Mantle Benchmark Thread

Diagro · 3 Feb 2014 at 18:30

What I like about mantel Is that i can run my cards at 800/1200 and get same performance from bf4 as I would at 1100/1400 on DX11!

Makes system run quieter and cooler.

Bargin

will bench this later

LtMatt · 3 Feb 2014 at 18:32

Diagro said:
What I like about mantel Is that i can run my cards at 800/1200 and get same performance from bf4 as I would at 1100/1400 on DX11!

Makes system run quieter and cooler.

Bargin

will bench this later

A part of Rusty dies every time you call it Mantel.

LtMatt · 3 Feb 2014 at 19:06

EL_JOCK said:
Just done the mantle run twice to check back the results (yep they are good) and they where within a 0.1 of each other on both runs, so happy its within error margin. Also re ran the DX11 numbers to get a good error margin.

This was done by restarting the level once it loaded to avoid the cut scene & then turning on the bench data right at the level start so no need to cut data out to allow for the load...then turning off once i got to the boat as per the video LtMatt put up, also double checked crossfire was working for both dx11 & mantle

Ultra preset 4xaa @1080p motion blur off FOV @90
i5 3570k @ 4.2 (h100i) water cooling
crossfire Tri-x 290's (stock 1000)

Mantle
Max 199
Avg 129.6
min 20.7

DX11
Max 122.4
Avg 66.2
Min 25.3

That's 95.7% improvement on the avg

Are you 100% sure that's accurate?

It looks like crossfire may not have been working on the DX11 run. I scored higher than that at 1440P at the same settings with 290 Pro crossfire.

Scooby-DoobyDoo · 3 Feb 2014 at 19:16

I've rolled back totally trashed my framerates in Payday and Payday 2!

BF4 15% more with mantle. But terrible for other games.

SieB · 3 Feb 2014 at 19:25

[email protected] & Sapphire 290 Tri-X 1000/1300

1920x1080

DX11

MIN:25
MAX:98
AVG:65

Mantle

MIN:25
MAX:103
AVG:73

LtMatt · 3 Feb 2014 at 19:28

SieB said:
[email protected] & Sapphire 290 Tri-X 1000/1300

DX11

MIN:25
MAX:98
AVG:65

Mantle

MIN:25
MAX:103
AVG:73

What resolution please?

SieB · 3 Feb 2014 at 19:31

LtMatt said:
What resolution please?

Forgot to add it, was at 1080p. Edited it in my post above

snips86x · 3 Feb 2014 at 19:32

Disable the IGP on my 2500k and ran the DX & Mantle test.

Results = Mantle:

== Configuration ==========================================
API: Mantle
Scenario: ScenarioFollow.csv
User Input: Disabled
Resolution: 1920x1080
Fullscreen: True
GameCore Update: 16.6 ms
Bloom Quality: High
PointLight Quality: High
ToneCurve Quality: High
Glare Overdraw: 16
Shading Samples: 64
Shade Quality: Mid
Deferred Contexts: Disabled
Temporal AA Duration: 16
Temporal AA Time Slice: 2
Detailed Frame Info: Off
================================================== =========
== Results ================================================
Test Duration: 360 Seconds
Total Frames: 19074
Average FPS: 52.98
Average Unit Count: 4687
Maximum Unit Count: 5739
Average Batches/MS: 878.44
Maximum Batches/MS: 2596.75
Average Batch Count: 18767
Maximum Batch Count: 134425
================================================== =========

Results = DX:

== Configuration ==========================================
API: DirectX
Scenario: ScenarioFollow.csv
User Input: Disabled
Resolution: 1920x1080
Fullscreen: True
GameCore Update: 16.6 ms
Bloom Quality: High
PointLight Quality: High
ToneCurve Quality: High
Glare Overdraw: 16
Shading Samples: 64
Shade Quality: Mid
Deferred Contexts: Disabled
Temporal AA Duration: 16
Temporal AA Time Slice: 2
Detailed Frame Info: Off
================================================== =========
== Results ================================================
Test Duration: 360 Seconds
Total Frames: 10720
Average FPS: 29.77
Average Unit Count: 4104
Maximum Unit Count: 5427
Average Batches/MS: 582.68
Maximum Batches/MS: 1407.05
Average Batch Count: 22335
Maximum Batch Count: 135304
================================================== =========

Looks to be promising!! However, the new AMD drivers are obviously biased to mantle. When I ran the DX test on the Star Swarn using the 31.11 drivers, the DX results were far higher.

