The AMD Mantle Thread

Page 221 - 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.

Grooveriding

Diamond Member
Dec 25, 2008
9,108
1,260
126
http://battlelog.battlefield.com/bf4/news/view/bf4-mantle-live/

Mantle renderer now available in Battlefield 4

The new game update for Battlefield 4 brings the Mantle renderer that DICE has developed in conjunction with AMD. With this first release of Mantle, you can expect performance increases in the game if you have the hardware to support it. In this blog post, the Technical Director for Frostbite details this exciting new step for the Battlefield series.


Hi,
My name is Johan Andersson and I’m one of the Technical Directors in the Frostbite team and I have something very exciting to share with all Battlefield players: the latest Battlefield 4 update on PC includes the long-awaited new renderer in Frostbite that is using AMD’s Mantle.
Mantle is a new low-level graphics API that we’ve been working very closely with AMD on over the last 2 years and it is a major change & improvement to how we are able to program & use modern GPUs in order to get the most out of them.


Battlefield 4 on PC is already quite heavily optimized using DirectX 11 and DirectX 11.1, but with Mantle we are able to go even further: we’ve significantly reduced CPU cost in our rendering, efficiently parallelized it over multiple CPU cores and reduced overhead in many areas.
The biggest performance gains can be seen when the game is bottlenecked by the CPU which can be quite common even on high-end machines and this was main goal to improve on with Mantle. We’ve also been able to streamline and optimize some of the GPU workload. The end result is that game performance is improved in virtually all scenarios in Battlefield 4 on both Windows 7 and Windows 8 when running with Mantle!


This update is the first release of our new Mantle renderer, and we will continue to provide more tuning for different types of configurations in future updates. Still, we are really happy with the performance improvements that we are seeing with this update and we believe you will be as well.


Read on for the full details on how to enable Mantle, and some information on the kind of performance gains you can expect.
How to enable Mantle

In order to use Mantle you will need the following:

-AMD Catalyst 14.1 Beta drivers. AMD will post them here.
-AMD Radeon GPU with Graphics Core Next (Note: AMD Catalyst™ 14.1 initially supports the AMD Radeon™ R9 290X, R9 290, R7 260X and “Kaveri” APU (learn more))
-64-bit Windows (7, 8 or 8.1)


If you have the above you will see a new “Graphics API” option in the in-game Graphics Options settings after starting the game, here you can switch between using DirectX 11 and Mantle (remember to restart the game after changing).





If the “Graphics API” option is missing, that means you do not have a compatible GPU and driver and the game will use the standard DirectX 11 graphics API instead.
Performance tests

To compare performance with Mantle vs DX11 we in Frostbite have done a couple of benchmarks with Battlefield 4 on 3 different configurations & test scenes:
Test case 1: Low-end single-player
CPU/GPU: AMD A10-7850K (‘Kaveri’ APU), 4 cores @ 3.7 GHz
Settings: 720p MEDIUM settings.
OS: Windows 7 64-bit
Level: Singapore “Beach”
This area is heavy on the CPU as it is quite a lot of action going on, but the game is primarily GPU-bound both on DX and with Mantle but thanks to GPU optimizations with Mantle we get a good performance improvement.
Result: 26.6 ms/f -> 23.3 ms/f = 14% faster




Test case 2: Standard 64-player multiplayer
CPU: AMD FX-8350, 8 cores @ 4 GHz
GPU: AMD Radeon 7970 3 GB (AMD will add support for the AMD Radeon™ HD 7970 in a later stage of Mantle’s release schedule, learn more)
Settings: 1080p ULTRA 1x MSAA
OS: Windows 8 64-bit
Level: Siege of Shanghai
Level was tested with 64 “pseudo players” that we have for our own internal testing that simulates heavy game workload that we have in multiplayer in order to get more deterministic results compared to full real multiplayer. 64 players on the large Battlefield levels is really demanding of the CPU so this test case is primarily CPU-bound.
Result: 18.87 ms/f -> 15.08 ms/f = 25.1% faster




Test case 3: High-end single-player with multiple GPUs
CPU: Intel Core i7-3970x Extreme, 12 logical cores @ 3.5 GHz
GPU: 2x AMD Radeon R9 290x 4 GB
Settings: 1080p ULTRA 4x MSAA
OS: Windows 8 64-bit
Level: South China Sea “Broken Flight Deck”
This single-player scene is heavy on both the CPU and GPU with lots of action going on. Test was done on the highest end Intel CPU on Windows 8, which is the fastest option before Mantle thanks to DirectX 11.1. Still this CPU is not fast enough to keep the 2 290x GPUs fed at 1080p on Ultra settings so we get a significant CPU performance bottleneck which results in major performance improvement when enabling Mantle.
Result: 13.24 ms/f -> = 8.38 ms/f = 58% faster




