Jump to content
Sign in to follow this  
binkster

ArmAII-Mark

Recommended Posts

ICE-Raver,

Can you confirm a couple more settings for me:

1. Have you disabled HT on your BIOS?

2. What switches have you launched arma2.exe with? -maxmem? -winxp?

3. Are your OS's patched to latest from MS?

4. Do you have any CPU or Memory perf stats to see utilization?

Cheers, T

Edited by Tupolov

Share this post


Link to post
Share on other sites
ICE-Raver,

Can you confirm a couple more settings for me:

1. Have you disabled HT on your BIOS?

2. What switches have you launched arma2.exe with? -maxmem? -winxp?

3. Are your OS's patched to latest from MS?

4. Do you have any CPU or Memory perf stats to see utilization?

Cheers, T

1. HT Disabled

2. No switches, I run the game as installed.

3. Vista is patched to SP 2, Windows 7 is also updated current

4. No I don't have anything logged to see utilization

Share this post


Link to post
Share on other sites

Windows XP SP3 32-bit

Just thought I would share my results...

PC:

Gigabyte EX58-UD5

i7 920 CPU OC'd to 3.2 (BCLK 160MHz)

ATI Radeon HD 4870X2 1GB GPU with ATI Catalyst 9.7

6 gig DDR 3 1600 Corsair Dominator Triple Channel RAM (WinXP 32bit sees 2.75GB RAM)

850 Watt Corsair PSU.

HyperThreading - Disabled

OS: Clean Build, with game installed patched to version 1.03, no anti-virus etc.

ATI Control Panel Settings:

Catalyst AI = Advanced

AF = Application Controlled

Antialiasing = Application Controlled

Adaptive Antialiasing = Disabled

Midimap Detail = Performance

Triple Buffering = OFF

Wait for Vertical Refresh = Always OFF

arma2.exe -nosplash -world=empty -winxp -maxmem=2047 -cpuCount=4

Test 3 All settings Very High, AA off, 1920x1080, fillrate 100%

OFPMark = 3756 (vs ICE-Ravers Vista64 - 3266 and Win7 64 - 4338)

No switches = 3515 and default video memory = 3726

Switches and Video Memory selected as Default = 3730

Test 4 All settings Very High, Plus AA High (Very High isn't available?), 1920x1080, fillrate 100%

OFPMark = 2770 (vs ICE-Ravers Vista64 - 2750 and Win7 64 - 2815)

Test 5 All settings normal, PP High (default setting for Quality preference Normal), AA off, 1920x1080, fillrate 100%

OFPMark = 5043 (vs ICE-Ravers Win7 64 - 7192 that was clocked to 4.2)

No switches = 5090 and default video memory = 4698

Switches and Video Memory selected as Default = 5080 and PP as Low = 5324

Switches and Very High Video Memory (PP as High) = 5042...

The first two results show how, even with increased screen resolution, WinXP 32 bit is beating Vista64. I'm not using two graphics cards either - just the dual GPU 4870X2 card. Also I'm only using 2GB of RAM.

The final test is a funny one. Video Memory settings seem to make little difference. Looks better to have it as default in the last test.

Clearly I need to upgrade to Win 7 64 bit and overclock the CPU a bit more!

Edited by Tupolov
Updating as I test

Share this post


Link to post
Share on other sites

Hi zaira I do have some numbers I have worked with on visibility however I dont have them all compiled yet but I will share my raw data with you very quickly and if anyone else can help verify this i would appreciate it. if you are meaning Fill is the same as visibility.

Visibility Benchmarks as to FPS penalty.

This benchmark was done in a open field looking northwards towards a wooded treeline and mountains. Testing the impact that trees have on FPS. I carefully balanced the sweet spot to 60 FPS to give me a good scale and then increased in value until I reached a 2 FPS drop and then recorded the results. As you can see I found a penalty between 500 to 690 with 695 to 1260 being optimal and the drop in FPS from there. How this will scale with other systems and locations I have no idea. It was only to study the impact of Visibility on trees, how ever there does seem to be some changes to the vegetation in the foreground up to 50m I could not really tell if more was added or the vegetation was just turning or rearranging slightly.

