DICE Posts its Own Battlefield 4 DirectX vs. Mantle Performance Numbers

[btarunr]

Along with its highly anticipated game patch that includes an AMD Mantle renderer for Battlefield 4, DICE posted numbers from its own testing, pointing out the performance difference between DirectX 11.1 and Mantle. DICE put Battlefield 4 through three test scenarios, entry-level gaming, mainstream gaming, and enthusiast gaming. The entry-level test-bed comprised of an AMD A10-7850K APU, with its integrated Radeon R7 200 series GPU (512 stream processors, 720 MHz GPU clock). This is a CPU and GPU limited scenario, in which the game was tested at 1280 x 720 pixels resolution. DICE notes that with Mantle, the game yielded about 14 percent higher frame-rates.

Next up, is mainstream gaming. The test-bed runs an AMD FX-8350, which offers roughly the same gaming CPU performance as a Core i5-3570K. A Radeon HD 7970 is in charge of graphics, and the game is run at 1920 x 1080 pixels resolution, with 1x MSAA and "Ultra" preset. DICE found that the setup yields about 26 percent higher frame-rates. Lastly, there's the enthusiast test-bed, running an Intel Core i7-3960X CPU, and dual Radeon R9 290X (CrossFire) graphics. The resolution stayed at 1920 x 1080, settings at "Ultra" preset, but the anti-aliasing was cranked up to 4x MSAA. The result? A stunning 58 percent higher frame-rates. It's important to note here that in addition to settings, the other thing that's not constant between the three setups is the test scene. Even if DICE' assessment is most generous towards AMD's claims, there really does seem to be a performance increment on offer, with Mantle. Can't wait to check it out for ourselves. For more details and notes from the developer, check out the source link.



View at TechPowerUp Main Site

 

[Supercrit]

I 'm kind of disappointed at the lower end system improvements. The whopping 58% higher performance is where it doesn't matter, not to mention the test is not done by a neutral 3rd party. At least it seems that the stuttering problem is gone.

 

[Jack1n]

Not bad.

 

[Fluffmeister]

Battlefield 4 + Mantle

• AMD FX-8350 + R9 290X +23% Performance increase Vs. Direct X
• Core i5-4670k + R9 290X +7.5% Performance increase Vs. Driect X

• Core i7-4960X CPU + R9 290X GPU
  • 1080p, Ultra Preset, 4xAA: 9.2% improvement with Mantle
  • 1600p, Ultra Preset, 4xAA: 10% improvement with Mantle
• Core i7-4960X CPU + R7 260X GPU
  • 1080p, Ultra Preset, 4xAA: 2.7% improvement
  • 1600p, Ultra Preset, 4xAA: 1.4% improvement
• A10-7700K CPU + R9 290X GPU
  • 1080p, Ultra Preset, 4xAA: 40.9% improvement
  • 1600p, Ultra Preset, 4xAA: 17.3% improvement
• A10-7700K CPU + R7 260X GPU
  • 1080p, Ultra Preset, 4xAA: 8.3% improvement
  • 1600p, Low Preset: 16.8% improvement

Source: http://wccftech.com/amd-launches-mantle-beta-low-end-cpus-not-gpus/

 

[john_]

I 'm kind of disappointed at the lower end system improvements. The whopping 58% higher performance is where it doesn't matter, not to mention the test is not done by a neutral 3rd party. At least it seems that the stuttering problem is gone.

From AMD's results you get a 40% with an APU and an 290X which is the most important scenario today that most people prefer to spend money on a new smartphone than buying a new cpu and motherboard. I expect many who stayed with older cpus (phenom, older i5/i3, core2quad) or build recently systems with sub $150 cpus(Richland/AM3+ quad-six cores/Pentium/i3) and where thinking a full system upgrade in the near future to improve gaming experience by removing cpu bottlenecks, instead of changing their cpu+motherboard alongside the gpu upgrade, to just wait and see if the cpu+motherboard upgrade is really necessary with Mantle. Any positive news and broad adoption of Mantle will drive these people to just buy the fastest GCN card they can afford instead of spending money on a cpu+motherboard upgrade AND also buy a new graphics card. They will either buy a better card or save over $100-$200.

 

[RCoon]

The low end is pretty much where it's supposed to matter though... I mean for all us high end users a few extra percent is nice, but the difference between 54 FPS and 57 FPS minimum doesn't really have much of an effect on me. But for a low end user, it can mean a great deal more.

