The AMD Mantle Thread

Page 225 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.

Paul98

Diamond Member
Jan 31, 2010
3,732
199
106
Thats simply a hype. The only benefit is essentially CPU due to API. But AMDs DX driver already leaves much to be desired.


As I posted earlier from BF4 mantle page.

GPU/CPU graphs are rather interesting in BF4, GPU usage gets very smooth. In the 8350+7970 in directX ms time
GPU avg: 18.55, min 15.70, max 23.93
CPU avg: 18.87, min 14. 32, max 24.66
in Mantle
GPU avg: 11.61, min 11.44, max 11.86
CPU avg: 15.08, min 12.88, max 21.86
 

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
I find it hilarious how many people are off the slower but smoother bandwagon in the face of those frame time graphs.

I'm still waiting for minimum fps numbers to come out, but they should be massively improved.
 

Teizo

Golden Member
Oct 28, 2010
1,271
31
91
Well, the advantage for AMD if this is so is that it can help them with CPU sales.

Question, and this may be a dumb question this late in the game, but is Mantle for AMD GPU's only?

What I mean is...can someone take advantage of Mantle's CPU gains using an Nvidia GPU?

I ask this because like I mentioned...this could help AMD sell more CPUs and gamers could save money since the more expensive Intel CPU's won't be needed as badly.

I know AMD wants to use Mantle to help bolster GPU sales, but I wonder if they will in the future be willing to allow cross-vendor utilization...with motherboard drivers that will enable AMD CPUs to use Mantle irregardless of whatever GPU the person has.
 
Last edited:

Unoid

Senior member
Dec 20, 2012
461
0
76
I wish I could throw my two 290's in my main rig and test, but I don't wanna spend the hour swapping them out of my mining rig.
 

moonbogg

Lifer
Jan 8, 2011
10,637
3,095
136
Seems this could be great for 120hz users with dual high end cards. CPU has always been the problem there, and if Mantle lets those cards take off then that's super great. We'll have to wait and see for real though.
 

Paul98

Diamond Member
Jan 31, 2010
3,732
199
106
I will be interested to see what sort of CPU usage we get out of 8 thread+ cpus

Also vram usage between DX and Mantle
 
Last edited:

Noctifer616

Senior member
Nov 5, 2013
380
0
76
Question, and this may be a dumb question this late in the game, but is Mantle for AMD GPU's only?

What I mean is...can someone take advantage of Mantle's CPU gains using an Nvidia GPU?

I ask this because like I mentioned...this could help AMD sell more CPUs and gamers could save money since the more expensive Intel CPU's won't be needed as badly.

AMD pointed out that Mantle can run on more than GCN as long as the architecture supports the Mantle feature set (there is an abstraction layer just like in DX and OpenGL, but thinner). But it's up to Intel and NVidia to support Mantle.
 

Arkadrel

Diamond Member
Oct 19, 2010
3,681
2
0
The people that got ~25-58% faster performance outta this have to be happy.

I would consinder those amounts to be "big".

The people with fast CPUs and slow GPUs..... tough luck.

Seems like its much better to go fast GPU and slow CPU if needed.
 

Imouto

Golden Member
Jul 6, 2011
1,241
2
81
Then why were developers, like the guys from Dice, asking for this? Do you know something more about DX that they don't?

https://twitter.com/repi/status/428896915049414656

@repi Solutions that benefit all DX11 cards, like Driver Command Lists, have a bigger impact. Get AMD to implement it, for all developers!!!

@firefreak111 we support it but it is fundamentally broken in DX. check my slide #34 from 3 years ago: http://www.slideshare.net/DICEStudio/directx-11-rendering-in-battlefield-3
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
I get that, I'm talking about modern, cutting-edge games. You know, the games most of us here build our PCs to run...

Anyways, looking forward to review site results, and see how they differ from the results AMD gave.

There are lots and lots of modern games that Play just fine with $100 CPUs.

Just a few samples,

Metro Last Light, $100 CPUs are fine even with Titan
http://gamegpu.ru/images/stories/Test_GPU/Action/Metro%20Last%20Light/test/nastr/m%20proz.jpg

Tom Clancy's Splinter Cell Blacklist, $100 CPUs are fine even with Titan
http://gamegpu.ru/images/stories/Te...linter Cell Blacklist/test/Blacklist proz.jpg

F1 2013, $100 CPUs are fine even with Titan
http://gamegpu.ru/images/remote/htt...es-Test_GPU-Simulator-F1_2013-test-f1proz.jpg

Batman AO, $100 CPUs are fine even with Titan
http://gamegpu.ru/images/remote/htt...ction-Batman_Arkham_Origins-test-bao_proz.jpg

Assassins Creed Liberation,
$100 CPUs are fine even with Titan
http://gamegpu.ru/images/remote/htt...ins_Creed_Liberation_HD_-test-ac_hd_proze.jpg

Actually, the games that need more than $100 CPU to be playable is less than 1%.
 