Vis. FPS Range

500 55 penalty

546 53 penalty

598 57 penalty

635 58 penalty

695 60 1260

1266 58 2370

2374 56 2925

2928 54 3300

3301 52 4070

4071 50 5020

5021 48 5659

5660 46 6375

6380 44 7410

7411 42 7865

7869 40 8609

8609 38 9415

9419 36 9999

Share this post


Link to post
Share on other sites

So is there a general feel here the game's running better with higher OCs and HT disabled? If so, great! I have a C0/C1 stepping i7 920 and therefore it can't crack 4GHz with HT enabled but it sure does with HT disabled :D

Maybe I'll start doing some more ArmA2Mark tests.

That said, I've actually dropped my OC down (from what's in my sig) to 3.48GHz @ 1.25v (simply to have a more efficient PC) and I'm happy with how the game's running.

Jero.

Edited by doakwolf

Share this post


Link to post
Share on other sites

Test 4 All settings Very High, Plus AA High (Very High isn't available?), 1920x1080, fillrate 100%

OFPMark = 2770 (vs ICE-Ravers Vista64 - 2750 and Win7 64 - 2815)

On your test 4 for me Very high AA is an option. AA High/Very High were 2 different settings. Probably SLI related. I never ran arma mark with AA set to high, only very high/normal. But yea, something is really up with Vista on this game. Some people run it great, for others (like me) it is a big turd.

arma22009-08-1418-33-39-35.jpg

Edited by ICE-Raver

Share this post


Link to post
Share on other sites
Hi zaira I do have some numbers I have worked with on visibility however I dont have them all compiled yet but I will share my raw data with you very quickly and if anyone else can help verify this i would appreciate it. if you are meaning Fill is the same as visibility.

Visibility Benchmarks as to FPS penalty.

This benchmark was done in a open field looking northwards towards a wooded treeline and mountains. Testing the impact that trees have on FPS. I carefully balanced the sweet spot to 60 FPS to give me a good scale and then increased in value until I reached a 2 FPS drop and then recorded the results. As you can see I found a penalty between 500 to 690 with 695 to 1260 being optimal and the drop in FPS from there. How this will scale with other systems and locations I have no idea. It was only to study the impact of Visibility on trees, how ever there does seem to be some changes to the vegetation in the foreground up to 50m I could not really tell if more was added or the vegetation was just turning or rearranging slightly.

Vis. FPS Range

500 55 penalty

546 53 penalty

598 57 penalty

635 58 penalty

695 60 1260

1266 58 2370

2374 56 2925

2928 54 3300

3301 52 4070

4071 50 5020

5021 48 5659

5660 46 6375

6380 44 7410

7411 42 7865

7869 40 8609

8609 38 9415

9419 36 9999

Nice table showing clear corelation fps/VD. :)

Could someone with i7 with 4+Ghz bench with this setings

arma22009081508401249.th.jpg

Share this post


Link to post
Share on other sites

Texture Detail - Normal

Anisotropic Filtering - High

Terrain Detail - Normal

Objects Detail - Normal

Shadow Detail - Normal

PostProcess Effects- Low

Cpu - Q6600 OC (3.2ghz)

Ram - OCZ Gold PC6400 @ 852mhz

GPU - Asus 8800GTS 512

OS - Win7 x64

Resolution - 1440 x 900

Test 1: 35.5352

Test 2: 39.6092

Test 3: 35.5796

Test 4: 47.8851

Test 5: 13.4667

Xclusiv8's OFPMark is 3441.51

Share this post


Link to post
Share on other sites

arma2marksettingsnoramlpostlow16009.png

662106.png

my rig :rolleyes: image is link

xp 32

1600x900 resolution

gfx driver 190.89

default settings: all normal, post processing low

Edited by [DirTyDeeDs]-Ziggy-

Share this post


Link to post
Share on other sites

Just to show you that is something really wrong with my computer....

