Star Swarm - Oxide bench thread

[Gregster]

Ok guys, post your bench results here for the Star Swarm - Oxide benchmark.

Rules.

1. List your GPU and CPU and what the clock speeds are.
2. Add "Official entry" to the top/title of your post.
3. Users can alter the deferred contexts number to 1 (from 0) if you wish (steam/steammaps/common/star swarm benchmark/assets/settings_Extreme.ini
4. Click Follow and click Extreme and then choose either "Start Mantle" or "Start D3D" (as shown in the spoiler)
5. Copy and paste you complete results with the rest of the info.

Spoiler

Link to how your entry should look like.
http://forums.overclockers.co.uk/showpost.php?p=25813564&postcount=260

Any questions, fire away and enjoy

 

[Gregster]

Top 25 DX results

1. 109.09 - Andybird - 4930K @ 4.5 - Titan @ 1140
2. 84.14 - Ian Evey - 3570K @ 4.8 - 780 @ 1215
3. 81.93 - Gregster - 3930K @ 4.625 - Titan @ 1015
4. 76.70 - Spankingtexan - 4770K @ 4.5 - 780 @ 1150
5. 66.23 - los - 4670K @ 4.5 - 970 @ 1427
6. 60.78 - Coupe69 - 3820 @ 4.2 - 680 @ 1202
7. 56.79 - SkeeterUK - 3570K @ 4.5 - 670 @ 1150
8. 41.65 - Humbug - 1090T @ 3.2 - 7870 @ 1150

Top 25 Mantle results

1. 84.43 - Geeman1979 - 3930K @ 4.5 290X @ 1300
2. 82.26 - Humbug - FX9590 @ 4.7 - 290 @ 1100
3. 73.27 - theRealDeal - I7-920 @ 4.0 290 @ 1100
4. 64.66 - Humbug - 1090T @ 4.2 - 290 @ 1171
5. 57.19 - Humbug - 1090T @ 3.2 - 7870 @ 1150
6. 55.02 - Fs123 - 3770K @ 4.2 - 7950 @ 1200
7. 53.75 - Thebennyboy - FX8320 @ 4.2 - 280X @ 1200

 

[Gregster]

What is this bench and where do you get it from.

Steam. Type in search "Star Swarm" and install the bench

 

[Gregster]

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\Greg\Documents\Star Swarm\Benchmark_14_01_30_1909.txt
Version 0.95
01/30/2014 19:09
===========================================================

== Hardware Configuration =================================
GPU: NVIDIA GeForce GTX TITAN
CPU: GenuineIntel
Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz
Physical Cores: 6
Logical Cores: 12
Physical Memory: 17122181120
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: 7117

Average FPS: 59.30
Average Unit Count: 3905
Maximum Unit Count: 5393
Average Batches/MS: 794.71
Maximum Batches/MS: 2151.02
Average Batch Count: 15495
Maximum Batch Count: 185700
===========================================================

That is with a single Titan and a 3930K @ 4.625Ghz

@ Recce, I will add settings to the OP cheers

 

[Gregster]

Np pal.

Strange how the Unit/batch counts are different in the benchmarks...

Yer, not 2 results the same It will make it hard to be accurate but fun none the less

 

[Gregster]

My run some parts was low 6fps slideshow.

Spoiler

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\Frankie\Documents\Star Swarm\Benchmark_14_01_30_1914.txt
Version 0.95
01/30/2014 19:14
===========================================================

== 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: 17137754112
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: 3120

Average FPS: 25.98
Average Unit Count: 3788
Maximum Unit Count: 5473
Average Batches/MS: 676.59
Maximum Batches/MS: 865.13
Average Batch Count: 26507
Maximum Batch Count: 153908
===========================================================

I had it down to 8 fps in one spot that I noticed. Flys and slows and flys and slows

I am very interested to see how you Mantle users get on compared to what you are posting

 

[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.

 

[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

 

[Gregster]

Lol, it's amazing to see people attempt to remove this as a valid test because, "it doesn't speed up with Mantle, it slows down DX".

I think if you run the test, you will see there is big differences between them and at the same settings, so results would be all over the place and no consistancy. I enjoy a good bench but need some consistancy for it to be valid.

My mistake, I thought this would be a good bench thread to give it some but at present, the GPU usage is so low, it doesn't give any consistancy.

 

[Gregster]

Well done DL. If people want to run that and post results, I will start to do a results system

Just one thing, I noticed my unit count was quite a difference to yours but see how it goes. Here is my run for verification.

