tonedog12 2 Posted February 26, 2011 hi, ive read that a good way to test if your cpu is bottlenecking your system is the following. test the game at an extremely low resolution. If your framerates improve, then CPU bottleneck is not your issue. If framerates stay the same, then CPU bottleneck could be your issue. is this true? Share this post Link to post Share on other sites
pauliesss 2 Posted February 26, 2011 I don´t know anything about this "test", but what I know for sure, is that you you can have GTX580 and for example C2D E6850@3.6Ghz(I had this cofiguration) and you will be bottlenecked. Then I bought Core i5-2500k, OCed at 4.5Ghz and my FPS went up dramatically. It did not help only with FPS, but also with stuttering etc. I bought GTX580 before the release of i5-2500k because of low price at that time. Share this post Link to post Share on other sites
Gunter Severloh 4051 Posted February 26, 2011 Hi, There is a thread for computer questions here: Will my PC Run this? What CPU/GPU to get? What settings? System Specifications. http://forums.bistudio.com/showthread.php?t=72181 Share this post Link to post Share on other sites
Leon86 13 Posted February 26, 2011 In arma it doesn't really work that way with resolution, because the object drawing distance is lowered when resolution is lowered. A good test is turning anti-aliasing on and off. That's 100% gpu dependant. Share this post Link to post Share on other sites
cotabucky 10 Posted February 27, 2011 (edited) I don´t know anything about this "test", but what I know for sure, is that you you can have GTX580 and for example C2D E6850@3.6Ghz(I had this cofiguration) and you will be bottlenecked. Then I bought Core i5-2500k, OCed at 4.5Ghz and my FPS went up dramatically. It did not help only with FPS, but also with stuttering etc. I bought GTX580 before the release of i5-2500k because of low price at that time. Had the opposite results with my I7 2600k and 6970. When I got my gtx580 with my 3.0 ghz-3.3ghz I7 950 has no stutter or at least no bad stutter. I wouldn't be so sure.That gtx580 would make almost any system rock imo. I think if the system jives it jives. My particular system(I sent back) was the i7 2600k with a faulty asus pro board .Come to find out it was my 6970 causing the issues with stutter in ARMA2.WHY??? I don't know because my friend who has the identical system had no issues.I had some wierd stuff happeneing with my 6970. I don't think that bottlenecking test you mentioned would work because I have tried every resolution there is with my 6970 and i7 2600k and it seemed the lower resolutions suffered more than the higher resolutions when it came to performance. Bottlnecking at 4.4 ghz??? I don't think so. I'm just saying my system had serious issues all around(motherboard/cpu failing and in the end a suspect bad 6970).There can be so many reasons why performance may suffer.I built a complete new rig(cpu/MB/SSD,etc) and the only thing that changed anything was going down in cpu speed and using my gtx580.That made the absolute biggest performance increase.I know I would have better performance with the newer I7 2500-2600k but it wouldn't be by much that much I do know because I have owned both and my friend has the system I owned before and we have compared our benchmarks with 6970's with both i7 2600k OC' 4.4 and my current i7 950 with same card.Not much difference between the i7 950 and i7 2600k.The gtx580 made the biggest difference.I may be off on all my facts but not my end results. jmo Edited February 27, 2011 by cotabucky Share this post Link to post Share on other sites
bushlurker 46 Posted February 27, 2011 I had an interesting experience while comparing before and after benchmarks during my old recent system upgrade... I did it in two stages, so it highlighted the whole cpu/gpu thing nicely... My previous system was an AMD x64 3.0Mhz / nvidia 8800GTS 320mb - 1920x1200 res in all cases... (all video options on low/normal) With this I was getting low 20's fps in benchmark 1 & E08... on that wierd non-graphics intense benchmark 02 it was 8 fps!!! Not brilliant... As an experiment - I did a fairly hefty OC on my 8800 - ran the benchmarks again... This time I scored mid-high 20's/ on benchmarks 01 & E08... definitely an improvement... that OC definitely had an effect...but... Not a single FPS increase on that benchmark 02... no matter what I set the graphics to - benchmark 02 says 8 fps... clearly graphics ain't playing much of a part in that one... The first part of my upgrade arrived... an i7 950, 6gb of ram and a matching mobo... I installed all that - still with my old 8800GTS, and ran the benchmarks again - with the same mixture of low/normal video settings as before... Interestingly - there wasn't that much improvement in benchmarks 01 & E08 - I was up there in the low 30's fps - an improvement, but not a gobsmacking one... In that wierd benchmark 02 it was different... 8fps before - 26fps with the new CPU!!!! The final stage of my upgrade appeared a couple of days ago - a shiney new gigabyte 560 OC... I installed that - cranked every video option to "very high" except PP & AA, and ran the benchmarks again... As you'd expect - benchmark 01 & E08 freaked out - a solid 60fps in the first, 58fps in the second... brilliant!!!! just what I was hoping.... However.... once again - benchmark02.... 26fps! - not a single fps of difference between the 8800GTS 320mb, and the OC'd 560 GTX 1gb.....!!! So.... long story short... I'm not sure exactly what benchmark 02 is supposed to actually measure - but I've found it to be a great relative indicator of general CPU capability for running the game... it doesn't seem to care about graphics cards, or graphics settings much - all it seems to care about is CPU... B Share this post Link to post Share on other sites
