• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

guru3dDoom Vulkan Benchmarks

Page 8 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
If your hardware takes five years to be fully utilized, you didn't design a forward looking piece of hardware, you designed a dud.

Oh look, our OpenGL drivers was so rubbish that if we move all the work on the developers shoulders this game gets 50% faster! Wow! What an achievement 🙂

Small developers with must be absolutely thrilled. Actually middleware companies must be thrilled for real as the entry bar has been raised even more and their products are now even more important.

BTW, why stop here? Let's give access to the bare metal. Also compilers are for losers, time to write those thousands and thousands of shaders in assembly 🙂

Sorry, I'm having a hard time hearing you over the sound of my 290X burying its original competition, the 780 Ti, in Doom running in Vulkan, and gliding past its newer competition, the 980. 😛 That's what I call forward looking -- competitive on release, as the whole original GCN lineup was, and remaining competitive against newer hardware thanks to new software released years later.

And DirectX 11 isn't going anywhere, small developers can still use it. And if a developer is so small that it doesn't have the resources to code a game in DX12, they likely aren't going to make a game with spectacular visuals worth benchmarking against, anyway.
 
That and nVidia making it as difficult as possible for AMD to be able to optimize for their GW features.
Aren't they closed libraries for most part? Plus, any changes by devs to closed code need to be approved by Nvidia. Nvidia doesn't have to get their code writers working intensively with devs, but only when needed. I don't know why people here were lauding the damned curse which gameworks really is. Now compare that with Vulkan, and you could see why devs like latter, or any derivatives from mantle better than they do gameworks.
 
Sorry, I'm having a hard time hearing you over the sound of my 290X burying its original competition, the 780 Ti, in Doom running in Vulkan, and gliding past its newer competition, the 980. 😛 That's what I call forward looking -- competitive on release, as the whole original GCN lineup was, and remaining competitive against newer hardware thanks to new software released years later.

Haha 10/10.
I recommend Gelid Icy Vision GPU cooler. I have it on my 290 and is super silent. Compared to stock cooler gives additional 10% more performance then accounting throttling of reference cooler 😉
 
I don't understand what he's saying is wrong. Are the textures loading slower with the nvidia drivers?

Yes, on the Nvidia cards the hi res textures are loading slower so as you progress through the map, new areas are blurry for a few seconds before the hi res textures load. This has been an issue in the past with idtech 5 games. They tested the 970 and a 780 against a 480 with 8gb. It'd be nice if they could investigate a bit further. Do 3 gb amd cards do the same? I don't remember seeing it when testing a 280 but it's not my main card. Does the same thing happen in ogl and vulkan? Perhaps in vulkan some of the async work on amd cards is helping mitigate the issue?
 
Yes, on the Nvidia cards the hi res textures are loading slower so as you progress through the map, new areas are blurry for a few seconds before the hi res textures load. This has been an issue in the past with idtech 5 games. They tested the 970 and a 780 against a 480 with 8gb. It'd be nice if they could investigate a bit further. Do 3 gb amd cards do the same? I don't remember seeing it when testing a 280 but it's not my main card. Does the same thing happen in ogl and vulkan? Perhaps in vulkan some of the async work on amd cards is helping mitigate the issue?

I have observed this behavior on my FuryX when running Ultra at 3440x1440, so it may well be related to VRAM limits. That said, it occurs way more often using the OpenGL API compared to Vulkan.
 
I haven't reinstalled DOOM to test Vulkan, but I honestly don't recall texture pop in outside the start of a level loading.

Start of a map, for sure that thing reminded me of Bioshock haha. But after that initial hiccup I can't say I recall it.

And I'm a stickler for this kind of stuff, Dark Souls 3 is maddening how much texture pop-in/flickering it has. Drives me crazy, but thankfully the game has means to distract you of this flaw. Infuriating means.

EDIT: They only tested a GTX 970? Is that pesky 3.5/0.5 VRAM issue to blame?
 
Interesting that no other reviewer mentioned this:

https://www.youtube.com/watch?v=FVb25eomcrI

This guy got some flak for saying the image quality on the RX 480 during his review looked better than the 970. But it's true.


Whoa! 😱 MAJOR Texture Pop issues with nvidia, but not AMD?

nvidia users waiting though seconds of game time before textures finish loading?! WTF?!

That's a pretty big issue, and could potentially also affect FPS if the frames are allowed to be completed and shown before the textures load completely. That could mean an artificial boost to FPS count, at the expense of Visual Quality. Could just be a bug.... or not. Might be present for AMD too, but if the function is being performed asynchronously then the issue disappears since the bottleneck could be removed by AMD's real async scheduler.

Very interesting indeed!
 
Whoa! 😱 MAJOR Texture Pop issues with nvidia, but not AMD?

if I had to guess I would look at 3.5GB vs 8GB

on the Digital Foundry Fury X vulkan video he didn't use the highest texture quality settings, so my guess is you really need a lot of vram if you want max settings on this game
 
Interesting. Definitely an issue with texture streaming. If I'm not mistaken consoles have a similar issue. Not sure if to this extent.
 
if I had to guess I would look at 3.5GB vs 8GB

on the Digital Foundry Fury X vulkan video he didn't use the highest texture quality settings, so my guess is you really need a lot of vram if you want max settings on this game
I doubt that it is ram related at 1080p. That said, there are many with Fury/ 290 cards here, they could easily disprove what's being said about memory being the issue.
 