Spoiler

===========================================================
Oxide Games
Star Swarm Benchmark - ©2013
C:\Users\Greg\Documents\Star Swarm\Benchmark_14_01_31_0555.txt
Version 0.95
01/31/2014 05:55
===========================================================

== Hardware Configuration =================================
GPU: NVIDIA GeForce GTX TITAN
CPU: GenuineIntel
Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz
Physical Cores: 6
Logical Cores: 12
Physical Memory: 17122181120
Allocatable Memory: 8796092891136
===========================================================


== Configuration ==========================================
API: DirectX
Scenario: ScenarioCustom.csv
User Input: Disabled
Resolution: 1920x1080
Fullscreen: True
GameCore Update: 8.4 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: 4724

Average FPS: 39.36
Average Unit Count: 3651
Maximum Unit Count: 5330
Average Batches/MS: 1360.94
Maximum Batches/MS: 2034.88
Average Batch Count: 35721
Maximum Batch Count: 75023
===========================================================

 

[Gregster]

Can't take this bench seriously yet, gpu usage really low, no sli, My gpu hit 99% for a millisecond then the rest of test was at like 30-40% usage.

Same for me. I tried PGI's fix but the bench locks up at the very start

 

[Gregster]

I've done a little reading up on this 16.6ms thing this morning...

From what I can make out 16.6ms stems from 1-60th of a second, which is the ideal frame time refresh, the bench is defaulted to this, essentially its a form of sync/frame rate cap. Decreasing it forces the game engine to update frames on a more regular basis, though decrease too far and the game engine crashes (D3D.exe error).

It may not be put there intentionally to cripple DX in favour of mantle, more a case of the DX game coding is **** poor, heck, if mantle shows the same traits the game engine as a whole may just be poo.

Yer Paul, I don't think they would purposefully cripple DX or if they did, they would be mugs, as the mass majority of people don't have a GCN card. It is an Alpha of course, so very early in development and even when I watch the demo, I am not sure what I am looking at I ran it in 3D for fun and can see this being a great 3D game when it is out and profiled.

 

[Gregster]

I will sort something out men. It is a bit of fun

 

[Gregster]

16.6 is fixed from the get go, 9.3 doesn't work for everybody, at least one instance got it to work with 9.1 when 9.3 didn't. So those 9.x differs on CPU basis I believe, since this bench is actually closer to a CPU bench. Overall 16.6 works for everone and 9.3 works for most, those two can make a comparison base I suppose. Edit the first page for people to include 9.3 ms along with standart and post side by side for that cpu. I think you will see that people with fast intel cpus will get high results, however fx 8 cores may net more unit counts and/or batches while not so high fps results perhaps.

I think the best way is to keep a standard of 16.6 and run it on both DX and Mantle and I will add scores and names. I will possibly add it for brands as well, as in 290/290X - 780/Titan/780Ti - 280X/7970/7950 - 770/680/670 etc . Like I say, just a bit of fun

 

[Gregster]

Added some scores to the 2nd post and for any I have missed, could you please let me know and let me know what CPU and GPU you was using. Lots of entries, so possibly missed a few

As for new entries, please could you follow the rules in the OP. Don't take it too serious guys, as consistent this bench is not but no harm in giving it a blast

 

[Gregster]

Greg can you edit your tables to also show single and Multi GPU's if poss? also what clocks people run etc

I can but SLI/single makes no odds for me and in fact lowers frames. When it is multi GPU supported, I will add that in.

 

[Gregster]

just clocked mine at 1400/1950 and result was worse

Yer, I don't think there is much point taking this bench serious yet or even adding scores. I ran it a couple of times with different clocks but because the GPU is only working around 40% and even with the CPU at 5Ghz, it is making the results worse.

Maybe if it gets sorted but in the state it is, I will not be buying it.

 

[Gregster]

Hmm, changed the results format.

No combined results.

I smell fear....

?

 

[Gregster]

Mantle looking good. It is just a shame that this is the only game that doesn't allow the GPU past 50% using DX. Results obviously looking great for Mantle and poor for DX.

 

[Gregster]

Its the nature of this bench.

Evey ship, Turret and point of Light is a CPU Physics calculation. in this it keeps building and building up until the CPU gets completely bogged down.

In a Game like BF4 the Physics calculations are fixed, and they are generally fixed to a limit that DX can cope with, so mantle isn't going to make a massive difference as the render is designed to run with the DX weakness in mind.
That is, unless you have a very weak CPU, that very weak CPU will suddenly become a powerful CPU in Mantle BF4

What it shows, IMO, is the possible future, now Devs have much more CPU power to play with if they use Mantle.

If that was the case, with me clocking my 3930K to 5Ghz would result in more frames but it didn't, it actually gave me lower frames. I am claiming this is purposefully crippling anything using DX to make Mantle look good. I don't have a problem with this at all and Oxide know what they are doing.