Nice Nvidia now beats AMD on Dx12

Im still curious if Nvidia can address what appears to be a fundamental archictecture problem with how Nvidia cards work? Note this is from my super laymans interpretation what was the initial controversy over DX12 benchmarking.

I believe the fundamental problem with DX12 was driver based, not some architectural problem.  Much of Nvidia's lead over AMD has been in massively efficient drivers, not just their superior hardware.  Efficient drivers that get around design problems in DX11 don't help if you fix those flaws for DX12.

Their lack of asyncronous computing is the only major problem I know of, and according to people that know, it's a driver issue, not a hardware issue.  Once they fix the drivers, they should be able to maintain superior performance with the current cards.  How long they'll stay superior?  No clue.

Cost wise, a Fury X isn't exactly a bargain deal, it's 600+, in the same price range as the 980 TI, so it's not like there's a major problem with having purchased an Nvidia card even at their current performance levels.  Based on history, performance optimizations will be much bigger from Nvidia than AMD, it's hard to tell how that will pan out in DX12 but even if they only stay slightly superior, they're still superior.

I'm sure this is all well and good, but it's not something that I care about. What I care about is, does the benchmark run better under DX12 than DX11? Prior to the latest driver release, the answer for my GTX 770 was No! By 10-20%.

But I re-ran the benchmarks with the latest driver:

Light-Medium-Heavy-Average...

DX11: 47.1 - 41.5 - 37.1 - 41.5

DX12: 45.6 - 42.5 - 37.1 - 41.4

The results are essentially a wash. Until they release a driver where it runs noticeably better under DX12 than DX11, as promised by everyone, I won't be satisfied.

?? eviator i am sorry but what are you talking about?

I have the same GTX770 and I do see a lot of difference

         Light-Medium-Heavy-Average
DX11:  21.9 - 18.4 - 16.1 - 18.5
DX12:  66.1 - 62.5 - 53.1 - 60.0

What will you say about that? its 3 to 4 times better in DX12.

This astounds me. Could it be CPU makes that much difference? What CPU do you have? Are you certain you are running the same graphics settings and resolution? I paid extra attention to make sure I was.

I have my Benchmarks in that link using 0.55 build

https://forums.ashesofthesingularity.com/472235/page/1/#3594691

i think it has to do with the cpu then.

Can you look at my benchmark results here: https://forums.ashesofthesingularity.com/472336/page/2/#3595397

The results indicate I am GPU bound.

Duplicate post

I think its a Win10 problem or need more Ram

Your CPU is old but Its good, and you have my same GPU.

It is the CPU, but it's because it's good, the problem is in reverse guys.

The FX 8350 should perform extremely poorly compared to a 2500k under the right circumstances.  The more that CPU is taxed, the more poorly it performs on the logical core level, down to about 50% of what it should be.

If the benchmark is taxing both logical cores, then DX11 is getting a little more than half a physical core, and a single 2500k core is more powerful than a single 8350 core to start with.  ASADDF, that's why your DX11 numbers are so poor, you can't feed even a 770 with a partial core running off already loaded shared assets.  Under DX12, it spreads that utilization across the four sets more evenly, and you're GPU bound without even trying.

The 2500k is probably enough juice to feed a 770 GTX with just one core.  Powerful individual cores mixed with lightweight GPU's aren't going to get much out of DX12, they're already making the most of their assets to begin with.  Having a 2500k myself, with a 680 GTX, very close in processing power, I overclocked to 4.3Ghz and saw only the slightest of improvements in DX11, I was only just barely cpu bound at stock clocks.

There does seem to be something wrong here though.  Eviator should have much better scores, going by yours.  Do you perhaps have a 4GB 770 GTX, and Eviator has a 2GB version?  It would take much more work to run off 2GB of ram.  If you guys actually have the same card, something is majorly hosed on his system, massive performance problem somewhere.

I do have the 4GB. Everything you said could explain some things, if true. My DX11 and DX12 numbers are very close as you predict.