• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

**Official 3DMark API Overhead Leaderboard**

@ Triss and Layte

Different vendors and different overclocks have different results. just like any application which calls its self a benchmark, there fore it can be used as a benchmark.

Its no different to using games as benchmarking tools.

You have made your point, i don't agree, so i would appreciate it if you didn't derail this any further.

well found a press release about the driver i need but i'm sure i tried to download it through windows update before and it pretty much made the card stop working.

Maybe contact Nvidia? fire off an Email to them. or use Twitter.
 
Last edited:
Pointing out the obvious flaw in the basic premise of the thread is not a derailment. Ranking scores in such a way makes no sense if you understood what FM have given us. The only way this tool can provide meaningful data is for people to post the results for each API on their own system as a unique entity. Ranking them doesn't work, the developers themselves have stated that.

This post tells me far more than a random bunch of different setups sorted by score. http://forums.overclockers.co.uk/showpost.php?p=27832904&postcount=5
 
Last edited:
@ Triss and Layte

Different vendors and different overclocks have different results. just like any application which calls its self a benchmark, there fore it can be used as a benchmark.

Its no different to using games as benchmarking tools.

You have made your point, i don't agree, so i would appreciate it if you didn't derail this any further.



Maybe contact Nvidia? fire off an Email to them. or use Twitter.

I didnt know mentioning what the designer of the feature said was derailing, You yourself posted a SS from them in the first post, i had simply assumed you wernt aware of what they said :rolleyes:
 
@ Triss and Layte

See how many draw calls your PC can handle with each API before the frame rate drops below 30 fps
Our systems all perform differently, we can improve the performance of our own systems by overclocking.

We have been doing that and comparing our work to eachothers since the dawn of Computer Hardware, since when did that become a crime? just now it seems.
Really?

If everyones system performed the same i can understand why you two would be so against it, but its not... this thread is about doing exactly what we as Hardware enthusiasts do all of the time and have been for a very very long time.

Why are you guys even here telling me i shouldn't do this, its as if your not even a part of it... dictating something you don't understand.
 
Last edited:
@ Triss and Layte

Our systems all perform differently, we can improve the performance of our own systems by overclocking.

We have been doing that and comparing our work to eachothers since the dawn of Computer Hardware, since when did that become a crime? just now it seems.
Really?

If everyones system performed the same i can understand why you two would be so against it, but its not... this thread is about doing exactly what we as Hardware enthusiasts do all of the time and have been for a very very long time.

Why are you guys even here telling me i shouldn't do this, its as if your not even a part of it... dictating something you don't understand.

"Dictating something you don't understand"
Would you care to show me where i have done this?

What i did is post what the designer of said software said about it ,As i said i assumed you hadn't seen it. But somehow you decided to turn that into a personal attack.. :( and even decided posts from 3dmark own guide on the software was derailment. You often have good points and info Humbug to offer so i will forgive you this time
 
Drivers: 15.3 BETA
CPU: i7-3770 @ Default
GPUs: R9 290 @ 1000/1260
OS: Win 7 X64

720P:

DX11 Multi-threaded draw calls per second:722929

DX11 Single-threaded draw calls per second:761208

DX12 draw calls per second:0.0

Mantle draw calls per second:13543035

http://www.3dmark.com/3dm/6406145
 
Layte it's clear from your many posts you have disdain for AMD, if these guys want to do these tests that's their choice, I'm not sure why you think you have the right to tell them they can't?

If you don't like the info don't read the thread, but don't haunt the thread with your negativity it makes you look a bit childish and petty. Simply move on mate

Humbug I have a 290 and a 280x I can test on my 4770k later once I get win10 installed etc
 
Layte it's clear from your many posts you have disdain for AMD, if these guys want to do these tests that's their choice, I'm not sure why you think you have the right to tell them they can't?

If you don't like the info don't read the thread, but don't haunt the thread with your negativity it makes you look a bit childish and petty. Simply move on mate

Humbug I have a 290 and a 280x I can test on my 4770k later once I get win10 installed etc

While I agree if people want to rank the results then go for it (doesn't harm anything, unless you begin to extrapolate from it), but I also appreciate the fact that what the tools intended purpose is for is brought to light.
 
5820k @ stock 1 x 980 @ 1329Mhz

6cores / 12Threads

720p - http://www.3dmark.com/aot/5303

DX 11 ST - 1 , 425 , 939
DX 11 MT - 2 , 533 , 054
DX 12 -16 , 843 , 506

1080p - http://www.3dmark.com/aot/5318

DX 11 ST - 1 , 059 , 887
DX 11 MT - 2 , 506, 695
DX 12 -16 , 404 , 070

1440p - http://www.3dmark.com/aot/5332

DX 11 ST - 1 , 088 , 005
DX 11 MT - 2 , 587 , 362
DX 12 -16 , 911 , 238

4k - http://www.3dmark.com/aot/5356

DX 11 ST - 1 , 073 , 942
DX 11 MT - 2 , 503 , 742
DX 12 -18 , 319 , 925

4Cores / 8Threads

720p - http://www.3dmark.com/aot/5382

DX ST - 1 , 260 , 382
DX MT - 2 , 303 , 671
DX 12 - 15 , 894 , 249

1080p - http://www.3dmark.com/aot/5403

DX ST - 1 , 262 , 470
DX MT - 2 , 368 , 451
DX 12 - 15 , 944 , 485

1440p - http://www.3dmark.com/aot/5415

DX ST - 1 , 272 , 788
DX MT - 2 , 367 , 598
DX 12 -15 , 931 , 828

4k - http://www.3dmark.com/aot/5431

DX ST - 1 , 277 , 452
DX MT - 2 , 373 , 996
DX 12 -15 , 943 , 406

2Cores / 4Threads

720p - http://www.3dmark.com/aot/5442

DX ST - 1 , 214 , 325
DX MT - 1 , 324 , 063
DX 12 - 8 , 543 , 926

1080p - http://www.3dmark.com/aot/5456

DX ST - 1 , 252 , 082
DX MT - 1 , 265 , 427 ,
DX 12 - 8 , 588 , 916

1440p - http://www.3dmark.com/aot/5470

DX ST - 1 , 214 , 796
DX MT - 1 , 305 , 444
DX 12 - 8 , 583 , 612

4k - http://www.3dmark.com/aot/5477

DX ST - 1 , 236 , 779
DX MT - 1 ,312 , 037
DX 12 - 8 , 266 , 831
 
Last edited:
Good job Bug, about time we found a way to measure and test API performance. Makes me considering installing Windows 10 to test DirectX 12.

Will put up some results over the weekend.
 
What's Abundantly clear is DX12 has massive gains as an API over DX11, that alone should be enough to get gamers excited.

One thing is also apparent it will breathe a lot of life into lower end pc hardware that supports it.
 
I'll just submit here the same results I added in the amd driver thread using the benchmark default settings. I may go back and do a 1080p run if needed later.

Here are my results listed for both windows 8 and 10.

Powercolour R9 290 PCS+ (1170/1600) (Elpida)
4790k @4800

Windows 10 Build 10041
Driver version 15.200.1012.2 (latest from windows update)
http://www.3dmark.com/aot/3268

DX11 Multi-threaded draw calls per second 1233983
DX11 Single-threaded draw calls per second 1298470
DX12 draw calls per second 22198217
Mantle draw calls per second 21425637


Windows 8.1
Catalyst 15.3 Beta (14.502.1014.0)
http://www.3dmark.com/aot/3336

DX11 Multi-threaded draw calls per second 1052212
DX11 Single-threaded draw calls per second 999679
Mantle draw calls per second 16211641
 
Back
Top Bottom