Arma 3 CPU vs RAM performance comparison 1600-2133= up to 15% FPS gain

[Punisher5555 0]

I had 8 gig of RAM 1600 and I went for 32 gig 1600 (for After Effect, not Arma) and I saw no difference

I guess performance gains are more about the 1600mhz => 2333 mhz

Now that you have 32GB of RAM your need to setup a 12GB RAMDRIVE, copy the Addons directory to it, set your -mod=<drive letter>:\Addons Then you will performance.

[beavis_5000 10]

Hey guys can you dumb down this thread for me and explain it like im a 10 year old lol. Also tell me how I can apply this.

[roshnak 41]

Does switching from 8 GB PC 16000 CAS 7 to 16 GB PC 16000 will also increase performance ?

Because i read on Arma 3 feedbacktracker, arma can go over 3,4, or 5 GB of RAM due to some tweak already into the game. I'm not a master of that so i can't explain how it works. But i never see anything testing the game with 8 GB or 16 GB of RAM.

thank you.

16 GB from 8 GB isn't going to make a difference in Arma. 8 GB is already more than the program can use. From what I understand it can use like 4 GB of RAM on a 64-bit OS and then the other gig is from your pagefile I could be slightly off on this, but either way 8 GB should be sufficient for Arma.

Hey guys can you dumb down this thread for me and explain it like im a 10 year old lol. Also tell me how I can apply this.

Faster RAM makes the game run better. To "apply" it you'd have to buy faster RAM.

[beavis_5000 10]

hank you Roshnak.

[dasa 20]

TOTALLY IRREVELANT here. anand bench was made using IGP ! where RAM speed has all its importance !

with a discret GPU , GPU bandwith doesn't depend of cpu RAM speed anymore.

totally irrevelant ...

read it again

they also run some tests with a 6950 and some more with 5970 & 5870 for tri-fire which sees up to 13% gain moving from 1333c9 to 2133c9\2400c10 but the reason i linked it wasnt for the performance gains but due to the lengths they went to when testing timings vs frequency

it shows that 1600c8=2133c11 in performance so you need to hit a balance between the two going all for one may not yield great results

Faster RAM makes the game run better. To "apply" it you'd have to buy faster RAM.

or overclock it but not all ram overclocks well and you want to make sure its stable with something like memtest86+ usb before booting into windows as unstable ram can corrupt your os install

ram sub timings still confuse the hell out of me after years of overclocking so for a noob it could be a little daunting

[griffz 1]

read it again

they also run some tests with a 6950 and some more with 5970 & 5870 for tri-fire which sees up to 13% gain moving from 1333c9 to 2133c9\2400c10 but the reason i linked it wasnt for the performance gains but due to the lengths they went to when testing timings vs frequency

it shows that 1600c8=2133c11 in performance so you need to hit a balance between the two going all for one may not yield great results

yeah my bad , for crossfire 2 and 3 card , there is some performance improvement . especially with minimum framerate. up to 20%.

But all that has been ran with 1360x768 display resolution. we can't extrapolate that to 1920x1080 or higher !

that said, the single card analyse show what i meant :

The final discrete GPU test shows a small 5% difference from 1600 C11 to 2400 C11, although other kits perform roughly in the middle.

5% average max difference, and 0% difference with "sleeping dog"...

ps: Do not overclock the RAM. either you will kill it with overvoltage, or you will broke your operating system.

[arotished 10]

When the FPS is so terrible as it is in ARMA, no wonder small changes gives a higher % increase...

[dasa 20]

But all that has been ran with 1360x768 display resolution. we can't extrapolate that to 1920x1080 or higher !

that said, the single card analyse show what i meant :

5% average max difference, and 0% difference with "sleeping dog"...

ps: Do not overclock the RAM. either you will kill it with overvoltage, or you will broke your operating system.

'

you left out that nobody will notice the difference between 170fps and 190fps

like i said they are testing console ports which are normally totally gpu bottlenecked and the way they make it more of a cpu limit is to drop the res but with a single gpu even dropping the res isnt enough to move the bottleneck away from the gpu with those crap games

while it in no way shows any real world usage benefits it is still a good ram benchmark that i believe shows close to the correct % differences you would see in a game like arma that is cpu limited even at higher res from tighter timings

my ram is rated 1600 11-11-11 1.25v made by samsung but the same chips were used in ram sold by other brands at 2400 10-12-12 1.65v

i run it at 2133 10-10-10 1.425v so overclocking ram can be fine but it does depend on what chips it uses

[Maxon 1]

Thanks for this Test!

i have 1600-9-9-9-28 and ordered just now because your posting 2400-10-11-10-30 and will bench. If it will have this gain, then i am very happy with that. :)

