Star Swarm Benchmark

[cooperman]

Here is anuther mantle run becase i am a dumbass and forgot to add my overclocks to the card after updateing to the new driver lol (1100mhz core 1500mhz mem)

==========================================================
Oxide Games
Star Swarm Stress Test - ©2013
C:\Users\coop\Documents\Star Swarm\Output_14_02_02_1413.txt
Version 1.00
02/02/2014 14:13
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon HD 7900 Series
CPU: AuthenticAMD
AMD FX(tm)-8320 Eight-Core Processor
Physical Cores: 4
Logical Cores: 8
Physical Memory: 8502214656
Allocatable Memory: 8796092891136
===========================================================


== 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: 17487

Average FPS: 48.57
Average Unit Count: 4320
Maximum Unit Count: 5608
Average Batches/MS: 703.63
Maximum Batches/MS: 1886.56
Average Batch Count: 16999
Maximum Batch Count: 156146
===========================================================

 

[Damien c]

Right done another run with it after abit of messing around after, I found that one of my cards was only running at 8x and not 16x also added the PCI-E 3.0 patch.

===========================================================
Oxide Games
Star Swarm Stress Test - ©2013
C:\Users\damien\Documents\Star Swarm\Output_14_02_02_1429.txt
Version 1.00
02/02/2014 14:29
===========================================================

== Hardware Configuration =================================
GPU: NVIDIA GeForce GTX 780 Ti
CPU: GenuineIntel
Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz
Physical Cores: 6
Logical Cores: 12
Physical Memory: 17117908992
Allocatable Memory: 140737488224256
===========================================================


== 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: 16409

Average FPS: 45.58
Average Unit Count: 4262
Maximum Unit Count: 5445
Average Batches/MS: 727.74
Maximum Batches/MS: 1806.76
Average Batch Count: 16568
Maximum Batch Count: 110385
===========================================================

Max CPU usage was: 48.1%
Max GPU usage was: 46.7% + 8.5%

This cannot be called a benchmark if it does not use all the resources available to it in my opinion.

 

[barnsley]

Max CPU usage was: 48.1%
Max GPU usage was: 46.7% + 8.5%

This cannot be called a benchmark if it does not use all the resources available to it in my opinion.

The CPU usage on the direct X does not use all of the CPU cores currently. They said they'll patch it in. The DX 11 version needs work



I'm currently sorting out drivers so that I can get a recorded run on DX11 then on Mantle. I'll edit and upload the video when I'm done

 

[Damien c]

Will be interesting to see that video.

 

[barnsley]

Sadly mantle isn't being picked up by my recording software (playclaw, I use it because it doesn't hog frames like fraps).
Anyway, bench time.
settings:

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

DX11 score:

== Results ================================================
Test Duration: 360 Seconds
Total Frames: 8522

Average FPS: 23.67
Average Unit Count: 3949
Maximum Unit Count: 5437
Average Batches/MS: 362.71
Maximum Batches/MS: 771.24
Average Batch Count: 19005
Maximum Batch Count: 125632
===========================================================

Mantle score:

== Results ================================================
Test Duration: 360 Seconds
Total Frames: 15778

Average FPS: 43.83
Average Unit Count: 4330
Maximum Unit Count: 5906
Average Batches/MS: 681.51
Maximum Batches/MS: 1527.98
Average Batch Count: 17749
Maximum Batch Count: 148791
===========================================================

I'll try and play with playclaw to get it to record the mantle run. I chose the scenario because its supposedly the most accurate.

 

[SieB]

I've done 5 runs on DX11 and 5 runs on Mantle. My average FPS under DX11 was between 26-31 FPS for all 5 runs.
My Average FPS during 5 runs under Mantle was between 57-61 FPS.

That's a massive "upto" 100% increase

 

[CalVic]

Very nice!

DX11: Average FPS: 29.18

Mantle: Average FPS: 50.33

 

[vorticalbox]

i3 2100, 7770

DX

== Results ================================================
Test Duration: 360 Seconds
Total Frames: 6606

Average FPS: 18.35
Average Unit Count: 4022
Maximum Unit Count: 5508
Average Batches/MS: 335.92
Maximum Batches/MS: 871.16
Average Batch Count: 20302
Maximum Batch Count: 104388
=========================================================


Mantle

== Results ================================================
Test Duration: 360 Seconds
Total Frames: 3536

Average FPS: 9.82
Average Unit Count: 3957
Maximum Unit Count: 5715
Average Batches/MS: 249.14
Maximum Batches/MS: 1209.80
Average Batch Count: 27226
Maximum Batch Count: 164211
=========================================================

got -50% on mantle....


Edit: minecraft didn't work so i used this Fix this also increased my mantle scores to 23 FPS average so a slight improvement.

 

[WYP]

Did a quick test on the star swarm demo, impressive gains for the first runs.

mantle avg FPS - 40.10

DX11 avg FPS - 20.32

That's near 100% gains on my 1090t , i'll need to do a few more runs and take an average but wow, that's an awesome gain there.

 

[Jinxxter]

Still can't get it to work even with the 14.1 Betas...
Did a clean uninstall with ddu first and reloaded star swarm to make sure my installation wasn't faulty...
In D3D-mode it's the same as in my previous post and using mantle it crashes with n error-message from windows...
Seems like I need to get my Asrock X79 Extreme6 @3930K back up and running...