 

[Mathragh]

Since multiplayer is usually a lot more taxing for the CPU, the performance uplift in that scenario should also be significant. Lets hope benchmarks done by review sites include those results aswell

 

[Lionheart]

Battlefield 4 + Mantle

• AMD FX-8350 + R9 290X +23% Performance increase Vs. Direct X
• Core i5-4670k + R9 290X +7.5% Performance increase Vs. Driect X

• Core i7-4960X CPU + R9 290X GPU
  • 1080p, Ultra Preset, 4xAA: 9.2% improvement with Mantle
  • 1600p, Ultra Preset, 4xAA: 10% improvement with Mantle
• Core i7-4960X CPU + R7 260X GPU
  • 1080p, Ultra Preset, 4xAA: 2.7% improvement
  • 1600p, Ultra Preset, 4xAA: 1.4% improvement
• A10-7700K CPU + R9 290X GPU
  • 1080p, Ultra Preset, 4xAA: 40.9% improvement
  • 1600p, Ultra Preset, 4xAA: 17.3% improvement
• A10-7700K CPU + R7 260X GPU
  • 1080p, Ultra Preset, 4xAA: 8.3% improvement
  • 1600p, Low Preset: 16.8% improvement

Source: http://wccftech.com/amd-launches-mantle-beta-low-end-cpus-not-gpus/

Wow! Those results make me want to get an APU now

 

[pidgin]

hoped a bit more for 7850K

 

[jigar2speed]

These are not bad results

 

[uuuaaaaaa]

One thing is obvious, for the same "performance hit" you can do much more with the same CPU without any bottleneck, which is clearly shown by the Star Swarm demo. My point is that probably BF4 is not heavy enough on the CPU side (high end CPU's in single player mode) to make a huge difference. This should be tested in multiplayer in a crowded server and I bet the gains will be greater.

 

[Mathragh]

One thing is obvious, for the same "performance hit" you can do much more with the same CPU without any bottleneck, which is clearly shown by the Star Swarm demo. My point is that probably BF4 is not heavy enough on the CPU side (high end CPU's in single player mode) to make a huge difference. This should be tested in multiplayer in a crowded server and I bet the gains will be greater.

Yeah, when you think about it, BF4 isn't really an optimal game to showcase mantle on when it comes to performance, since it was already quite CPU-agnostic compared to lots of other games. Would've probably been a lot better for something like the newest Total War game, or Civilization V.

 

[derwin75]

Wow!! Top pictures look better than the bottom pictures.

 

[Mathragh]

Wow!! Top pictures look better than the bottom pictures.

Indeed! the bottom ones look more foggy mostly. I sincerely hope they didn't resort to shortening the view distance(or something like it) in order to reach the performance improvements.

 

[TheMailMan78]

Indeed! the bottom ones look more foggy mostly. I sincerely hope they didn't resort to shortening the view distance(or something like it) in order to reach the performance improvements.

That would be funny if all Mantle did was turn all your games in Turok 64. EPIC FOG FTW!

As a matter of fact that worked for all N64 games.

 

[Prima.Vera]

"AMD FX-8350, which offers roughly the same gaming CPU performance as a Core i5-3570K"
Wait... What??

 

[Mathragh]

"AMD FX-8350, which offers roughly the same gaming CPU performance as a Core i5-3570K"
Wait... What??

http://www.techspot.com/review/734-battlefield-4-benchmarks/page6.html

 

[RCoon]

"AMD FX-8350, which offers roughly the same gaming CPU performance as a Core i5-3570K"
Wait... What??

In reality it does. I owned both processors at the same time, both stock and overclocked were pretty much identical.

 

[v12dock]

This is horrible I can't believe a beta driver with a beta API on a game with a beta patch only gives free performance improvements...

 

[tokyoduong]

Indeed! the bottom ones look more foggy mostly. I sincerely hope they didn't resort to shortening the view distance(or something like it) in order to reach the performance improvements.

It looks like there's just more fog effects in that picture. Usually when they use fog to limit draw distance, they would thicken the fog enough so they don't have to draw those far off buildings at all.

What's interesting is the graph.

 

[Ed_1]

In first shot mantle is on bottom and slower one (you can read the rendering in text at top which is which ) .
on pic 2 and 3 i can't make out but think Mantle is on bottom of all .

 

[Prima.Vera]

http://www.techspot.com/review/734-battlefield-4-benchmarks/page6.html

All I can see is some CPU/GPU bottleneck at 96-97fps. That's all.

 

[Ed_1]

It looks like there's just more fog effects in that picture. Usually when they use fog to limit draw distance, they would thicken the fog enough so they don't have to draw those far off buildings at all.

What's interesting is the graph.

Yes ,much less variance but the patch today has some optimizations for GPU/CPU too , across all hardware .

 

[tokyoduong]

All I can see is some CPU/GPU bottleneck at 96-97fps. That's all.

All I saw was GPU bottleneck. An increase in 1ghz for the 4770k yielded in 2 fps or under 2%.

 

[Slomo4shO]

So it delivers exactly what was claimed: it reduces CPU bottlenecks and enables higher utilization of the GPU...

From Johan Andersson's Mantle blog:

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.

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

Also,

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.