ArmA3Mark - Benchmark your ArmA 3

[Brady 1]

- RESULTS -\n-----------\nTest One - 59.7464\nTest Two - 66.5528\nTest Three - 48.2984\nTest Four - 69.7269\nTest Five - 159.432\nBrady 319's OFPMark is 8075.13!\n\n - press 'ESC' to quit -

Intel Core i7 3960x

Asus P9x79 Deluxe MB

64gb pc 24000 gskill ram

coolermaster cosmos ii case

Corsair 1100 watt psu

Dual EVGA Classified gtx 690's in quad sli

Edited May 2, 2013 by Brady

[i_2_i 10]

1920x1200 w video settings where possible disabled or set to lowest possible value!

- RESULTS -

-----------

Test One - 35.9181

Test Two - 42.4742

Test Three - 32.8018

Test Four - 44.2478

Test Five - 64.7722

my's OFPMark is 4404.28!

amd 1090t (oc 3.6 ghz x 6 cores)

amd 890 chipset

amd hd 6870 1gb (oc 950/1150)

16 gb ddr 3 1333

128 gb sata 3 ssd (system)

2 tb sata 3 7200 rpm (programs)

2 tb sata 3 7200 rpm (users)

windows 8 64-bit (windows 8 sucks)

Edited April 16, 2013 by i_2_i

[dasa 20]

i would have run the ultra preset but it leaves texture detail below ultra

so here is my preferred settings at the moment

hmm first post cant link the pic just yet

2600k@4.7ghz 1.38v 2133 9-10-10-1t 7970@1150\7000 1.16v

2560x1440 cf\single no difference

http://i3.photobucket.com/albums/y83/dasa09/arma3alphabench_zpsf423ec48.png (270 kB)

been playing with settings some more and for some reason the ultra texture setting doesnt seem to make any difference over very high at least not on the landscape i was looking at maybe it affects different items but hopefully the textures just arnt there yet in the alpha

the other thing is fxaa seems to sharpen things which is the opposite to how it works in most other games where it blurs things

smaa which is also usually sharper than fxaa is blurry just like with fxaa disabled

its a shame we dont have more control over the post process effects from the menu as some things it does look decent but other parts of it look horid

Edited April 20, 2013 by dasa

[tonschuh 3]

Motherboard: Asus P8Z77-V

CPU: 3770k @4200MHz

RAM: 16GB (4x4GB) G.Skill @2133MHz

GPU: 2x Gigabyte GTX680-SOC-2GB in 2-Way-SLI

OS-Drive: Intel 520S 240GB SATA-3 SSD

Game-Drive: Seagate Constellation 3TB 7200RPM 64MB Cache

OS: Windows-8-Pro-WMC-x64

GPU-Driver: NVIDIA GeForce v320.00 Beta

Screen-Resolution: 1920x1080

Benchmark: v0.7

NVIDIA-Control-Panel:

Version-1: Default Settings (Global / ARMA-3) [*]

Version-2: see below [**]

Version-3: Default Settings, but "Adaptive VSync" + "Prefer maximum performance" + "Threaded Optimisation: On" [***]

ARMA-3-Settings:

Ambient Occlusion: Quality

Anisotropic filtering: 16x

Antialiasing - FXAA: On

Antialiasing - Gamma correction: On

Antialiasing - Mode: Enhance the application settings

Antialiasing - Setting: 32x CSAA

Antialiasing - Transparency: 8x supersample

CUDA - GPUs: All

Maximum pre-rendered frames: Use the 3D application settings

Multi-display / mixed-GPU acceleration: Single display performance

Power management mode: Prefer maximum performance

SLI rendering mode: NVIDIA recommended

Texture filtering - Anisotropic sample optimation: On

Texture filtering - Negative LOD bias: Clamp

Texture filtering - Quality: Quality

Texture filtering - Trinilinear optimisation: On

Threaded optimisation: On

Triple buffering: On

Vertival sync: Adaptive

SLI configuration: Maximise 3d performance

PhysX settings: Auto-select (GTX680(2))

Alpha 0.54.103957 (Dev-Build 0.55.104588 doesn't work)

Low:

Test-1: 71.9062*

Test-2: 73.5311*

Test-3: 50.1362*

Test-4: 78.3801*

Test-5: 119.392*

OSPMark: 7866.91*

Standard:

Test-1: 60.8005*

Test-2: 56.2493*

Test-3: 41.5038*

Test-4: 59.5829*

Test-5: 92.3815*

OSPMark: 6210.36*

High:

Test-1: 50.6289*

Test-2: 45.3636*

Test-3: 33.5287*

Test-4: 45.1977*

Test-5: 68.8529*

OSPMark: 4871.44*

Very High:

Test-1: 46.8263*

Test-2: 37.4684*

Test-3: 27.7348*

Test-4: 35.0263*

Test-5: 52.0806*

OSPMark: 3982.73*

Auto-Detect:

Test-1: 43.2567*

Test-2: 33.3628*

Test-3: 24.1408*

Test-4: 29.5712*

Test-5: 41.6111*

OSPMark: 3438.85*

Ultra:

Test-1: 42.2393*

Test-2: 33.6436*

Test-3: 24.9626*

Test-4: 27.8422*

Test-5: 41.3416*

OSPMark: 3400.59*

Custom / maximum Quality:

Test-1: 41.9934*

Test-2: 33.0739*

Test-3: 24.6296*

Test-4: 28.0965*

Test-5: 30.0907*

OSPMark: 3157.68*

Low:

Test-1: 59.8887**

Test-2: 59.9152**

Test-3: 52.0236**

Test-4: 60.241**

Test-5: 59.9346**

OSPMark: 5840.06**

Standard:

Test-1: 59.1894**

Test-2: 57.0385**

Test-3: 49.3928**

Test-4: 60.3318**

Test-5: 59.2588**

OSPMark: 5704.23**

High:

Test-1: 52.1327**

Test-2: 44.2546**

Test-3: 39.3661**

Test-4: 46.0123**

Test-5: 58.9333**

OSPMark: 4813.98**

Very High:

Test-1: 47.8983**

Test-2: 36.4332**

Test-3: 33.4**

Test-4: 35.7782**

Test-5: 52.2876**

OSPMark: 4115.95**

Auto-Detect:

Test-1: 45.2281**

Test-2: 31.2831**

Test-3: 28.6829**

Test-4: 31.4136**

Test-5: 42.5475**

OSPMark: 3583.11**

Ultra:

Test-1: 45.5115**

Test-2: 31.775**

Test-3: 28.7583**

Test-4: 31.3972**

Test-5: 40.977**

OSPMark: 3568.38**

Custom / maximum Quality:

Test-1: 43.267**

Test-2: 30.7556**

Test-3: 28.6389**

Test-4: 29.2255**

Test-5: 28.2207**

OSPMark: 3202.15**

Auto-Detect:

Test-1: 46.7019***

Test-2: 32.5529***

Test-3: 30.1336***

Test-4: 31.7041***

Test-5: 42.0308***

OSPMark: 3662.47***

There are far too many factors / settings to concider for a comparison.

:)

[baaaard 10]

This mission is broken for me atm on the dev build. It starts up, the ifrits drives, but stops and dont proceed at the end of test1

[helo 10]

That´s odd. However, I keep my fingers fro mthe dev builds sicne quite some time. :)

[novahot 1]

Is it broken due to the new stable build today?

[Tonci87 163]

Broken...

[wormeaten 0]

I got Q6600 quad intel cpu OC on 3Ghz, 4Gb (2x2Gb DDR2 800Mhz), HD 6870 Radeon and I cant set it up for proper PvP gameplay so I decide to upgrade my rig.

I planing to upgrade my rig with new MB, CPU and 16Gb RAM plus maybe SSD.

Latter I'm planing to upgrade GC to GTX770 when it will release.

I don't know should I go AMD FX-8350 or i5 3570K and what is real power difference in ArmA 3 between i5 3570K and i7 3770K?

I'm not interested in some theoretic difference in speed zipping file or in Photoshop. This difference is for professionals who have multiple pictures or zip files to encode or compress or hours of video where this difference is worth the money.

I'm asking in game real experience is this difference is noticeable?

In this video go forward on 7:00 min they clame fx8350 on same GTX670 is twice better than i5 3570K in ArmA 2. Can someone confirm this?

And in the end pleas keep some sort of standards in this benchmarks so we could have point of this hardware tests otherwise all this effort you are doing is pointless and useless.

To be able to compare hardware and performance we need to establish some sort of standards for example:

• CPU, Core number and Mhz
  
• MB
  
• RAM and Mhz
  
• GPU ( here should be which model only one GPU, with SLI/CF turned off )
  
• 2xGPU ( this is to compare SLI/CF performance when it is turned on )
  

That is about hardware. If we do it right that way we could track data how some differences in hardware affect in performance for example if two testers have same MB CPU GPU and RAM but only difference is RAM speed.

One is run 16Gb DDR3 on 1333Mhz and other have 16Gb DDR3 on 1866Mhz we could calculate is this difference is significant or not and how much it is.