Here is a summary of the test cases:


Test case 1: Low-end single-player 2: 64-player multi-player 3: Multi-GPU single-player CPU AMD A10-7850K (‘Kaveri’ APU), 4 cores @ 3.7 GHz AMD FX-8350, 8 cores @ 4 GHz Intel Core i7-3970x Extreme, 12 logical cores @ 3.5 GHz GPU N/A AMD Radeon 7970 3 GB 2x AMD Radeon R9 290x 4 GB Settings 720p Medium Ultra 1080p Ultra 1080p 4x MSAA OS Windows 7 64-bit Windows 8 64-bit Windows 8 64-bit Level Singapore Siege of Shanghai South China Sea DX11 avg 26.6 ms/f (37.6 fps) 18.87 ms/f (52.9 fps) 13.24 ms/f (78.4 fps) Mantle avg 23.3 ms/f (43 fps) 15.08 ms/f (66.3 fps) 8.38 ms/f (121.5 fps) Improvement 14% faster 25.1% faster 58% faster
These results are concrete examples of what we have seen in heavy scenarios. But it is important to point out that on PC there are of course many different types of CPU, GPU and OS configurations as well as the workload in the game is highly varying throughout both single-player and in multi-player, so one can’t say say a single fixed number of how much faster it is.


Fortunately there are a quite a few PC hardware review sites that are planning to do their own benchmarks of Battlefield 4 running with Mantle to compare a much wider set of PC configurations and gameplay situations, and I’m sure a lot of Battlefield players will do the same and share.
Performance tools

To simplify measuring performance in the game we’ve added a new tool to the in-game console to record frame times for later analysis. Simply run “PerfOverlay.FrameFileLogEnable 1″ to start saving frame times and “PerfOverlay.FrameFileLogEnable 0″ to stop. The resulting .csv file will be located in Documents/Battlefield 4 which can be opened & graphed by Excel or other applications for viewing.Another in-game tool that is useful to use is “Render.DrawScreenInfo 1″ that will now show additional on-screen information about your CPU & GPU config, resolution and as well as if Mantle or DirectX 11 is used for rendering.



Feedback

If you try out Mantle in BF4 we would really like to hear your feedback about it! If you have any feedback about performance or your experience in general with Mantle in Battlefield 4, please post on the Battlelog forums or in the comments section below.


If you are having any technical problems when running with Mantle, please let us know by posting on answers.ea.com and we’ll look into it. It is still possible to switch back to the existing DirectX 11 rendering.


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.


I hope you are as excited as we are about the release of the new Mantle renderer. If you have any questions, ask them below and we’ll do our best to answer them.
 

Spidre

Member
Nov 6, 2013
146
0
0
Seems like Mantle has a greater effect on high end components as opposed to low end. This is the exact opposite of what I imagined.

Bad for the average user, good for most of us ^^
 
Aug 11, 2008
10,451
642
126
Now we find?
This was known from the very first day AMD shared any information about Mantle.
Currently the API is made only for GCN.

We also knew that the it helps most when in CPU limited cases, when comparing to DX11.

The limitation of to the x series of cards at beginning was a surprise, but I'm sure we will see patch to fix this as well.


Hopefully they will release new patch in future which will give some new graphics options, increased programmability should give some interesting possibilities.

One of the primary things the developers were hyping over and over was how easy mantle was to implement. So the lack of support for all except 2xxx cards is much more than a "surprise". It is a major failing. Even after all the time and effort put into the game by AMD/Dice, the vast majority of *GCN* cards, much less any other architecture, does not work. Since mantle will inevitably compete with Dx, it is critical that it is easy and seamless to implement. Granted things will undoubtedly improve, but it is not a promising beginning.