[dasa 20]

Thanks for this Test!

i have 1600-9-9-9-28 and ordered just now because your posting 2400-10-11-10-30 and will bench. If it will have this gain, then i am very happy with that. :)

i look forward to your results

what is your hardware?

i cant even find 2400-10-11-10-30 ram

if you wouldnt mind i would like to see benchmarks with old ram then the new ram at the same speed as the old ram and then its rated speed

thanks

[Maxon 1]

My Gaming-Systems are:

System 1: MSI X79A-GD65 - Intel Core i7 3820 @ 4,5Ghz - 16 GB DDR1600 "GEIL" - AMD Radeon HD7970 OC

System 2: ASUS M5A97 R2.0 - AMD FX-8350 @ 4,2Ghz - 8 GB DDR1333 "Mushkin" - AMD Radeon HD5870 OC

System 3: XFX X58 - Intel Core i7 920 @ 3,6Ghz - 6 GB DDR1600 "OCZ" - AMD Radeon HD6970

so i can do some test (if i find time for that) but the new memory should take place in System 1.

Sorry i have done a mistake while writing the latency, the right one is: Team Group Xtreem Series White, DDR3-2400, 10-12-12-31.

I have already run the Stratis and Altis Benchmark, make some notes, and will do this again with the new memory.

[galzohar 31]

you left out that nobody will notice the difference between 170fps and 190fps

like i said they are testing console ports which are normally totally gpu bottlenecked and the way they make it more of a cpu limit is to drop the res but with a single gpu even dropping the res isnt enough to move the bottleneck away from the gpu with those crap games

Good point there. Plus, if a game bottlenecks the GPU in pretty much every possible configuration, why the hell should anyone cares how it could scale with better cpu/ram/etc in some hypothetical unrealistic situation?

Edited February 14, 2014 by galzohar

[mattldempsey 10]

So 2133 seems exceptionally better than 1600.

Seems to be the best option for those without enless money, or has anyone seem significant gains between 2133 & 2400?

[Maxon 1]

Hello,

here my first Result. Tested withSystem 1: MSI X79A-GD65 - Intel Core i7 3820 @ 4,5Ghz - 16 GB DDR1600 "GEIL" - AMD Radeon HD7970 OC

changing the Memory to 2400-10-12-12-31 give a performance increase of 4% (Altis)

I have to say this System has Quad Channel Memory interface, maybe the gain is lower then expected. At the moment im benching System 2 with Dual channel.

Oh wow i got the Test from my other System:

System 2: ASUS M5A97 R2.0 - AMD FX-8350 @ 4,2Ghz - 8 GB DDR1333 "Mushkin" - AMD Radeon HD5870 OC

changing the Memory just to 1600-9-9-9-28 will give a performance increase of 9% (43 vs 47 FPS).

Test with 2400 Memory have given performance increase to 50 FPS. If i took a look from 43 now to 50 FPS then this is a big increase!

Important for AMD is Ganged Mode in Memory!

Edited February 16, 2014 by Maxon

[dasa 20]

im surprised your even seeing 4% gains on quad channel ram

interesting to see it is there though thanks for the reply

MattLDempsey

2133 9-11-10 and 2400 10-12-12 often cost about the same amount its just my old cpu cant run 2400 so i cant test it but as you go faster the gains do drop of as can be seen by the big jump Maxon got from 1333-1600 then less from there on

the rate at which it drops off depends on just how starved for bandwidth the cpu is and it becomes more starved the higher you oc

Edited February 16, 2014 by dasa

[dasa 20]

i have run some new tests on ram timings but there is obviously some inconsistency with the tests since i only bothered to run them all once with 1600 8-8-8 performing the same as 9-9-9 in one test and 9-9-9 performing the same as 10-10-10 in another this is just a hickup

one 7970 is now dead so im back to a single gpu for these tests

minimum fps in altis was to inconsistent to bother including

current ram prices in au

2x4g

1333 9-9-9 $93

1600 11-11-11 $89

1600 9-9-9 $103

1600 8-8-8 $109

1866 9-10-9 $105

2133 11-11-11 $102

2133 9-11-10 $115

2400 10-12-12 $125

2x8g

1333 9-9-9 $152

1600 11-11-11 $175

1600 9-9-9 $203

1866 10-11-10 $199

1866 9-10-9 $205

2133 11-11-11 $205

2133 9-11-11 $225

2400 10-12-12 $229

Edited February 18, 2014 by dasa

[dnk 13]

You've got a 4.9GHz here, which the vast majority of players don't have. It also doesn't look like there's much of a difference between 1600 and 2133.

