Quick test: Futuremark 3DMark v2.3.3663 Vulkan API Overhead Benchmarks
Click here to post a comment for Quick test: Futuremark 3DMark v2.3.3663 Vulkan API Overhead Benchmarks on our message forum
mtrai
You mean this?
http://i.imgur.com/x9yLEWX.jpg
mtrai
For kicks I just thought of something with how my system is set up and thought I would test it out.
PC Specs: I5 6600 OC 4.9 core/4.8 Unicore
16 GB ram at 3200
ASUS Z170-A
GPUs: RX 480 8 GB Red Devil
and
HD 7770 with just bios flashed to R7 250X
Since we are looking at a lot of numbers and comparing them across manufactures and gpus etc...I thought I would run the API on just the HD 7770 GPU and see the numbers...I disabled the RX 480 to force it 3dmark to use the HD 7770. Very Interesting DX 11 results.
http://www.3dmark.com/3dm/18805193
Comparison to my previous run this morning with the RX 480. Now I am very eagerly awaiting explanations on especially on the DX 11. I am not being a smart you know what. I am being serious.
http://www.3dmark.com/compare/aot/199856/aot/198443
THIS IS THE HD 7770 API just a few minutes ago.
http://i.imgur.com/DPLPkDV.jpg
THIS IS THE SCREENSHOT for the RX 480 this morning.
[img]http://i.imgur.com/x9yLEWX.jpg[/IMG]
Redemption80
https://s28.postimg.org/6p80dzzvh/Untitled.png
Everything pretty much at stock though.
The Doom comparison is completely irrelevant when this is the drawcall test, even if it was Doom has no DX12 or even DX11 version for comparison, so how is this showing the opposite?
Doom runs well because it's 1080/60 console game than could run on a phone, Vulkan only get the credit because OpenGL was so poor.
chumanga1
Does anyone with i7 and NVIDIA card already noticed major difference with HT enabled vs disabled for DX11 multithreading in this test?
This happens to me since first time 3Dmark overhead come out, take a look:
Results from today with HT disabled:
DirectX 11 Multi-threaded draw calls per second: 2 707 663
DirectX 11 Single-threaded draw calls per second: 2 286 665
Results with HT enabled:
DirectX 11 Multi-threaded draw calls per second: 2 233 946
DirectX 11 Single-threaded draw calls per second: 2 346 820
It's always like that for me, hyperthreading decrease DX11 multi-threading draw calls result.
Redemption80
HT Off
https://s28.postimg.org/6p80dzzvh/Untitled.png
HT On
https://s13.postimg.org/77wk9m66v/Untitled.png
chumanga1, yep same thing here as well.
mtrai
I am just finding this interesting that with the Vulkan support added we are also noticing the DX11 discrepancies.
Redemption80
chumanga1 mentioned it has always been like this with HT enabled.
HeavyHemi
As a note to others, the Vulkan API test has the same bug as does Doom using the Vulkan render in game: if you are running another nvidia model GPU, it will select that one to render vice your primary GPU.
Srsbsns
Why is the Fury X coming in so much behind the RX 480 when the Fury X is a much stronger card?
Robbo9999
May as well post my results too, interesting that your DX12 results are higher than Vulkan, for me it's the other way around, yet we both have Intel/NVidia.
DX11 mult:i 4,507,018
DX12: 27,231,630
Vulkan: 29,736,051
Anyone know why our results are swapped around & so different?
tensai28
Irrelevant since all of the games are dx11 and will be that way for the foreseeable future. That's why I made the switch to nvidia a few years ago.
Redemption80
I wasn't aware using MGPU made a difference, but appears to with DX12.
http://www.3dmark.com/compare/aot/200305/aot/200302
https://s27.postimg.org/f1xj3giyb/1_X970.png
https://s21.postimg.org/5ga91fu3b/2_X970.png
Also noticed the scores aren't comparable to old versions of the test, either that or the overhead in DX12 has gotten much higher and DX11 much lower.
October 2015
https://s28.postimg.org/lqbtlwl7h/Old.png
Undying
removed
beta-sama
http://www.3dmark.com/aot/200368
b101uk
for me,
DX12 was consistently hitting just over 17,000,000 draw-calls per second. (most 17,220,483)
Vulcan was consistently hitting just under 16,000,000 draw-calls per second. (most 15,985,491)
some other observations, the CPU use in DX12 MT was to within 1% across all 6 cores 12 threads, hitting 81% total cumulative CPU use and creating 12 equally shaped peeks , Vulcan MT the deviation in CPU use across 6 cores 12 threads was 15%, ranging from 76% to 91% CPU load, with an 84% cumulative total CPU load, creating 4 sets of shaped CPU usage peaks.
Dygaza
Something happening on Hilbert's setup. My Fury X at stock is getting 20M on both Vulkan and DX12. And I'm running 3770k @ 4.1GHz.
3dmark api in general has a lot of variance in performance between different drivers, though it has never really made any difference. On one driver I can have 18M , and on next driver I can have 22M under DX12. Or even DX11 atm seems to be somewhat bugged. Your fps is supposed to constantly drop when app adds more drawcalls. But at the end I have like 45000 drawcalls at 35fps, and next I get fps boost to 37fps when drawcalls are added to 48000. This shouldn't happen.
Overall I seem to be getting 5M drawcalls with 1C/2t, which is great throughput on both Vulkan and DX12. Since that's pretty much the max game developers have for their budget, as games do need to run a lot more than just rendering commands. So max 10M drawcalls per second for games until we get rid of 4C/4T or even 8t cpu's.
Redemption80
It's a fun benchmark....but it isn't even a benchmark as we can't even compare the scores lol.
Hopefully they can get Timespy onto Vulkan as well.
Undying
Or Firestrike. 😉
For now its just a another api overhead test.
Redemption80
Yeah, i think Timespy might be doing more technically, but Firestrike looks more impressive.
It's a nice test, but i don't get excited about the results when i got 17 million a few years back, but someone with a 1080ti gets 22 million now.
GeniusPr0
DirectX 11 single-thread
1 278 132 Draw calls per second
DirectX 11 multi-thread
1 868 381 Draw calls per second
DirectX 12
27 583 473 Draw calls per second
Vulkan
29 721 036 Draw calls per second