LtMatt · 3 Feb 2014 at 19:38

SieB said:
Forgot to add it, was at 1080p. Edited it in my post above

Scores added.

Guys im open to suggestions if you can think of a better way for me to layout the scores.

EL_JOCK · 3 Feb 2014 at 19:40

LtMatt said:
Are you 100% sure that's accurate?

It looks like crossfire may not have been working on the DX11 run. I scored higher than that at 1440P at the same settings with 290 Pro crossfire.

hi matt,

Crossfire was defo working had the AB overlay on for the DX run, however there was a problem with the sheet i was using before the app so the Avg for mantle was well off

Did a few re-runs with the app on mantle to take human error out, think i had a problem in my setup on the excel sheet,

in short the avg should have been about 76fps on mantle so about a 30% uplift for me vs DX so still happy with that,

I will try to get some screen shots done ASAP, however like a tool i hit restart lvl not checkpoint.

Cambofrog · 3 Feb 2014 at 19:47

LtMatt said:
Scores added.

Guys im open to suggestions if you can think of a better way for me to layout the scores.

Dunno, maybe put the DX and Mantle scores together in a single post so it's easier to compare DX v Mantle.

LtMatt · 3 Feb 2014 at 19:52

EL_JOCK said:
hi matt,

Crossfire was defo working had the AB overlay on for the DX run, however there was a problem with the sheet i was using before the app so the Avg for mantle was well off

Did a few re-runs with the app on mantle to take human error out, think i had a problem in my setup on the excel sheet,

in short the avg should have been about 76fps on mantle so about a 30% uplift for me vs DX so still happy with that,

I will try to get some screen shots done ASAP, however like a tool i hit restart lvl not checkpoint.

Nice one.

And Lol at the schoolboy error, thats going to cost you. Now you have to sit through all the tedious crap that goes on before that sequence lol.

Cambofrog said:
Dunno, maybe put the DX and Mantle scores together in a single post so it's easier to compare DX v Mantle.

Could do that. Need to make it stand out a bit ideally.

uksoldierboy · 3 Feb 2014 at 19:53

Going to run the bench now, i'll report back with the differences asap.

sockky · 3 Feb 2014 at 19:55

How's everyone running these benchmarks? I usually use fraps but I am having issues with BF4 crashing every time I start the program.

LtMatt · 3 Feb 2014 at 19:56

sockky said:
How's everyone running these benchmarks? I usually use fraps but I am having issues with BF4 crashing every time I start the program.

Your lack of reading the OP and following posts brings shame upon us all.

sockky · 3 Feb 2014 at 20:00

Okay ... Let's all ignore that and love me when I return with some results.

LtMatt · 3 Feb 2014 at 20:01

sockky said:
Okay ... Let's all ignore that and love me when I return with some results.

Atta boy.

Cambofrog · 3 Feb 2014 at 20:03

sockky said:
How's everyone running these benchmarks? I usually use fraps but I am having issues with BF4 crashing every time I start the program.

Not sure about the crashing, but I read people stating elsewhere, that fraps doesn't work with Mantle.

I haven't tried it myself though ( I don't have BF4 or an AMD card ).

LtMatt · 3 Feb 2014 at 20:09

Cambofrog said:
Not sure about the crashing, but I read people stating elsewhere, that fraps doesn't work with Mantle.

I haven't tried it myself though ( I don't have BF4 or an AMD card ).

Afterburner OSD and fraps both make mantle/bf4 crash for me.

snips86x · 3 Feb 2014 at 20:14

LtMatt said:
Afterburner OSD and fraps both make mantle/bf4 crash for me.

I was advised to disable IGP, worked like a charm.