Star Swarm - Oxide bench thread

[TOL247]

crossfire enabled 280X and 7970

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

== Hardware Configuration =================================
GPU: AMD Radeon HD 7900 Series
CPU: GenuineIntel
Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 8555474944
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: 2290

Average FPS: 19.07
Average Unit Count: 3703
Maximum Unit Count: 5460
Average Batches/MS: 678.76
Maximum Batches/MS: 836.99
Average Batch Count: 36430
Maximum Batch Count: 138277
===========================================================

crossfire disabled 280X

Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\craig\Documents\Star Swarm\Benchmark_14_01_30_1944.txt
Version 0.95
01/30/2014 19:44
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon HD 7900 Series
CPU: GenuineIntel
Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 8555474944
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: 3772

Average FPS: 31.43
Average Unit Count: 3699
Maximum Unit Count: 5442
Average Batches/MS: 961.64
Maximum Batches/MS: 1383.64
Average Batch Count: 32405
Maximum Batch Count: 98531
===========================================================

Doesn't seem to like crossfire very much.

 

[Avenged7Fold]

Funny it will be used as a Mantle vs DX11 benchmark, when it isnt consistent enough to compare Dx11 vs Dx11 with the same components.

Hopefully they fix the amount of units and their actions, or at least devise some sort of point score.

 

[Kaapstad]

Funny it will be used as a Mantle vs DX11 benchmark, when it isnt consistent enough to compare Dx11 vs Dx11 with the same components.

Hopefully they fix the amount of units and their actions, or at least devise some sort of point score.

CF support needs work.

 

[pgi947]

Well, I'm calling BS on this bench.

GPU utilisation is around 30%, my scores line up with both sides of the fence so presume both AMD and NV are getting poor usage on the DX bench. CPU usage is sitting around 40% too.

Whats the odds that when mantle is released it magically hits 99% GPU usage showing huge gains in the process?

 

[4737Joker]

this benchmark is a waste of time looking at the random results.

I hope it doesn't get used for mantle comparisons as people will just manipulate the results.

edit: saying that if you look through the results the people with 2600/2700/4770 seems to getting higher scores than 3570/4670 users. My gpu usage is 64% at best so it weould be interesting to see if the mantle bench makes it hit 100%

But i woulld not be surprised if it has been engineered that way

 

[ICDP]

Well, I'm calling BS on this bench.

GPU utilisation is around 30%, my scores line up with both sides of the fence so presume both AMD and NV are getting poor usage on the DX bench. CPU usage is sitting around 40% too.

Whats the odds that when mantle is released it magically hits 99% GPU usage showing huge gains in the process?

I'm calling BS on your BS calling

 

[layte]

Well, I'm calling BS on this bench.

GPU utilisation is around 30%, my scores line up with both sides of the fence so presume both AMD and NV are getting poor usage on the DX bench. CPU usage is sitting around 40% too.

Whats the odds that when mantle is released it magically hits 99% GPU usage showing huge gains in the process?

Didn't one of the Oxide devs let slip a little while ago just how hilariously un optimised the DX part of the engine was?

 

[ICDP]

No, he stated DX is horribly unoptimised. Not even remotely the same as saying our game is horribly unoptimised in DX.

 

[SimBy]

Didn't one of the Oxide devs let slip a little while ago just how hilariously un optimised the DX part of the engine was?

Au contraire mon capitan. This engine is one of the best optimized engines for DX. Their words.

Mantle was tacked on later.

 

[Sgt_Pepper]

Didn't one of the Oxide devs let slip a little while ago just how hilariously un optimised the DX part of the engine was?

Probably purposefully, to make mantle seem better than it really is.

It must be lovely and snuggly in AMD's pocket.

 

[triss]

Ran it 5 times so far have a 15fps difference between tests imo needs stabilized somehow before its a efficient benchmark

 

[GoogalyMoogaly]

Have people tried all of the scenarios to see if any are more consistent than others?