Other condition is to use only game preset and same resolution so this should be:

• 1920:1080 resolution is mandatory
  
• View distance 2600 m mandatory
  
• Draw objects 1600 m mandatory
  
• game preset Standard (could be also used other game preset like for example Low or Ultra but Standard should be mandatory as middle set up and starting point for compare)
  

This way we will keep some standards to get more precise results which could be really compared and we will get right answers about some hardware. About View Distance and Draw objects distance I put this value as some middle range between low end hardware and high end hardware. This effect on performance allot most people reduce this distance to improve performance. Also comparing with other games like Bf3 or Crysis 3 this is still twice as much this game use it in this game so for better results this is more than good distance which most of the players use.

From now on this will be good starting point and standard what we should use in publish our benchmark results.

Right now 80% of published results are hard to compare or useless because not good standardisation.

I'm appealing on you from now on use this standard as mandatory for benchmark results so we could have better and more precise comparison of results.

Thank you very much on using it and understanding.

[pr0ph3tswe 1]

I got Q6600 quad intel cpu OC on 3Ghz, 4Gb (2x2Gb DDR2 800Mhz), HD 6870 Radeon and I cant set it up for proper PvP gameplay so I decide to upgrade my rig.

I planing to upgrade my rig with new MB, CPU and 16Gb RAM plus maybe SSD.

Latter I'm planing to upgrade GC to GTX770 when it will release.

I don't know should I go AMD FX-8350 or i5 3570K and what is real power difference in ArmA 3 between i5 3570K and i7 3770K?

I'm not interested in some theoretic difference in speed zipping file or in Photoshop. This difference is for professionals who have multiple pictures or zip files to encode or compress or hours of video where this difference is worth the money.

I'm asking in game real experience is this difference is noticeable?

In this video go forward on 7:00 min they clame fx8350 on same GTX670 is twice better than i5 3570K in ArmA 2. Can someone confirm this?

And in the end pleas keep some sort of standards in this benchmarks so we could have point of this hardware tests otherwise all this effort you are doing is pointless and useless.

To be able to compare hardware and performance we need to establish some sort of standards for example:

• CPU, Core number and Mhz
  
• MB
  
• RAM and Mhz
  
• GPU ( here should be which model only one GPU, with SLI/CF turned off )
  
• 2xGPU ( this is to compare SLI/CF performance when it is turned on )
  

That is about hardware. If we do it right that way we could track data how some differences in hardware affect in performance for example if two testers have same MB CPU GPU and RAM but only difference is RAM speed.

One is run 16Gb DDR3 on 1333Mhz and other have 16Gb DDR3 on 1866Mhz we could calculate is this difference is significant or not and how much it is.

Other condition is to use only game preset and same resolution so this should be:

• 1920:1080 resolution is mandatory
  
• View distance 2600 m mandatory
  
• Draw objects 1600 m mandatory
  
• game preset Standard (could be also used other game preset like for example Low or Ultra but Standard should be mandatory as middle set up and starting point for compare)
  

This way we will keep some standards to get more precise results which could be really compared and we will get right answers about some hardware. About View Distance and Draw objects distance I put this value as some middle range between low end hardware and high end hardware. This effect on performance allot most people reduce this distance to improve performance. Also comparing with other games like Bf3 or Crysis 3 this is still twice as much this game use it in this game so for better results this is more than good distance which most of the players use.

From now on this will be good starting point and standard what we should use in publish our benchmark results.

Right now 80% of published results are hard to compare or useless because not good standardisation.

I'm appealing on you from now on use this standard as mandatory for benchmark results so we could have better and more precise comparison of results.

Thank you very much on using it and understanding.

http://gamegpu.ru/action-/-fps-/-tps/arma-iii-alpha-test-gpu/testovaya-chast.html

if you scroll down you have cpu benchmark for arma 3, should give you an idea how the different cpu's perform

sadly not many people seem to post in here with amd fx benchmarks :(

[helo 10]

The view distance is set by the mission. It overrides the user gfx settings regarding the viewing distance :)

[wormeaten 0]

http://gamegpu.ru/action-/-fps-/-tps/arma-iii-alpha-test-gpu/testovaya-chast.html

if you scroll down you have cpu benchmark for arma 3, should give you an idea how the different cpu's perform

sadly not many people seem to post in here with amd fx benchmarks :(

Yes, Thank you man I even understand some of that Russian text.

The view distance is set by the mission. It overrides the user gfx settings regarding the viewing distance :)

Yes maybe we should think about to set it up inside mission that way.

I was even think about to reduce it even mor to 1600/1000m to be more comparable with BF3 performance. In Bf2 distance is shorter.

[babydoc 1]

The bench stops after the first Ifrit mission

I use the 0.56 build.