Settings:

PP to low

Everything other to normal (mean everything)

1680x1050 res

ArmA2 Mark Score : 940

arma22009082303434422.th.png

Specs:

C2D E6420 @2.66GHz*

3.2GB 667MHz CL4*

GF 9800GT OC (675/1550/1998)

Seagete 7200.12 500GB (as ArmA 2 disk)

*And now the bottleneck of hole rig (CPU and RAM can't run at any higher speeds, MoBo doesn't read all 4GBs of RAM too)

Gigabyte GA-945P-S3 Motherboard

But i really don't know if it's only MoBo holding me up... Wating for new one, but it's hard to explain my sponsors that i need new MoBo ASAP ;)

Share this post


Link to post
Share on other sites
;1416515']Just to show you that is something really wrong with my computer....

Settings:

PP to low

Everything other to normal (mean everything)

1680x1050 res

ArmA2 Mark Score : 940

arma22009082303434422.th.png

Specs:

C2D E6420 @2.66GHz*

3.2GB 667MHz CL4*

GF 9800GT OC (675/1550/1998)

Seagete 7200.12 500GB (as ArmA 2 disk)

*And now the bottleneck of hole rig (CPU and RAM can't run at any higher speeds' date=' MoBo doesn't read all 4GBs of RAM too)

[b']Gigabyte GA-945P-S3 Motherboard[/b]

But i really don't know if it's only MoBo holding me up... Wating for new one, but it's hard to explain my sponsors that i need new MoBo ASAP ;)

You're right, there's something wrong with your setup obviously. Try changing memory timings and see if that helps.

I have similar setup than yours (you have slightly faster cpu & mem ) and i got 1860 arma mark score. I even used PP high setting.

Share this post


Link to post
Share on other sites

Well, problem is, if i change any setting the result may be totally opposite - my MoBo is so much fu**ed up. It's a miracle that it allow me to have CPU OCed to 2.66 (default is 2.13), sometimes it even don't accept FSB changes and i have double restart all the time... So i'm pretty happy with current setting, but still want to throw this Gigashit piece of crap far, far away... My memory have default speed higher than current - 800MHz CL4, but... Well, you know the "but"... And that is all - this isn't topic about PC problems ;)

Share this post


Link to post
Share on other sites
Windows XP SP3 32-bit

Just thought I would share my results...

PC:

Gigabyte EX58-UD5

i7 920 CPU OC'd to 3.2 (BCLK 160MHz)

ATI Radeon HD 4870X2 1GB GPU with ATI Catalyst 9.7

6 gig DDR 3 1600 Corsair Dominator Triple Channel RAM (WinXP 32bit sees 2.75GB RAM)

850 Watt Corsair PSU.

HyperThreading - Disabled

OS: Clean Build, with game installed patched to version 1.03, no anti-virus etc.

ATI Control Panel Settings:

Catalyst AI = Advanced

AF = Application Controlled

Antialiasing = Application Controlled

Adaptive Antialiasing = Disabled

Midimap Detail = Performance

Triple Buffering = OFF

Wait for Vertical Refresh = Always OFF

arma2.exe -nosplash -world=empty -winxp -maxmem=2047 -cpuCount=4

Test 3 All settings Very High, AA off, 1920x1080, fillrate 100%

OFPMark = 3756 (vs ICE-Ravers Vista64 - 3266 and Win7 64 - 4338)

No switches = 3515 and default video memory = 3726

Switches and Video Memory selected as Default = 3730

Test 4 All settings Very High, Plus AA High (Very High isn't available?), 1920x1080, fillrate 100%

OFPMark = 2770 (vs ICE-Ravers Vista64 - 2750 and Win7 64 - 2815)

Test 5 All settings normal, PP High (default setting for Quality preference Normal), AA off, 1920x1080, fillrate 100%

OFPMark = 5043 (vs ICE-Ravers Win7 64 - 7192 that was clocked to 4.2)

No switches = 5090 and default video memory = 4698

Switches and Video Memory selected as Default = 5080 and PP as Low = 5324

Switches and Very High Video Memory (PP as High) = 5042...

The first two results show how, even with increased screen resolution, WinXP 32 bit is beating Vista64. I'm not using two graphics cards either - just the dual GPU 4870X2 card. Also I'm only using 2GB of RAM.

The final test is a funny one. Video Memory settings seem to make little difference. Looks better to have it as default in the last test.

Clearly I need to upgrade to Win 7 64 bit and overclock the CPU a bit more!

