• 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**

It would be interesting if someone compared the performance of a mantle game, between Win 10 (10041) and an older os, considering mantle seems to have a performance boost on win 10, from what the api tests all show.

Around 3-5M extra draw calls per second from what I have seen. Still 2-3x more than what dx11 pulls :P
 
Last edited:
was it purely a draw call issue they suffer from ?

I suspect No API can make up for questionable coding and rushed deadlines

me thinks this will just lead to more laziness and dependency's on Nvidia

I remember a lot of mentions that Ubi added more draw calls then what DX11 could handle.

Take with grain of salt I'm not 100% sure.
 
Windows 10 Build 10041
Driver version 15.200.1012.2
290P 1100/1400
[email protected]

http://www.3dmark.com/3dm/6381405

DX11 Multi-threaded draw calls per second 1135766
DX11 Single-threaded draw calls per second 1179079
DX12 draw calls per second 22018815
Mantle draw calls per second 20217273

I'm assuming those are at 720P ? what i just realised is that the 3DMark link doesn't say.

I may have to link to posts... :o

Anyway. Done for now. please let me know if you spot any mistakes :)
 
Last edited:
I'm assuming those are at 720P ? what i just realised is the the 3DMark link doesn't say.

I may have to link to posts... :o

Anyway. Done for now. please let me know if you spot any mistakes :)

Yes 720P. My 3dmark link shows me under "show result details" and then "settings" that it was run in 720P but maybe it does not show that for you.
 
Last edited:
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

Nvidia guys can take part in this with DX12, in fact we have one on the board now. :)

Anyway, looking forward to your results :)
 
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?

Why do people always try to drag things down to this level?

This isn't a benchmark, FM have explicitly stated it was not a benchmark and that it should not be used to compare different systems. Looking at one very specific subset of a beta API using beta drivers is not going to tell you the full story. Just look at what happens in a DX12 benchmark (an actual benchmark) that uses more than this one specific function. http://www.anandtech.com/show/8962/the-directx-12-performance-preview-amd-nvidia-star-swarm

It's a shame, but not a surprise that this tool is being twisted to fit an agenda. So rather than a discussion over how the different API's perform in comparison to each other on a set hardware platform (and why no DX11 results?) we get a meaningless ranking chart that gives away little.
 
Last edited:
It's a shame, but not a surprise that this tool is being twisted to fit an agenda. So rather than a discussion over how the different API's perform in comparison to each other on a set hardware platform (and why no DX11 results?) we get a meaningless ranking chart that gives away little.

Why do people always try to drag things down to this level?

:cool:

**Official 3DMark API Overhead Leaderboard**

Is the thread title and great insight to what's coming, well done Humbug.:cool:
 
Back
Top Bottom