B-Riz

Golden Member
Feb 15, 2011
1,530
676
136
Serious questions is serious: Will Windows version (8.1 vs 7) impact performance gain or will it be comparable?
 

PPB

Golden Member
Jul 5, 2013
1,118
168
106
No, he is just clueless. The green camp guys had been hitting the "command lists" drum for quite some time.

Guess what, according to johan andersson, command lists are "fundamentally broken" on dx. So no way you could have done the same, even if amd implemented command lists on dx.
 

ASM-coder

Member
Jan 12, 2014
193
0
0
...I wonder if they will in the future be willing to allow cross-vendor utilization...with motherboard drivers that will enable AMD CPUs to use Mantle irregardless of whatever GPU the person has.

AMD and Dice have stated it is open, and Johan(Dice) has said he hopes
others(Nvidia & Intel) will implement it.

Based on the Nvidia fan base on this thread, they have no use for it, so it probably won't happen.
 

HurleyBird

Platinum Member
Apr 22, 2003
2,725
1,342
136
Ok, so here's some quick number crunching.

Single player Kaveri example

DX:
26.60ms CPU time
26.44ms GPU time
Equally CPU and GPU bound.

Mantle:
23.29ms CPU time (14.21% improvement)
22.67ms GPU time (16.63% improvement)
Slightly CPU bound

Multiplayer 7970 + FX8350 example

DX:
18.87ms CPU time
18.55ms GPU time
Equally CPU and GPU bound

Mantle:
15.08ms CPU time (25.13% improvement)
11.61ms GPU time (59.78% improvement) !!!
Massively CPU bound

Single player 290X CF + 3970X example

DX:
13.24ms CPU time
18.38 / 2 = 9.19 ms GPU time
Massively CPU bound

Mantle:
8.38ms CPU time (57.99% improvement) !!!
14.02 / 2 = 7.01ms GPU time (31.10% improvement)
Slightly CPU bound

Interesting to say the least.
 
Last edited:

moonbogg

Lifer
Jan 8, 2011
10,637
3,095
136
Should be great for 120/144hz users.

Yeah, I can't believe people don't appear more exited about this actually. I find this possibility to be fantastic if it pans out, and according to the numbers it should. This could be the perfect solution for 120+hz high end users. If it works like they say, then AMD needs to get a freesync solution going because that would be perfection.
 

cmdrdredd

Lifer
Dec 12, 2001
27,052
357
126
I'm not from the states, so I don't pay the mining tax and neither does the rest of the world.

Are your prices in line with the US before the "mining tax"? Also I seem to remember that some of the world doesn't have real access to computer hardware in their home country no? Like some people have to order from elsewhere? Probably a small percentage.

I am not really familiar with how supply and pricing is in Europe for example.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
Here's a screenshot comparison on my rig to the example DICE/AMD gave on the battlelog website.

First off, here are the examples AMD/DICE gave, which were run on a 3960x and 290x CF rig at 1080p ultra:

Mantle off:



Mantle on:



And now here's mine, run at 1080p ultra. I have no idea why my screenshot is so clear, whilst AMD's so blurry. Motion blur is set to default, which is 50%. CPU is overclocked to 4.5ghz (although I'm pretty sure it wasn't running full speed) and GTX 770 SLI 4GB GPUs were at stock clocks, which is 1241 with boost. If you notice, it says I'm running Direct3D 11 even though I'm on Windows 8.1 Pro:

 
Last edited:

tonyfreak215

Senior member
Nov 21, 2008
274
0
76
Your CPU is 1 GHZ faster, that would make pretty big difference.

Also what MSAA? They were running at 4X.

Also try maxing out the blur (to bring it to the same look)
 
Last edited:

DiogoDX

Senior member
Oct 11, 2012
746
277
136
I will be interested to see what sort of CPU usage we get out of 8 thread+ cpus

Also vram usage between DX and Mantle
This first path seems focusing on CPU optimizations. GPU side like crossfire and memory will come later.

This patch is the first release of our Mantle renderer. We will continue to add improvements, based on your feedback, in upcoming Battlefield 4 updates as well. In particular we plan to focus next on adding multi-GPU frame pacing and a few additional GPU performance and memory optimizations.
http://battlelog.battlefield.com/bf4/news/view/bf4-mantle-live
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
The screenshots I posted were very telling, and just goes to show how under optimized AMD's drivers are when running the Direct3D 11.1 path.

Even my GTX 770s, which aren't even top of the line, outperformed the 290x CF rig. Only with Mantle do they offer comparable performance, and it's not like they are a huge leap over mine anyway.

Basically, this just goes to show what I, and others have been saying is true. AMD's drivers have inferior CPU utilization.. It's not that the GPUs aren't capable, it's the drivers which aren't properly optimized for multicore processors.

That's the only way I can explain such a massive difference..
 
Status
Not open for further replies.
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |