Star Swarm Benchmark

[Cru]

well this just popped up, so post results!

http://store.steampowered.com/app/267130/

just ran the attack scenario on extreme
1080p
single gtx 780
4770k at 4.5ghz
16gb dominator gt at 2133

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
D:\Documents\Star Swarm\Benchmark_14_01_30_1940.txt
Version 0.95
01/30/2014 19:40
===========================================================

== Hardware Configuration =================================
GPU: NVIDIA GeForce GTX 780
CPU: GenuineIntel
Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 17121165312
Allocatable Memory: 8796092891136
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 4207

Average FPS: 35.05
Average Unit Count: 3799
Maximum Unit Count: 5355
Average Batches/MS: 1186.31
Maximum Batches/MS: 2180.05
Average Batch Count: 34825
Maximum Batch Count: 182546
===========================================================

 

[barnsley]

The drivers for Mantle aren't actually released right now. Yes, the benchmark is out before the software. GG AMD.

DirectX run:

== Hardware Configuration =================================
GPU: AMD Radeon R9 200 Series
CPU: AuthenticAMD
AMD FX(tm)-8320 Eight-Core Processor
Physical Cores: 4
Logical Cores: 8
Physical Memory: 8532623360
Allocatable Memory: 8796092891136
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 1996

Average FPS: 16.62
Average Unit Count: 3607
Maximum Unit Count: 5587
Average Batches/MS: 667.40
Maximum Batches/MS: 888.04
Average Batch Count: 42831
Maximum Batch Count: 103011
===========================================================

aha that average FPS. I really hope mantle is good to me as I love space RTS games

I'll post my Mantle go as well when it comes out, although its only appears to be for the 290s currently.

 

[Cru]

well it doesn't seem to like SLI

just ran it with SLI enabled and got 4fps more!

 

[Wraith]

9.74 Fps laughed so hard! I really need to consider upgrades.

== Hardware Configuration =================================
GPU: AMD Radeon HD 6700 Series
CPU: AuthenticAMD
AMD FX(tm)-4100 Quad-Core Processor
Physical Cores: 4
Logical Cores: 4
Physical Memory: 8587370496
Allocatable Memory: 8796092891136
===========================================================

== Configuration ==========================================
API: DirectX
Scenario: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================

== Results ================================================
Test Duration: 120 Seconds
Total Frames: 1170

Average FPS: 9.74
Average Unit Count: 2877
Maximum Unit Count: 4558
Average Batches/MS: 100.87
Maximum Batches/MS: 315.99
Average Batch Count: 11434
Maximum Batch Count: 55938
===========================================================

EDIT: Something isn't right, I'm getting a mixed bag of FPS scores:
Extreme 9.74
High 9.62
Medium 9.66
Low 8.28......:huh:

 

[SieB]

Seems to be quite CPU intensive and likes more cores.

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\Sie\Documents\Star Swarm\Benchmark_14_01_30_1836.txt
Version 0.95
01/30/2014 18:36
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon R9 200 Series
CPU: GenuineIntel
Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
Physical Cores: 4
Logical Cores: 4
Physical Memory: 8545779712
Allocatable Memory: 140737488224256
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 3817

Average FPS: 31.81
Average Unit Count: 3834
Maximum Unit Count: 5536
Average Batches/MS: 779.74
Maximum Batches/MS: 1156.05
Average Batch Count: 25130
Maximum Batch Count: 75969
===========================================================

 

[cooperman]

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\coop\Documents\Star Swarm\Benchmark_14_01_30_1932.txt
Version 0.95
01/30/2014 19:32
===========================================================

== 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: DirectX
Scenario: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 1964

Average FPS: 16.36
Average Unit Count: 3618
Maximum Unit Count: 5615
Average Batches/MS: 636.82
Maximum Batches/MS: 869.13
Average Batch Count: 41386
Maximum Batch Count: 144784
===========================================================

 

[Damien c]

Seems it doesn't like my pc

Run 1 before I added a profile to Nvidia Control Panel.

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\damien\Documents\Star Swarm\Benchmark_14_01_30_2000.txt
Version 0.95
01/30/2014 20:00
===========================================================

== 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: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 3205

Average FPS: 26.71
Average Unit Count: 3819
Maximum Unit Count: 5434
Average Batches/MS: 1155.66
Maximum Batches/MS: 1792.11
Average Batch Count: 44149
Maximum Batch Count: 215294
===========================================================

Run 2 with a standard profile with no changes.

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\damien\Documents\Star Swarm\Benchmark_14_01_30_2010.txt
Version 0.95
01/30/2014 20:10
===========================================================

== 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: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 3861

Average FPS: 32.16
Average Unit Count: 3797
Maximum Unit Count: 5560
Average Batches/MS: 1063.68
Maximum Batches/MS: 1687.41
Average Batch Count: 32394
Maximum Batch Count: 194195
===========================================================

Run 3 with "Alternate Frame Rendering 1" set in the control panel

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\damien\Documents\Star Swarm\Benchmark_14_01_30_2015.txt
Version 0.95
01/30/2014 20:15
===========================================================

== 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: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 1630

Average FPS: 13.58
Average Unit Count: 3282
Maximum Unit Count: 5624
Average Batches/MS: 495.61
Maximum Batches/MS: 1399.73
Average Batch Count: 52565
Maximum Batch Count: 199902
===========================================================

Run 4 with "Alternate Frame Rendering 2"

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\damien\Documents\Star Swarm\Benchmark_14_01_30_2019.txt
Version 0.95
01/30/2014 20:19
===========================================================

== 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: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 2787

Average FPS: 23.22
Average Unit Count: 3732
Maximum Unit Count: 5589
Average Batches/MS: 740.77
Maximum Batches/MS: 1552.19
Average Batch Count: 35769
Maximum Batch Count: 192511
===========================================================

It was ran with my cpu at 4.6ghz and my 780Ti's at stock.

 

[QuietOne]

Here be mine.

== Hardware Configuration =================================
GPU: NVIDIA GeForce GTX 780
CPU: GenuineIntel
Intel(R) Core(TM) i7-2700K CPU @ 3.50GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 17145413632
Allocatable Memory: 8796092891136
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 3143

Average FPS: 26.18
Average Unit Count: 3767
Maximum Unit Count: 5549
Average Batches/MS: 997.26
Maximum Batches/MS: 1635.37
Average Batch Count: 36206
Maximum Batch Count: 178529
===========================================================

 

[Cru]

so, i ran the benchmark a couple times now and it doesn't seem to be consistent at all

those are the average results i got without changing any settings:

36.32
25.15
31.18
35.31
46.43
29.46

anyone mind running the benchmark a couple times to confirm that?

 

[chrismjurd]

== Hardware Configuration =================================
GPU: NVIDIA GeForce GTX 660
CPU: GenuineIntel
Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 8551895040
Allocatable Memory: 140737488224256
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 3960

Average FPS: 33.00
Average Unit Count: 3788
Maximum Unit Count: 5467
Average Batches/MS: 569.24
Maximum Batches/MS: 1215.71
Average Batch Count: 19110
Maximum Batch Count: 68511
===========================================================

ran it 5 times so far
33.00
22.01
22.06
27.98
25.51

the bench is not the same preconfigured run, its dynamic so variance will be quite large, and quite worthless.

 

[Davva2004]

Are you ready for this jelly?

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
D:\Dave\Documents\Star Swarm\Benchmark_14_01_30_2057.txt
Version 0.95
01/30/2014 20:57
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon HD 5800 Series
CPU: AuthenticAMD
AMD Phenom(tm) II X4 955 Processor
Physical Cores: 4
Logical Cores: 4
Physical Memory: 4293373952
Allocatable Memory: 8796092891136
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioAttract.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
Motion Blur Frame Time: 16
Motion Blur InterFrame Time: 2
Detailed Frame Info: Off
===========================================================


== Results ================================================
Test Duration: 120 Seconds
Total Frames: 1119

Average FPS: 9.32
Average Unit Count: 2813
Maximum Unit Count: 4559
Average Batches/MS: 110.14
Maximum Batches/MS: 312.72
Average Batch Count: 11827
Maximum Batch Count: 38619
===========================================================

FMMFL.

 

[Cru]

the bench is not the same preconfigured run, its dynamic so variance will be quite large, and quite worthless.

why would you even call something thats not repeatable a benchmark...

 

[chrismjurd]

dont know enough about it, just wierded me out when i ran it the second time and i was like did i blink and miss this cam view last time or something.

quick vid for those without it to take a peep, btw shadowplay didnt effect the variance, from what i can see.

 

[SieB]

I've just ran it three times and it is dynamic, it was different each time.

I've noticed it's also a lot more CPU intensive than it is GPU intensive.
If you have the MSI overlay on so you can see your GPU load and CPU load, during the intense parts where FPS drops to like 6-7, the GPU usage is around below and up to 50% while the CPU is around 80-90%

This is a benchmark designed to showcase Mantle though and what Mantle does is offload the work done by the CPU onto the GPU. So this benchmark is actually good for showing how well Mantle offloads the load on the CPU on to the GPU.

It is still dynamic though, so even when running under Mantle you are still going to get varied results.

 

[Cru]

It is still dynamic though, so even when running under Mantle you are still going to get varied results.

Which destroys the whole point
it's designed to show the increase with mantle and it can't even do that right

 

[SieB]

Which destroys the whole point
it's designed to show the increase with mantle and it can't even do that right

I agree, although the results will vary, you will still be able to see how well Mantle does it's job. But as far as using this benchmark to test and compare performance between yours, mine and other peoples rig goes, it's not accurate enough and shouldn't be used to gage performance of your rig vs another.

 

[barnsley]

Which destroys the whole point

I'm pretty sure the custom one is the same each time. Regardless I do hope their is an actual benchmark part of the program as at the moment its more of a "look how bad your average DirectX game is LOL XDDDDDD" as the cpu usage on my machine is as bad as DayZ and thats a game that doesn't do cpu optimisation. Battlefield is a much better benchmark (when its fully implimented) as that shows what a mostly talented bunch of programmers can do with DX.

 

[Cru]

I agree, although the results will vary you will still be able to see how well Mantle does it's job.

not really imo
as you just seem to get a random run the first bench you do without mantle can average at 20fps and the next one, in which you'd get 40fps without mantle, averages 55fps.
the performance increase would then be calculated to 275%(?) instead of 137.5%(?)

numbers might be wrong, bit tired

 

[SieB]

not really imo
as you just seem to get a random run the first bench you do without mantle can average at 20fps and the next one, in which you'd get 40fps without mantle, averages 55fps.
the performance increase would then be calculated to 275%(?) instead of 137.5%(?)

numbers might be wrong, bit tired

I think the only real way of telling will be in the average FPS. Those parts where it gets really intense and where the FPS drops is where Mantle will offload from the CPU onto the GPU. So although the FPS will be slightly different each time it should still be higher under Mantle than it will be under DX. You still can't use it to accurately judge performance though due to it being different each time.

I'm not sure if the other benchmarks are dynamic also, going to check them now.

 

[WYP]

I've just ran it three times and it is dynamic, it was different each time.

I've noticed it's also a lot more CPU intensive than it is GPU intensive.
If you have the MSI overlay on so you can see your GPU load and CPU load, during the intense parts where FPS drops to like 6-7, the GPU usage is around below and up to 50% while the CPU is around 80-90%

This is a benchmark designed to showcase Mantle though and what Mantle does is offload the work done by the CPU onto the GPU. So this benchmark is actually good for showing how well Mantle offloads the load on the CPU on to the GPU.

It is still dynamic though, so even when running under Mantle you are still going to get varied results.

Mantle does not offload cpu tasks to the gpu, it takes just takes away the driver overhead off the CPU and enables the cpu to talk to the gpu more openly, allowing a boat load more draw calls and allows the cpu to essentially do more with less.

Star swarm is a draw call heavy scenario with all the misc calculations (pathfinding, AI etc) and the power required to feed all the gpu tasks is simply too much for most cpus to handle.

though there is a possibility of using an APU (or rather the gpu in an APU) as a co-processor for physics or other tasks. It is described here on hardwareluxx.

I guess the only way to use this test as any indication of relative performance will be to use a large number of sequential runs. Can't wait for catalyst 14.1 to be released and for myself to get back to belfast so i may benchmark