I mean, the difference in latency between the fastest and slowest is a 38% reduction, which goes with a 17% reduction in avg frame times and 20% for minimum frames (meaning roughly a 50% ratio between RAM speed increases and FPS increases, which is still significant enough to justify a new purchase for players with old slow RAM).

However, it seems like once you reached a certain threshold for your system (1600/C10) your minimum frame times didn't improve with anything faster, while your averages continued on improving. Now, if you have a much slower processor (most players), that plateau point for min FPS might be considerably lower, and the avg FPS increases might be less than here (50% of RAM speed increase) as well.

So would a 3.9GHz or 2.9GHz require even 1600MHz RAM for best performance, or would 1333 or even 1066 work just fine for each, respectively?

[dasa 20]

i have my doubts about it being worth upgrading the ram over old ram just for the extra performance but it sure looks worth the extra cost to me for people that need new ram anyway

the older tests i ran back on the alpha i also tested at 3.8ghz http://forums.bistudio.com/showthread.php?153524-CPU-VS-RAM-Performance-amp-CPU-Threading-Benchmarked

my cpu is clocked high but its 3 years old now i guess the performance of a i7 4770k is roughly equal to my 2600k@4.3ghz and at ~4.3ghz a 4770k would be about as fast as my 2600k at 4.9ghz

the minimum fps are very random i wouldnt read to much into them i would need to average them over several runs to get a clear picture

the difference between 1600c9 and 2133c9 looks about as big as the jump from 1333\1600c11 to 1600c9 to me ~5fps in stratis and 4-5fps in altis

edit

here is the 1600c10 vs 2133c9 runs so you can see where the min fps are

edit again oops got them around the wrong way in that last graph stratis=altis and vice versa

Edited February 18, 2014 by dasa

[Maxon 1]

@dasa

how do you create this graphics? have you one sequence you will try every time? and then? how do you generate this graphic?

thank you.

My System 2 runs with the DDR1600 RAM much much better then with the slow DDR1333. and the ganged mode gives an additional increase of 3 FPS

[dasa 20]

run the benchmark altis & stratis found in the workshop

record the fps with fraps

chart them with libreoffice calc

[lao fei mao 21]

I still can't figure out why you guys are still persuing high FPS on ARMA, it's meaningless. IMHO, a stable 30 fps is far important than a fluctuation between 30~100+.

Buy a ssd and use the MSI-afterburner to cap the fps at 30, then you will feel really smooth no matter in sp or mp mission. If you obtain such notion, you don't even have to buy a luxury video card or some stupid SLI or crossfire thing, a 150USD one is good enough, and 2133hz RAM are totally unnecessary, 1600HZ is good enough. save the money to buy a 100usd 128GB SATA3 ssd.

[Maxon 1]

Thanks Dasa!

@Lao Fei Mao,

I agree with you. but getting higher FPS will mostly give more min FPS. And the most interesting are not the FPS peaks. more the stable min FPS.

Did this tests on my System 2 and the MinFPS have increased by 4 FPS with faster Memory (DDR1333 before).

Tests with my System 1 (QuadChannel) gives an FPS increase of 2FPS (DDR1600 vs DDR2400)

[SYH 10]

Stable 30FPS.

It's as almost as near to slideshow.

At least from what I can see.

If I can't get a stable 50-60FPS in the 2013-2014 game with my hardware it's a past tense in my book.

[Punisher5555 0]

Stable 30FPS.

It's as almost as near to slideshow.

At least from what I can see.

If I can't get a stable 50-60FPS in the 2013-2014 game with my hardware it's a past tense in my book.

You're still streaming the game from a Hard Drive. You have to increase that speed either with a fast SSD or utimately a RAM Drive.

My system has 1600Hz Ram, but I STREAM the Addons directory from a RAM Drive. The game is smooth as silk. I run 10,000 view distance.

Remember everything is CONSTANTLY streaming from that directory to the CPU. You have to increase that speed. And nothing is faster than RAM.

*1. Player sees chicken. 2. Goto Addons directory to stream in animals.pbo...........

Edited February 27, 2014 by Punisher5555

[dasa 20]

its a shame i dont have enough high speed ram to setup a ram drive for benchmarking arma as i suspect the head parking on my seagate 2tb 7200rpm may be throwing in the odd stutter despite the ssd cache which makes it very hard to get consistent min fps in my tests

i did try a ram disk when i had 16g 1600mhz ram in and it didnt really effect the average fps

edit

i have just run some similar tests on thief for anyone that is interested

http://forums.atomicmpc.com.au/index.php?showtopic=55771

Edited February 28, 2014 by dasa