You should be getting much better scores than that? Thats my opinion.

Maybe the 2G is holding you back, but I wouldnt have thought it would matter THAT much.

Ill be interested to see your Win7 results.....

Share this post


Link to post
Share on other sites

SCORE 4242.38

Resolution - 1680x1050

Aspect Ratio - 16:10

Texture Detail - Normal

Video Memory - High

Anisotropic Filtering - Normal

Antialiasing Filtering - Disabled

Terrain Detail - Normal

Objects Detail - Normal

Shadow Detail - Normal

PostProcess Effects- Low

Interface Size - Small

CPU - E8400 (3800MHz)

RAM - 2x2048MB DDR2 (475MHz)

GPU - ATi HD4890 1GB

OS - WinXP SP3

Share this post


Link to post
Share on other sites

SCORE 4656.91

Resolution - 1920x1200

Aspect Ratio - 16:10

Texture Detail - Normal

Video Memory - High

Anisotropic Filtering - Normal

Antialiasing Filtering - Disabled

Terrain Detail - Normal

Objects Detail - Normal

Shadow Detail - Normal

PostProcess Effects- Low

Interface Size - Small

CPU - i7-975 EE (4x3333MHz)

RAM - 3x2048MB DDR3 (1333MHz)

GPU - NVIDIA GTX295 1792 MB

OS - Win7 64bit

Share this post


Link to post
Share on other sites

Texture Detail - Normal

Anisotropic Filtering - Normal

Terrain Detail - Normal

Objects Detail - Normal

Shadow Detail - Normal

PostProcess Effects- Low

Cpu - E6850 3 ghz Overclocked to (3.6ghz)

Ram - 4 gig Kingston DDR3 1333mhz

GPU - MSI 260 GTX 216core Clock 700mhz Memory 1105mhz

OS - Vista 32bit

Resolution - 1600 x 1200

Score - 2791

ArmA2Mark.jpg

Share this post


Link to post
Share on other sites

Score 3207

Visability - 1530

Texture - High

Video Mem - High

Af - Normal

AA - Low

Terrain - Normal

Objects - Low

Shadows - High

Post Process - High

____________________

Intel core 2 Quad Q9550 @ 3.0 Ghz

BFG Tech GTX 285 OC to core 683 MHz. . mem 1247 MHz

Asus P5N-D 750i

4 GB Corsair DDR2 XMS 800

EZ COOL 700W psu

Maxtor HDD

Windows Vista 64bit

ARMA 2 MARK 3207 with high and med settings

Share this post


Link to post
Share on other sites

All settings Normal, except

Postprocessing: Low

Anti-Aliasing: Disabled

Resolution 1600x1020

arma22009090616303990.jpg

Its funny, in this game FRAPS will be telling me I am averaging 35-45 FPS, but to my eye it still looks jerky, why is that?

My system:

Core 2 Duo e8400 @ stock

HD 4850 512MB @ stock

4GB DDR 2 Ram

Win XP Home.

--------

We really need some standard settings here so we can compare.

I set mine to what it looked like the concensus for a standard was, others should do the same. ( However, we should prob use 1280x1024 as standard because some may run low res monitors).

Edited by thaFunkster

Share this post


Link to post
Share on other sites

Have made some tweaks recently that has finally brought my game up to spec for the hardware I am running.

Not sure if any of this info will be helpful to others but here are some points to consider.

HT off

Moved my swap file to my two spare drives, each drive has a 2-4gb swap. SSD has no swap file. This resolved my jerky/studdering gameplay.

Using ATT to turn off vsync (massive improvement with this, FPS is up and mouse lag is gone)

Using Windows 7 Professional 64 RTM

Using Cat 9.8

No switches in my start shortcut

For this test CPU was 'only' running at 3.8ghz, both GFX are overclocked to 800mhz on all four cores

Texture Detail - Normal

Anisotropic Filtering - Normal

Terrain Detail - Normal

Objects Detail - Normal

Shadow Detail - Normal

PostProcess Effects- Low

Fill rate 100%

Resolution - 1680 x 1050

arma22009090607532360.jpg

For my normal multiplayer settings (1600x1200) I get a score in the high 8k's. Can do a run and screen shot as such if needed ;)

Edited by Gamer

Share this post


Link to post
Share on other sites

score: 3475

thats 2nd time running, always use 2nd stat. it's much like going to a webpage you've already been to, thus loads faster. its not cheating.

i ran many different gfx options. each setting i ran the app fully, then restarted mission and record its results, change to next setting, reboot arma2 and try next setting. best result happened to be my preferred settings.