[azz_er 10]

I get a script error regarding a stop expression and the bench stops after the Ifrit mission :(

[Tigerentner 10]

I get a script error regarding a stop expression and the bench stops after the Ifrit mission :(

Same problem here.

[Mellonpopr 10]

same problem with latest dev build and .6 or .7 version of the benchmark tool

Edited June 11, 2013 by Mellonpopr

[Waffle_SS 11]

Why the hell is this using .sqs files?!?

Those are made for use in Operation Flash Point!

[St. Jimmy 272]

Why the hell is this using .sqs files?!?

Those are made for use in Operation Flash Point!

I think he has just ported this since OFP with minimal changes.

[helo 10]

Hi guys,

sicne BIS still hasn´t provided us with a Benchmark mission, I tried to adjust the mission from the alpha to the changes of the current Beta. However, everytime I run the mission it says (in the .rpt)

Error in expression <stop>

Error position: <stop>

Error Undefined variable in expression: stop

the init looks as follows:

;exit

tests = 0

scores = []

;goto "test5"

0 fadesound 0

0 fademusic 0

titlecut["ArmA3Mark v0.7 script credit to tntoadlife@toadlife.net ;gammadust","black in",10000000000]

~2

titlecut["","black in",10000000000]

[] exec "test1.sqs"

@stop

~5

stop=false

titlecut["","black out",3]

~2

titletext[format["Test One Score: %1",scores select tests],"plain"]

~5

tests = tests + 1

[] exec "test2.sqs"

@stop

~5

stop=false

titlecut["","black out",3]

~2

titletext[format["Test Two Score: %1",scores select tests],"plain"]

~5

tests = tests + 1

[] exec "test3.sqs"

@stop

~5

stop=false

titlecut["","black out",3]

~2

titletext[format["Test Three Score: %1",scores select tests],"plain"]

~5

tests = tests + 1

[] exec "test4.sqs"

@stop

~5

stop=false

titlecut["","black out",3]

~2

titletext[format["Test Four Score: %1",scores select tests],"plain"]

~5

tests = tests + 1

#test5

[] exec "test5.sqs"

@stop

~5

stop=false

titlecut["","black out",3]

~2

titletext[format["Test Five Score: %1",scores select tests],"plain"]

~5

#scoreloop

copyToClipboard format["- RESULTS -\n-----------\nTest One - %1\nTest Two - %2\nTest Three - %3\nTest Four - %4\nTest Five - %5\n%7's OFPMark is %6!\n\n - press 'ESC' to quit -",(scores select 0),(scores select 1),(scores select 2),(scores select 3),(scores select 4),(((scores select 0) + (scores select 1) + (scores select 2) + (scores select 3) + (scores select 4)) / 5) * 100,(name player)]

titletext[format["- RESULTS -\n-----------\nTest One - %1\nTest Two - %2\nTest Three - %3\nTest Four - %4\nTest Five - %5\n%7's OFPMark is %6!\n\n - press 'ESC' to quit -",(scores select 0),(scores select 1),(scores select 2),(scores select 3),(scores select 4),(((scores select 0) + (scores select 1) + (scores select 2) + (scores select 3) + (scores select 4)) / 5) * 100,(name player)],"plain"]

~5

goto "scoreloop"

What do I have to change to make the benchmark running again?

[griffz 1]

the command setviewdistance doesnt override the "Object view distance" from what i saw in multiplayer.

[helo 10]

Ah, okay, good to know. Thought it would be still the same.

Nonetheless, I still appreciate input regarding the stop command in ArmA 3.

---------- Post added at 11:29 AM ---------- Previous post was at 11:21 AM ----------

Ah, okay, good to know. Thought it would be still the same.

Nonetheless, I still appreciate input regarding the stop command in ArmA 3.

[helo 10]

I gonna release an entirely new benchmark mission for the Beta in some hours...

---------- Post added at 08:59 PM ---------- Previous post was at 07:10 PM ----------

Here´s the Arma 3 Benchmark mission for the Beta. It´s made from the scratch in cotnrast to the version released by me for the Alpha and this time it comes with no view distance presets meaning you can configure whatever distances suiting yourself.

Download link:

https://docs.google.com/file/d/0ByhZc2l2kSOaX0czdDhaRnMtbW8/edit?usp=sharing

Happy benchmarking.

[Guest]

New version frontpaged on the Armaholic homepage.

ArmA3Mark [bETA] v0.51

[jumpinghubert 48]

thanks very much! I would like to see a more infantry-perspective based benchmark.

43fps

Edited June 29, 2013 by JumpingHubert

[helo 10]

What do you mena with infantry perspective ? Possibly a benchmark with infantry AI only ?