It also shows minimal improvement with Kaveri without a discrete card, eliminating the claims of those who were expecting an APU without a discrete card to suddenly become a viable gaming platform.
 

cmdrdredd

Lifer
Dec 12, 2001
27,052
357
126
One of the primary things the developers were hyping over and over was how easy mantle was to implement. So the lack of support for all except 2xxx cards is much more than a "surprise". It is a major failing. Even after all the time and effort put into the game by AMD/Dice, the vast majority of *GCN* cards, much less any other architecture, does not work. Since mantle will inevitably compete with Dx, it is critical that it is easy and seamless to implement. Granted things will undoubtedly improve, but it is not a promising beginning.

It also shows minimal improvement with Kaveri without a discrete card, eliminating the claims of those who were expecting an APU without a discrete card to suddenly become a viable gaming platform.

I didn't even think of that angle. When you put it like that, it does seem like a bad start.

I want to see actual numbers from third party testing to put these percentages in perspective. What is the actual benefit to the min and avg fps as compared to other cards and setups.
 
Last edited:

Tmf

Senior member
Jan 15, 2014
247
1
81
Really bummed for no Tahiti support out of the gate but still good to see it in the wild.
 
Last edited:

Spjut

Senior member
Apr 9, 2011
928
149
106
One of the primary things the developers were hyping over and over was how easy mantle was to implement. So the lack of support for all except 2xxx cards is much more than a "surprise". It is a major failing. Even after all the time and effort put into the game by AMD/Dice, the vast majority of *GCN* cards, much less any other architecture, does not work. Since mantle will inevitably compete with Dx, it is critical that it is easy and seamless to implement. Granted things will undoubtedly improve, but it is not a promising beginning.

It also shows minimal improvement with Kaveri without a discrete card, eliminating the claims of those who were expecting an APU without a discrete card to suddenly become a viable gaming platform.

Well, Dice has implemented Mantle now. It's up to AMD to fix the drivers for it.
I get the impression that it's mostly AMD that's lagging behind, not the developers (Dice, Oxide)
 

selni

Senior member
Oct 24, 2013
249
0
41
Seems like Mantle has a greater effect on high end components as opposed to low end. This is the exact opposite of what I imagined.

Bad for the average user, good for most of us ^^

Too early to conclude that - crossfire apparently, but crossfire is very different to a single high end GPU, and the earlier AMD numbers showed practically no improvement in the high end gpu+cpu case. Who knows exactly what they tested though, what's going on won't be clear without a lot more data.
 

ASM-coder

Member
Jan 12, 2014
193
0
0
So the lack of support for all except 2xxx cards is much more than a "surprise". It is a major failing.
I disagree. A partial success after 3 months of work is not a major failing.
But I do think this is on AMD. You would think the basic architecture of
the R9 = R8 = R7. Something interesting going on there.

It also shows minimal improvement with Kaveri without a discrete card, eliminating the claims of those who were expecting an APU without a discrete card to suddenly become a viable gaming platform.
Again, my perception is different as is Johan's re: Test Case 1
"This area is heavy on the CPU as it is quite a lot of action going on, but the game is primarily GPU-bound both on DX and with Mantle but thanks to GPU optimizations with Mantle we get a good performance improvement.
Result: 26.6 ms/f -> 23.3 ms/f = 14% faster"

You say minimal, some of us say good.

Do you have an R8, and that is why you are so disappointed?
 

96Firebird

Diamond Member
Nov 8, 2010
5,712
316
126
Personally, after seeing the preliminary results, I am glad I didn't wait to purchase a new card. I was seriously considering the R9 290 (aftermarket cooler and at MSRP, of course) because of Mantle and BF4. It seems like my system wouldn't benefit too much from Mantle, which seems to be geared towards mismatched systems...

To me, it doesn't make sense to buy a low-end CPU and a high-end GPU. This seems to be where the biggest improvement is (until they fix CrossFire in Mantle). Assuming it is a gaming PC, most other games will be CPU-limited in DX. So why limit yourself in all these other games for the few games that have Mantle? I'd much rather have a more balanced setup.

Didn't live up to the hype IMO, at least not yet. Perhaps this will change over time.
 
Last edited:

Spjut

Senior member
Apr 9, 2011
928
149
106
To me, it doesn't make sense to buy a low-end CPU and a high-end GPU. This seems to be where the biggest improvement is (until they fix CrossFire in Mantle). Assuming it is a gaming PC, most other games will be CPU-limited in DX. So why limit yourself in all these other games for the few games that have Mantle? I'd much rather have a more balanced setup.

