. BTW you can clearly see that Tomb Raider is scaling well from DX11 to DX12 on a 8 Core/16T 1800X processor with a GTX1070:
DX11: https://youtu.be/nLRCK7RfbUg?t=405
DX12: https://youtu.be/nLRCK7RfbUg?t=351

You guys just need to chill. There is nothing wrong with Ryzen and nVidia. It is just a matter of early adapter DX12 games. And like someone said: These "problems" exist on Intel, too.

We can now see, that you are unaware of the actual argument.

This is not a scaling issue, or a game issue or even a DX12 issue. It is a Software issue with how Nvidia's drivers are handled in Windows 10. Nobody here, is discussing/comparing DX driver scores. We are discussing how NVidia's unified driver suite, is essentially a 32bit hackjob, stemming from FERMI & CUDA and was good enough, then.

Until now, which was illustrated when Reviewers started running Radeon 64bit drivers next to Nvidia's legacy drivers.





I am using laymen's terms here