 

[dladdin]

benchmark for star swarm

• CPU amd fx 4100 fx @ 4.2ghz
• Motherboard asus m5a78l-lx
• RAM 8gb 1333mhz
• GPU asus hd 7770 2gb oc
• Cooling corsair h80

direct x..................... mantle

(settings) ................ (settings)

low = 53.2................ low= 67.9

med = 23.2............... med=34.1

high = 22.9 .............. high=26.8

extreme = 17.03........ extreme=24.41


Ran for the 6 mins on follow.

thanks

dladdin
_________________

 

[WYP]

OK just did a few re-runs and taken an average over 6 runs.

DirectX - 21.44FPS

Mantle - 41.31FPS

This is on a 1090t @ 4GHz using the follow scenario

92.7% gains here for average framerates

I will likely try the other scenarios later.

 

[dladdin]

Sweet can't wait to get my new rig this year

 

[Wraith]

Well I've spent most of the day tinkering with the custom.ini mixing Ultra, High, Med and Low settings and getting really dreadful fps scores.... also something rather interesting CPU & GPU load won't go any higher than 55% while running the 6 minute tests... anyone else notice that it's low load.

 

[barnsley]

Well I've spent most of the day tinkering with the custom.ini mixing Ultra, High, Med and Low settings and getting really dreadful fps scores.... also something rather interesting CPU & GPU load won't go any higher than 55% while running the 6 minute tests... anyone else notice that it's low load.

Yep, I've noticed the DX11 one is especially bad for load. It hovers around 40% cpu 55% gpu. Core usage appears to only be on random cores as well so I'd imagine its got a long way to go.

 

[Wraith]

Yep, I've noticed the DX11 one is especially bad for load. It hovers around 40% cpu 55% gpu. Core usage appears to only be on random cores as well so I'd imagine its got a long way to go.

I kind of figured something wasn't right when my fans didn't speed up, going to leave it for now and see what comes later.

 

[barnsley]

I kind of figured something wasn't right when my fans didn't speed up, going to leave it for now and see what comes later.

I've reverted back to stable drivers. the beta was being a bit odd and for some reason it was making flash player use quite a bit of cpu . Best give it a month or so I guess.

 

[RadeonHDx]

===========================================================
Oxide Games
Star Swarm Stress Test - ©2013
C:\Users\Joe\Documents\Star Swarm\Output_14_02_02_2152.txt
Version 1.00
02/02/2014 21:52
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon HD 7800 Series
CPU: AuthenticAMD
AMD FX(tm)-8320 Eight-Core Processor
Physical Cores: 4
Logical Cores: 8
Physical Memory: 17081880576
Allocatable Memory: 8796092891136
===========================================================


== 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: 8765

Average FPS: 24.35
Average Unit Count: 4005
Maximum Unit Count: 5555
Average Batches/MS: 489.95
Maximum Batches/MS: 941.85
Average Batch Count: 21907
Maximum Batch Count: 106189
===========================================================

===========================================================
Oxide Games
Star Swarm Stress Test - ©2013
C:\Users\Joe\Documents\Star Swarm\Output_14_02_02_2200.txt
Version 1.00
02/02/2014 22:00
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon HD 7800 Series
CPU: AuthenticAMD
AMD FX(tm)-8320 Eight-Core Processor
Physical Cores: 4
Logical Cores: 8
Physical Memory: 17081880576
Allocatable Memory: 8796092891136
===========================================================


== 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: 14519

Average FPS: 40.33
Average Unit Count: 4274
Maximum Unit Count: 5539
Average Batches/MS: 730.79
Maximum Batches/MS: 1957.43
Average Batch Count: 19704
Maximum Batch Count: 110163
===========================================================

TBH since my CPU certainly isn't a bottleneck IMO, I'm very pleased with the results.

 

[Rogster]

===========================================================
Oxide Games
Star Swarm Stress Test - ©2013
C:\Users\Fred\Documents\Star Swarm\Output_14_02_03_0353.txt
Version 1.00
02/03/2014 03:53
===========================================================
== Hardware Configuration =================================
GPU: AMD Radeon R9 200 Series
CPU: AuthenticAMD
AMD FX(tm)-8350 Eight-Core Processor
Physical Cores: 4
Logical Cores: 8
Physical Memory: 8481812480
Allocatable Memory: 8796092891136
===========================================================

== 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: 18745
Average FPS: 52.07
Average Unit Count: 4697
Maximum Unit Count: 6164
Average Batches/MS: 805.63
Maximum Batches/MS: 1797.99
Average Batch Count: 18440
Maximum Batch Count: 88592
===========================================================

My results

 

[WYP]

Update on star swarm form oxide games (found here)

We just deployed a new build of our Star Swarm stress test that significantly improves the demo’s performance using AMD’s Mantle API. We suggest that those of you interested in benchmarking and performance numbers re-run your Mantle scenarios; you may be surprised at the results.

Did we crack some secret code or find a crazy new optimization? No, nothing so spectacular. The truth is that we made a mistake in our haste to deploy the build that stripped out the activation process. We didn’t follow our normal release process, and missed how a minor change in that build had disabled some of the Nitrous engine’s multi-threading features when using Mantle. Unfortunately we didn’t notice that at first, as nobody was running Mantle last week due to the beta driver being delayed.

It’s all fixed now, so Mantle users should see a noticeable boost in performance on most configurations (this fix doesn’t have a huge effect on powerful-CPU/slow-GPU systems). We’ve also addressed the “gray screen” issue that was affecting some non-English Windows systems.

We appreciate everyone’s patience and feedback as we hunt down some of these hard-to-reach bugs for this alpha test of the Nitrous Engine.

Happy blasting! For more info, check out our Star Swarm FAQ, the official forums, and the Oxide Games Twitter page.