I agree about that, but if talking strictly about BF4, Mantle can give older systems an upgrade path. Plenty of my mates have Phenom II X4/X6 CPUs, would be neat if Mantle makes them able to fully utilize a high-end GPU
 

TerryMathews

Lifer
Oct 9, 1999
11,473
2
0
I disagree. A partial success after 3 months of work is not a major failing.
But I do think this is on AMD. You would think the basic architecture of
the R9 = R8 = R7. Something interesting going on there.


Again, my perception is different as is Johan's re: Test Case 1
"This area is heavy on the CPU as it is quite a lot of action going on, but the game is primarily GPU-bound both on DX and with Mantle but thanks to GPU optimizations with Mantle we get a good performance improvement.
Result: 26.6 ms/f -> 23.3 ms/f = 14% faster"

You say minimal, some of us say good.

Do you have an R8, and that is why you are so disappointed?

I think list of supported cards explains what's going on.

No 77xx is supported except for 7790. As I understand it, the only feature difference there is double precision.

If I had to hazard a guess, I'd say that AMD hasn't worked out a feature reporting function yet and instead is only offering Mantle on cards that support the full feature set.
 

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
Again, my perception is different as is Johan's re: Test Case 1
"This area is heavy on the CPU as it is quite a lot of action going on, but the game is primarily GPU-bound both on DX and with Mantle but thanks to GPU optimizations with Mantle we get a good performance improvement.
Result: 26.6 ms/f -> 23.3 ms/f = 14% faster"

You say minimal, some of us say good.

Do you have an R8, and that is why you are so disappointed?

Yes, "good" if you like to play in 720p with MEDIUM details.
 

ASM-coder

Member
Jan 12, 2014
193
0
0
If I had to hazard a guess, I'd say that AMD hasn't worked out a feature reporting function yet and instead is only offering Mantle on cards that support the full feature set.

Perhaps, but the R7 260 is supported. Do you think think that supports the full feature set?
 

TerryMathews

Lifer
Oct 9, 1999
11,473
2
0
Perhaps, but the R7 260 is supported. Do you think think that supports the full feature set?

Yes.

Isnt the 260 the other card that does crossfire over PCIe instead of using the bridge?

Remember, feature complete != fast. Think Haswell Celeron against i7 990x.
 

Noctifer616

Senior member
Nov 5, 2013
380
0
76
yes only R9 290 and R9 290X GPUS are supported

Wrong. Right now nothing supports Mantle cause they driver isn't out.

The public Mantle driver only supports the new AMD cards. When the support for the 7000 series is coming is not known (though developers have the driver that supports all GCN based cards).
 

GaiaHunter

Diamond Member
Jul 13, 2008
3,634
180
106
Yes, "good" if you like to play in 720p with MEDIUM details.

It is an iGPU.
It is amazing something as weak as that 512core@720MHz is actually being bottleneck by the CPU (I would wager the performance gains is mostly to memory improvements than anything else).

I guess you were expecting 50% faster?
 
Last edited:

96Firebird

Diamond Member
Nov 8, 2010
5,712
316
126
R9 290, R9 290X, R7 260X, R7 260, and 7790 are GCN 1.1 GPUs. Those are the ones that appear to be supported by Mantle. Looks like GCN 1.0 GPUs are not supported yet.
 

NTMBK

Lifer
Nov 14, 2011
10,269
5,134
136
While the averages for that Kaveri test case have changed by "only" 14%, the frametime graph looks fantastic! It's ridiculously flat now, better than any of the others. That's a console level of consistency.
 

notty22

Diamond Member
Jan 1, 2010
3,375
0
0
It's been 3 months since a announcement, not the development process.
Hi,
My name is Johan Andersson and I’m one of the Technical Directors in the Frostbite team and I have something very exciting to share with all Battlefield players: the latest Battlefield 4 update on PC includes the long-awaited new renderer in Frostbite that is using AMD’s Mantle.
Mantle is a new low-level graphics API that we’ve been working very closely with AMD on over the last 2 years and it is a major change & improvement to how we are able to program & use modern GPUs in order to get the most out of them.
quote taken from a post above.
 

krumme

Diamond Member
Oct 9, 2009
5,956
1,595
136
Yes, "good" if you like to play in 720p with MEDIUM details.

No probably very good = aprox 40% uplift in very cpu limited scenarios, when you go from sp til 64man mp in huge maps. Mantle makes it a viable option to play bf4 in a realistic mp scenario, and will make it possible to game on the kaveri with acceptable min fps. Its very important and decides playability. Its basicly playable or not playable. Talk about a 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/    |