Being a good goy in the current year

>>59670752
intel isn't paying nvidia, nvidia is just that shit on their own

>>59670752
I guess that DX12 patch wasn't all it was cracked up to be

>>59672119
>nvidia
>telling the truth
pick 1

File: wtf.png (73KB, 1284x618px) Image search: [Google]

73KB, 1284x618px

The fuck is up with this game.

File: gotem.png (69KB, 1115x562px) Image search: [Google]

69KB, 1115x562px

>>59672224
It's not so much the game as it is the GPU
You know this because you watched the same video I did

why in the holy fuck would intel pay to have directX 12 gimped?

>>59670752
Shitty nvidia driver/api bottleneck can't feed Ryzens cores.

Jen-Hsun Huang hates intel more than AMD

>>59672280
>exclusively on Ryzen
You need to learn how to think for yourself

>>59672318
Cash is cash and intel has always been more willing to put down AMD through shaddy tactics rather than create new products worth buying

>>59672378
you need to learn to be less paranoid i bet you got behind solar roadways hardcore as well

>>59672403
And you need to stop being so naive and ignorant. It's not like Intel hasnt done this shit in the past a lot of times.

>>59670752
It isn't intentional gimping. Nvidia's driver is just shit in DX12. They have a massive advantage in DX11 due to the multithreaded nature of their driver, whilst AMD's operates on a single thread, but in DX12 that situation's flipped turned upside down, with Nvidia only running on a single CPU thread and thus requiring all the single core performance they can get.

It's incompetence, not malice. The fact that Nvidia are complete shit in DX12 isn't news, and it's something that won't change until Volta at the earliest.

>>59672403
>I would fall for the solar botnet
insulting to even think about

>>59672440
show me one time along with proof i do not belive you for a second are you going to say the earth is flat next? or possibly that the earth is thje center of the universe

i get you HATE thunderfoot and want him fired as well

>>59672472
>botnet that barely remains functional and generates no power all while having underwhelming LED lights

its not a good botnet

>>59670752
>that near 100% crossfire scaling
holy shit

damn shame this basically never happens though

>>59672481
Cant you just google it, idiot? There are plenty of articles on the net of such cases.

https://www.nytimes.com/2014/06/13/business/international/european-court-upholds-1-06-billion-fine-against-intel.html?_r=0

File: nvidia_vulkan.png (269KB, 1220x3000px) Image search: [Google]

269KB, 1220x3000px

Ryzen + Nvidia Vulkan on Linux seems just fine. Whatever fuckup there it, it is related to DirectX 12 implementation.

Intel+Nvidia Vulkan
https://www.phoronix.com/scan.php?page=article&item=nvidia-vlk-madmax&num=4

Ryzen+Nvidia Vulkan
https://www.phoronix.com/scan.php?page=article&item=amd-ryzen-madmax&num=2

>>59673195
duh, explains terrible fps in dx12 bf1 as well and perfect fps in DOOM Vulcan
no wonder everyone tested in OpenGL DOOM

everyone also tested either with 1080 or Titan X

This guy has undergone some serious smear campaigns, just because he's willing to challenge conventional wisdom and highlight how fucking incompetent Nvidia is when it comes to Direct X 12.

BUT a lot of the points he has made are a direct result of AMD not having a card in the high end. Vega better be fucking amazing.

>>59670752
so nvidya cards are dogshit?
or nvidya + amd is dogshit?

i dont understand, someone just tell me who the jews are

>>59670752
How is NVIDIA gimping DX12? The NVIDIA card performs better in DX12 than in DX11 in the image you posted, what are you even saying?

>>59673587

Nvidia is incompetent.

No Jewing, just pure, unadulterated incompetence.

>>59673605
look at 16:43.
titan xp with ryzen 1800x is 80 fps.
in the same place, he gets 90 fps with xfire 480 and ryzen 1800x. how the fuck could that even be possible?

>>59673605
watch the video, nigglet.

>>59673605

Watch towards the end of the video when he slaps in crossfire 480s, against a single overclocked 1070 and see what happens to Ryzen frame rates

>>59673636
But the problem seems to be related to Ryzen + NVIDIA in this game specifically, not to DX12. NVIDIA runs better in DX12 compared to DX11 both on Ryzen and on Kaby Lel.

>>59673649
I don't see what that has to do with NVIDIA gimping DX12 either. CF also doesn't exist in DX12, what you're being impressed by is the DX12 AFR implementation in RotTR, not by CF. CF is only applicable to DX11 and below.

>>59672503
Where are you getting single GPU results from to determine scaling? All I can see is DX11 CF (utter dogshit) and DX12 mGPU (performs very well). Both of those test use 2 RX480s.

>>59673901
>But the problem seems to be related to Ryzen + NVIDIA in this game specifically, not to DX12. NVIDIA runs better in DX12 compared to DX11 both on Ryzen and on Kaby Lel.
>Ryzen + NVIDIA
that's why we are talking about gimping. something's not right.