Interface Resolution: 1900x1200

3D resolution: 1900x1200 (100%)

Texture Detail: normal

Video Memory: very high

AF: high

AA: off

Terrain: normal

Objects: normal

Shadow: normal

PP effects: low

Interface size: very small

Aspect ratio: 16:10 widescreen

i wish i could get ATT to work so i could disable vsync. the CCC vsync option doesnt seem to be working.

p.s. no screenshot thanks to vista

Edited by Shooter McGavin

Share this post


Link to post
Share on other sites
always use 2nd stat. it's much like going to a webpage you've already been to, thus loads faster. its not cheating.

No, it's not cheating because it's not a competition - but it doesn't give an accurate rating to compare to other user's results on this thread (unless we all use run #2). When playing the 'game', there is no opportunity for the PC to 'pre load' the textures prior to entering a town or new part of the map etc (unless it is, in fact, coded that way, somehow) so using your second score in ArmA2-Mark doesn't give an accurate result, IMO.

But that said, half the ppl on this thread are quoting first run scores and the other half quoting second run scores so it's impossible to compare specs and see how your PC is going compared to other's.

My 2c :)

Share this post


Link to post
Share on other sites
score: 3475

thats 2nd time running, always use 2nd stat. it's much like going to a webpage you've already been to, thus loads faster. its not cheating.

i ran many different gfx options. each setting i ran the app fully, then restarted mission and record its results, change to next setting, reboot arma2 and try next setting. best result happened to be my preferred settings.

Interface Resolution: 1900x1200

3D resolution: 1900x1200 (100%)

Texture Detail: normal

Video Memory: very high

AF: high

AA: off

Terrain: normal

Objects: normal

Shadow: normal

PP effects: low

Interface size: very small

Aspect ratio: 16:10 widescreen

i wish i could get ATT to work so i could disable vsync. the CCC vsync option doesnt seem to be working.

p.s. no screenshot thanks to vista

I had to jump through a few hoops to get ATT installed on W7, you might need to do the same with Vista, not sure. The problem for me in W7 is that ATT uses an unsigned driver.

First off you need to disable UAC (User account control) in windows control panel/user accounts.

Then reboot

Then go to start menu/run type 'cmd'. You will then see 'cmd' on the top of the start program list, right click and choose 'run as administrator'.

Once opened type or copy and paste the following into CMD and press enter:

bcdedit /set testsigning on

bcdedit /set loadoptions DDISABLE_INTEGRITY_CHECKS

That will turn off unsigned drivers and allow you to install the ATT driver.

Hope that helps. ;)

Edited by Gamer

Share this post


Link to post
Share on other sites
No, it's not cheating because it's not a competition - but it doesn't give an accurate rating to compare to other user's results on this thread (unless we all use run #2). When playing the 'game', there is no opportunity for the PC to 'pre load' the textures prior to entering a town or new part of the map etc (unless it is, in fact, coded that way, somehow) so using your second score in ArmA2-Mark doesn't give an accurate result, IMO.
Doesn't the first run get slowed down by disk access and is not a true representation of the actual video performance?

Share this post


Link to post
Share on other sites

Ok, my previous results as per last page were ~2800-

Same settings today, having downloaded Win7: ~3800!

Thats over a 30% increase.

My FPS in the spin test went from ~30 to 50

While my FPS in 'Space Capsule' went from ~14 to ~30!

Pretty impressive.

The main factor is the new catalyst drivers which take full advantage of Win 7. I have also overclocked my processor from 3G to 3.6, but this will only make a slight difference, as those test wont bu pushing the cpu as much as GPU.

To prove, I will later on re-do both tests with the OC, and post screenies to prove it.

Recommend Win 7 to all struggling XP users! Now I can use FSAA and still play smooth.

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.
Sign in to follow this  

×