For any users running Eyefinity, Flawless Widescreen has released a patch that finally adds Vulkan support for proper FoV while playing DOOM with the Vulkan API.

The performance increase in the Vulkan API is mindboggling. Free performance upgrade and no loss in visual fidelity?! Yes please! I hope more major releases add Vulkan support and really let these GCN cards stretch their legs.
 
Someone should tell Sony, Microsoft and all those console developers they are doing it wrong.



No, they are doing fine, they need that stuff, its pc gaming that does not really need it, DX was created because we needed high level apis to make pc gaming possible, anyone could hace released a low level api at any point, yet, AMD does when DX9 [ST], glory days where over and they where unable to provide a proper DX11 or Opengl driver, the huge gains on AMD are bad opengl perf.. Whiout that it will be 5 to 10%...

Free perf now is fine, but im worried about the future, Kepler on Vulkan with its massive negative gains are a prime example of that, thats not Nvidia fault, what we gona do tomorrow when that stuff starts to happen to Maxwell, GCN 1.0... etc? specially when the list of arch to optimise for gets too large? the safest bet is to pick wharever the consoles have, and thats no good for us either!
The abtraction on DX was created to avoid those issues.

I think AMD may be trying to fight both Intel and Nvidia by dumbing down PC gaming to console level, that only good for THEM, no us.

MS having to create a abstracted mGPU solution for DX12 is also worriedsome.
 
Last edited:
No, they are doing fine, they need that stuff, its pc gaming that does not really need it, DX created because we needed high level apis to make pc gaming possible, anyone could hace released a low level api at any point, yet, AMD does when DX9 [ST], glory days where over and they where unable to provide a proper DX11 or Opengl driver, the huge gains on AMD are bad opengl perf.. Whiout that it will be 5 to 10%...

While yo uare right that AMD under performs in OpenGL that doesn't explain Fiji trouncing the 1070 in Vulkan.
computerbase-amdvulkanapiperf-1_674_57618.jpg
 
The performance difference between opengl and vulkan in this game is one of the oddest things I've seen when it comes to the subject of game performance.

cKQQnDa.png


http://www.hardwareunboxed.com/gtx-1060-vs-rx-480-in-6-year-old-amd-and-intel-computers/

The article wasn't even about Doom per se. I found it by chance, while searching for benchmarks for new GPUs running with old CPUs in old systems.

What is going on? How does CPU performance affect 480's numbers so drastically yet has so little effect on 1060?!
 
Last edited:
The performance difference between opengl and vulkan in this game is one the oddest things I've seen when it comes to the subject of game performance.





The article wasn't even about Doom per se. I found it by chance, while searching for benchmarks for new GPUs running with old CPUs in old systems.

What is going on? How does CPU performance affect 480's numbers so drastically yet has so little effect on 1060?!

Hard to say exactly, but it looks like AMD somehow has higher cpu overhead in vulkan than nvidia does even in opengl. Maybe it's an issue with AMDs vulkan driver, maybe it's just an issue with Doom? There's no way for us to know yet.
 
The performance difference between opengl and vulkan in this game is one of the oddest things I've seen when it comes to the subject of game performance.

cKQQnDa.png


http://www.hardwareunboxed.com/gtx-1060-vs-rx-480-in-6-year-old-amd-and-intel-computers/

The article wasn't even about Doom per se. I found it by chance, while searching for benchmarks for new GPUs running with old CPUs in old systems.

What is going on? How does CPU performance affect 480's numbers so drastically yet has so little effect on 1060?!
Very interesting. The 480 is not a card likely to be coupled with best CPUs, so that puts a dramatic new twist on things. Of course we cannot reach broad sweeping conclusions from one chart (nor can we with one Vulkanized game), despite earlier attempts by some to paint the Dooom/Vulkan results as somewhat of a representation of all future Vulkan (or DX12) performance.
 
Strange how little difference does vulkan make with slow CPU. I would expect the other way around. More scaling with slow CPU, less with the fast one.
 
That result is really strange because Doom isn't even fully maxing out my CPU (while getting great FPS) and I'm only on an i5, not even i7.

Is it do with SSE4/special instructions? Maybe a more modern i3, and the result would be completely different.
 
Strange how little difference does vulkan make with slow CPU. I would expect the other way around. More scaling with slow CPU, less with the fast one.
Looks like Doom really enjoys more (modern?) cores, especially under Vulkan.
Can someone finally test Phenom II x6 and dx12 titles?
As one of the comments mentioned, it'd be nice to see the X6 & Intel i7 9xx in there just to see how well the hex cores fare.
 
The performance difference between opengl and vulkan in this game is one of the oddest things I've seen when it comes to the subject of game performance.

cKQQnDa.png


http://www.hardwareunboxed.com/gtx-1060-vs-rx-480-in-6-year-old-amd-and-intel-computers/

The article wasn't even about Doom per se. I found it by chance, while searching for benchmarks for new GPUs running with old CPUs in old systems.

What is going on? How does CPU performance affect 480's numbers so drastically yet has so little effect on 1060?!

Those results don't match my own experience when I tested my r9 280 with a c2q 9300. I was getting easily 30-50% higher fps on Vulkan compared to ogl. Frame times were also much smoother just from the subjective feel. I tried the cpu at 3 GHz, 2.5 GHz, and 1.6 GHz and it wasn't until the 1.6 GHz underclock in Vulkan that the game felt as sluggish as it did in ogl with the cpu at 3 GHz.
 
Back
Top