 

[Gregster]

I have to be honest guys, in all the tests I did, i have seen some massive ups and downs and nothing even remotely consistent. It will be hard to do a bench thread with all these results and low GPU usage.

 

[SiDeards73]

Yeah Greg i dont think that demo can be relied upon for hard facts really, not unless you can control the batches and stuff manually, its a decent demo graphically.

I will run it in mantle purely to see what kinda of usage the gpu gets however.

 

[Gregster]

Yeah Greg i dont think that demo can be relied upon for hard facts really, not unless you can control the batches and stuff manually, its a decent demo graphically.

I will run it in mantle purely to see what kinda of usage the gpu gets however.

I am sure Mantle will have it running nice

 

[nutsnut]

================================================== =========
Oxide Games
Star Swarm Benchmark - ©2013
F:\My Documents\Star Swarm\Benchmark_14_01_30_2022.txt
Version 0.95
01/30/2014 20:22
================================================== =========

== Hardware Configuration =================================
GPU: AMD Radeon HD 7900 Series
CPU: GenuineIntel
Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 17145425920
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: 2174

Average FPS: 18.10
Average Unit Count: 3631
Maximum Unit Count: 5423
Average Batches/MS: 734.42
Maximum Batches/MS: 960.45
Average Batch Count: 40626
Maximum Batch Count: 138840
================================================== =========

 

[Finguz]

Don't forget the bench is still an Alpha, we shouldn't be expecting anything ' solid ' from it at this point.

 

[ICDP]

I did 3x tests at identical settings and got very consistent results. Not sure why some are claiming it is inconsistent. I find it no less inconsistent than most other benchmarks, they all give slight variations.

Test 1

Spoiler

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
Documents\Star Swarm\Benchmark_14_01_30_1956.txt
Version 0.95
01/30/2014 19:56
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon R9 200 Series
CPU: GenuineIntel
Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 17121140736
Allocatable Memory: 140737488224256
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioRTS.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: 1753

Average FPS: 14.61
Average Unit Count: 3411
Maximum Unit Count: 5614
Average Batches/MS: 1303.05
Maximum Batches/MS: 1600.43
Average Batch Count: 91122
Maximum Batch Count: 165678
===========================================================

Test 2

Spoiler

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
Documents\Star Swarm\Benchmark_14_01_30_2122.txt
Version 0.95
01/30/2014 21:22
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon R9 200 Series
CPU: GenuineIntel
Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 17121140736
Allocatable Memory: 140737488224256
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioRTS.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: 1735

Average FPS: 14.45
Average Unit Count: 3416
Maximum Unit Count: 5597
Average Batches/MS: 1289.43
Maximum Batches/MS: 1595.13
Average Batch Count: 91172
Maximum Batch Count: 166370
===========================================================

Test 3

Spoiler

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
Documents\Star Swarm\Benchmark_14_01_30_2126.txt
Version 0.95
01/30/2014 21:26
===========================================================

== Hardware Configuration =================================
GPU: AMD Radeon R9 200 Series
CPU: GenuineIntel
Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz
Physical Cores: 4
Logical Cores: 8
Physical Memory: 17121140736
Allocatable Memory: 140737488224256
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioRTS.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: 1720

Average FPS: 14.33
Average Unit Count: 3370
Maximum Unit Count: 5550
Average Batches/MS: 1274.92
Maximum Batches/MS: 1588.35
Average Batch Count: 91176
Maximum Batch Count: 156591
===========================================================

 

[SiDeards73]

Seems then that the RTS one is about the most reliable one you can get as batch and unit counts etc are all fairly close to each other.

 

[ICDP]

Seems then that the RTS one is about the most reliable one you can get as batch and unit counts etc are all fairly close to each other.

It seems so, as long as people agree on a standard then the test is valid. IMHO testing at full detail in the most demanding scenario is the only way to get accurate results.