>>59673901

They aren't gimping, they're just incompetent.

>>59673994
Should Jen-Hsun ask Lisa for engineering samples before launch?

>>59673994
ok, but why is this incompetence only shows itself in ryzen+amd setup? this bugs me.

>>59670752
No, DX12 is just a bad API

>>59673963
But it's not a problem on the DX12 side. It happens in DX11 too and only on Ryzen. It has something to do with how the NVIDIA driver runs on Ryzen (CPU and platform), it's not about DX12. Saying this sort of inaccurate shit is what muddles the waters even further and makes everything more confused.

On the 1070, Intel vs. AMD we get this:
>DX11: 7700K ~33.8% ahead of Ryzen
>DX12: 7700K ~33.5% ahead of Ryzen
The gap is identical across both APIs. Saying that "NVIDIA is gimping DX12" is fully retarded, whatever the problem is, it lies somewhere below the application-level graphics API.

The gap is also too large to be simply due to single-thread performance too (the 1800x is not that far behind), something else is going on with the NVIDIA driver. I can't even speculate as to what the real problem might be, I doubt anyone but NVIDIA could make an educated guess too. But the conclusion that it has something to do with DX12 specifically is completely stupid and unsupported by the results presented. If anything the problem is more serious than "just" DX12.

>>59674077
Nvidia optimized their code for Intel.

>>59674099
>>59674103
so, they generally fucked the driver up for dx12 and it fucks ryzen along the way.

>>59674138
No, it's completely unrelated to DX12 and it's fucked for Ryzen specifically. DX12 is irrelevant here, it doesn't show any different behavior on Ryzen compared to DX11.

>>59674078

t. Nvidia

>>59674138
nvidia drivers are fucking up on Ryzen, maybe even on purpose, but maybe not.

though, it does make sense for nvidia to hurt the AMD brand in every way possible, so gimping drivers on AMD's latest CPU isn't something beyond what they are willing to do.

>>59674066

AMD should have given them out to every tech company like popcorn when it was done with validation testing.

>>59672224
>>59672257
judaism, and if you complain about it thats antisemetic goyim

>>59674228
how come nobody thinked before to test with mGPU setup or with fury x? nvidia couldn't have predicted that. fucking tech press fucked this up.

>>59674103

This.

If they weren't so fucking incompetent at DX12 it would be a non-issue.

As it stands, AMD holds just as much responsibility for not releasing a high end GPU to compare it with.

>>59674267
*thought before

>>59674285

Because the tech press are lazy and incompetent who do nothing more than run benchmarks to verify and justify their incompetence?

>>59674302
fair enough.

>>59672224
Nothing, AMD is just shit.

>>59674437
great post noah, show those dumb goyim the light

I'm very curious to see if we'll consistently see better Ryzen performance in multi-gpu AMD setups or if a very powerful single card, i.e. Vega, will utilize the 8c/16t CPU to its fullest.

>>59674595
NO GOYIM AMD IS SHIT

>>59674636
Well, I mean, according to that video the 1080Ti does a pretty good job and even outperforms the 7700k setup, with the R7 1800X

>>59674687
OY VEY ITS NOT REAL, REMEMBER MY 20 GORILLIAN

>>59674753
Ordering 7700K as we speak, great Jew.

>>59674805
good goyim, greatest ally

People have started testing this.

https://www.youtube.com/watch?v=QBf2lvfKkxA&feature=youtu.be

>7700K is still 6-7% faster than a 1800X (not a R7 1700) using their own GPUs with Crossfire RX480

No you idiots, the 1800X is vastly better than the R7 1700, because you don't reach 4Ghz with normal OC without going crazy on the voltage.
The 1800X needs way less power.
And the 1800X is already running at its reasonable limits.

>>59675463
huh, neat

>>59675522
you can get it to 4.2 all cores

>>59675463
Interesting watch. Seems this may be a wider problem that affects both single and dual gpu setups.

>>59675463

novideo really put all their eggs into the DX11 basket

>>59676148

They still haven't released a DX12 optimized architecture. Volta won't be any different.

>>59675463

The thing that gets my hype train rolling is having an m.2 drive acting as cache for Vega and how infinity fabric will be making things even faster with Vega and Ryzen together.

>>59673901
mGPU isn't just enabled by default. Crossfire is still a thing in DX12.

>>59674077
You only see a big performance boost in the AMD rig when you jump from DX11 to DX12 because crossfire is remarkably better in DX12. On the other hand, the GTX 1070 won't gain that much from it because it's a single card.

>>59675522

The power consumption is the same on all CPUs. The only difference is literally the pre-OC. An 1800X can easily hold 3.9 out of the box on the cheapest B350, but the 1700 will barely OC to 3.8.

The 1700 is the same OC-d on a decent X370.