Hondacity Posted February 28, 2012 Posted February 28, 2012 hahahhaha after this 3d comp, i'll be trying my hands on some raid for some 2d points Quote
K404 Posted February 28, 2012 Posted February 28, 2012 I'm still not a PCMark05 guy, but I had an SSD to test so I started playing a little and the info in this thread has helped some of my scores quite a lot Thankyou! Quote
saint19 Posted February 28, 2012 Posted February 28, 2012 Lol, sweet in near to me ( at least in same continent) but hondacity is different. Well, I need bench 1st to know if i can win something. Quote
I.M.O.G. Posted March 2, 2012 Posted March 2, 2012 (edited) I don't have direct comparison with more than 4 cores with exact the same settings.Just found PCMark04 results with FX-4100 @ 6918 MHz and FX-8120 @ 6782 MHz: 4-core: 49.3 web page rendering, 317 fps WMV - http://hwbot.org/submission/2231053_ 8-core: 55.2 web page rendering, 362 fps WMV - http://hwbot.org/submission/2240067_ Don't look at Overall score, FX-4100 result was done with wrong Grammar Check. Windows was updated with AMD Bulldozer Performance Updates: Windows6.1-KB2592546-x64.msu Windows6.1-KB2645594-x64.msu Windows6.1-KB2646060-v3-x64.msu Do you use WMV settings suggested in PCMark05 Tweaking Thread? I mean this: "Force NumThreads"=dword:00000002 Try playing with other values Thanks. I did set the registry value for the number of threads on fx-8120. This was hard set to 8 threads (just look at audio/video): http://hwbot.org/submission/2246975_i.m.o.g._pcmark_2005_fx_8120_39611_marks This was hard set to 4 threads (just look at audio/video): http://hwbot.org/submission/2259791_i.m.o.g._pcmark_2005_fx_8120_42727_marks The 8 threaded score was better, but I believe that was the result of Vista over Win7 and I was running higher clocks on the 8 threaded submission (I spent more time on it, and higher clocks were easier on that motherboard). That isn't a good comparison because many settings as well as the OS was different. I don't personally have any direct comparisons with similar settings either, but from running the bench a few dozen times on the fx-8120, it seemed to me that scaling stopped more or less at 4 cores. That said, I don't believe the encoder actually supports splitting out more than 4 threads (reference here: http://www.microsoft.com/windows/windowsmedia/howto/articles/codecadvancedsettings.aspx - that isn't talking about wmencoder9, but even a newer version). Seeing as how it is doing a video and audio encode at the same time though, I suppose the video encode should be able to use 4 threads itself, and perhaps the audio can use several of its own... I don't know. Looking at scores however, the best 2600k scores are done with only 4 cores active and no HT. This tells me that its likely scaling stops at beyond 4 cores again. Not sure, just my inferences from my own benching and looking around. Edited March 2, 2012 by I.M.O.G. Quote
Crew pro Posted March 4, 2012 Crew Posted March 4, 2012 its funny how time changes things, 18-24 months ago most of these were considered cheats, now its all gravy, im very happy that this benchmark is progressing.. a tweak applied by one man is a cheat, a cheat applied by everyone is a tweak Quote
knopflerbruce Posted March 4, 2012 Posted March 4, 2012 its funny how time changes things, 18-24 months ago most of these were considered cheats, now its all gravy, im very happy that this benchmark is progressing.. a tweak applied by one man is a cheat, a cheat applied by everyone is a tweak If only one man knows the trick and it seems to good to be true, then the correct way of dealing with it is to assume it is what it looks like - something that fucks up the benchmark. Quote
I.M.O.G. Posted March 4, 2012 Posted March 4, 2012 On this topic, pro you had posted results on XS of replacing the browser. Scores of 60+ and 100+ on web page rendering. You never submitted to hwbot, as you knew it wasn't hwbot legal... but just out of curiousity, I tried to do browser replacement stuff to understand how you got those scores. I was never able to replace/rename iexplore.exe and get another browser to actually run the test. Any tips so others can understand how that was done? Quote
Crew pro Posted March 4, 2012 Crew Posted March 4, 2012 you dont need to replace web browser for 100+ pages/sec, ie9 is more than capable of it Quote
I.M.O.G. Posted March 4, 2012 Posted March 4, 2012 (edited) you dont need to replace web browser for 100+ pages/sec, ie9 is more than capable of it Are you saying you've done it? I haven't seen any results or discussion anywhere about it. When you did 60, you said you were using a different browser: http://www.xtremesystems.org/forums/showthread.php?240156-PCMARK05-testing&p=4135651&viewfull=1#post4135651 You didn't say anything when you did 110 on web page rendering a few days after that: http://www.xtremesystems.org/forums/showthread.php?240582-AMD-claims-a-global&p=4144958&viewfull=1#post4144958 Only one person doing 60+ on hwbot currently in web page rendering: http://hwbot.org/submission/2218295_mtech_pcmark_2005_core_i7_2600k_50458_marks Several people around 50+, which these tweak will do on Sandybridge: http://hwbot.org/forum/showthread.php?p=147431 Not saying you are wrong, and not challenging you. Your comment was vague, looking to understand. Edited March 4, 2012 by I.M.O.G. Quote
Crew pro Posted March 5, 2012 Crew Posted March 5, 2012 Are you saying you've done it? I haven't seen any results or discussion anywhere about it. sure When you did 60, you said you were using a different browser: http://www.xtremesystems.org/forums/showthread.php?240156-PCMARK05-testing&p=4135651&viewfull=1#post4135651 You didn't say anything when you did 110 on web page rendering a few days after that: http://www.xtremesystems.org/forums/showthread.php?240582-AMD-claims-a-global&p=4144958&viewfull=1#post4144958 can be done with a different browser or as ive recently worked out can be done with ie9 also Quote
saint19 Posted March 13, 2012 Posted March 13, 2012 (edited) Thanks to Christian Ney for create the thread and all the guys for sharing the tweaks and secrets, I did some quick test the last weekend and I had a jump of ~9k points. I was 100% sure that you can't disable cores until Massman confirmed me that you can do it but still competing in the same category of your CPU with all cores actives. Maybe I try again this, I'm just ~4k away from the 1st place in the rank of my CPU. Thanks again guys Edited March 13, 2012 by saint19 Quote
rebelion Posted March 13, 2012 Posted March 13, 2012 Thanks to Christian Ney for create the thread and all the guys for sharing the tweaks and secrets, I did some quick test the last weekend and I had a jump of ~9k points. I was 100% sure that you can't disable cores until Massman confirmed me that you can do it but still competing in the same category of your CPU with all cores actives. Maybe I try again this, I'm just ~4k away from the 1st place in the rank of my CPU. Thanks again guys But for PCMark more cores are better, so don´t disable it! Quote
saint19 Posted March 13, 2012 Posted March 13, 2012 But for PCMark more cores are better, so don´t disable it! I have a six core, until I read, the benchmark scales good until 4 cores. Quote
rebelion Posted March 13, 2012 Posted March 13, 2012 I have a six core, until I read, the benchmark scales good until 4 cores. Yes but someone pointed some post ago that some subtest scales with extra cores. Try it and let us know, please. Quote
saint19 Posted March 13, 2012 Posted March 13, 2012 Yes but someone pointed some post ago that some subtest scales with extra cores. Try it and let us know, please. As soon as I assembly the rig again, I will test that and post results. Quote
The Mutt Posted March 28, 2012 Posted March 28, 2012 Does the Bob tweak work every time for you guys? I sat all night yesterday and it took me around 40-50 trys before it worked, am I missing something or is it just my crappy old OS thats bugging? I'm going to do reinstall but I thought I would test some tweaks on the old OS first. Quote
SteveRo Posted March 28, 2012 Posted March 28, 2012 ^^ it works for me 99% of the time, occassionally it doesn't but that is the exception. - what OS? I can comfirm Bob's tweak works in both W7x64 and vistax64. I think Bob said it does not work in W8. Quote
The Mutt Posted March 28, 2012 Posted March 28, 2012 W7x64, I will reinstall and try again tomorrow. Quote
bob80 Posted March 28, 2012 Posted March 28, 2012 It doesn't work on W8....I did only some tests and it didn't work at all. For me it works every time Quote
K404 Posted March 28, 2012 Posted March 28, 2012 Not the best place to ask.... but i'm gonna go for it anyway I've been kicking my Celeron 352 around in PCM05 and it looks *alright* EXCEPT.... for MT Test 2- Image decompression. The score is in the toilet. Sometimes it looks like it's failed, but I get a score.... a really bad one (<2 MPixels/s) Any ideas? Quote
I.M.O.G. Posted March 28, 2012 Posted March 28, 2012 k404, its been a few weeks since I ran PCM05, is MTT2 the one that does text/uses fmpad? If so, you can go to properties of FMPad and set it to disable themes when it runs... See if that fixes the lousy score on that test. Quote
K404 Posted March 28, 2012 Posted March 28, 2012 (edited) Yea, the other part of the test is text edit Compatability -> Disable visual themes. Will try that first and see what happens My other PCM05 scores have normal MT Test 2 numbers, maybe it's an architecture thing, just needs rubbed a slightly different way Edited March 28, 2012 by K404 Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.