Digital Beats Posted December 30, 2010 Share Posted December 30, 2010 Hi i got Corsair Dominator 2x2gb - TWIN2X4096/CM2X2048-8500C5D XMS850X v1.2...PSC.F-Die.......128MX8 and i have some problems running beyond 1080mhz I have Gigabyte X48-DS5 and Q9450 Oc'ed >> 453mhz with ram 2.4B > 1087mhz with these settings, but i can't seem to tweak them right.. memory enhance: Turbo 5-5-5-15; advanced timings available: act to act delay (tRRD): 4 rank write to READ delay: 5 ((1) -i tried, 6,7,8,9,10) write to Precharge delay: 9 ((2) -i tried, 10,11,12,13,14) > but always like so ((1)read6>>(2)write10; (1)read8>>(2)write12... with at least 3-4 number space between each other) refresh to ACT delay: 68 Read to Precharge delay: 6 static tRead value: 8 (would like 7 if possible) static tread phase adjust: auto Command Rate (CMD): 2T ram voltage: 2.1v i tried 2.15v or even 2.2v but it didnt matter much.. mch: +0.300v from default and it fails windows memory diagnostics after 3-4run sometimes sooner.. i found out that those in bold are the most sensitive, read to precharge delay 6 should be ok, but i cant set other 2 right, maybe i need to tweak something else? any tips please. :): btw; atm im at 447mhz >> 2.4B ram at 1073mhz memory enhance: Turbo with these settings: 5-5-5-15 4 5 9 68 5 7 auto 2t ram: 2.1v mch: +0.275v Link to comment Share on other sites More sharing options...
Wired Posted December 30, 2010 Share Posted December 30, 2010 Moved to the overclocking area. Link to comment Share on other sites More sharing options...
Digital Beats Posted January 1, 2011 Author Share Posted January 1, 2011 ups sorry, i guess i posted in wrong section, thanks for changing it. edit: i think i found something, im testing with windows memory diagnostic tool as follows: default setting/cache- ON for all tests (its off by default); usually if its something wrong it fails at 23-28% 1-2nd round or 6 or 11-12th round.. anyway i tested 453mhz(fsb) with ram multi 2.4B = 1087mhz (2.1v) and it passed 22rounds with these settings on the picture.. hmm imo it looks 100% stable, well it happened that it failed two times at 21 round - 28%, with a bit different settings... http://i144.photobucket.com/albums/r163/ToxicfilteR/membiosinfo.jpg this is how it looks like with AIDA64 mem info http://i144.photobucket.com/albums/r163/ToxicfilteR/everest%20mem%20benchmark/memstRead81087.png if i change Rank write to READ delay above 7-8 it disappears from the settings -tWTR 13T >> 5T :confused: -tWTP is 8 in bios, real 16 -tRTP is 6 in bios, real 8 and bandwidth test; this is the best i can get with static tRead8, read speed and latency is a bit weaker compared to static tRead7 http://i144.photobucket.com/albums/r163/ToxicfilteR/everest%20mem%20benchmark/1087mhz.png and after it passed 22 rounds with static tRead8, i tried the same settings with static tRead 7 and it failed at 12 round:mad:..it took a long time though and that made me think it can be possible, i just need the right setting, maybe even more ram voltage, i used the same 2.1V? the thing is i've been tweaking for almost 8 months and i still can't bring this static tRead7 under control:whip: lol i hope that's enough info:D: Link to comment Share on other sites More sharing options...
Digital Beats Posted January 4, 2011 Author Share Posted January 4, 2011 weird, now its not stable anymore:mad: i tried a little lower Refresh to ACT delay - 64 and it started failing, raised to 66 still, raised back to default 68 and now it fails even on this one, i just don't get it anymore, it was fine 2-3 days..:wtfeh: Link to comment Share on other sites More sharing options...
Digital Beats Posted February 9, 2011 Author Share Posted February 9, 2011 Hi, just a quick update i updated my bios to F8H (latest) and i found out it wasn't stable anymore even at 1068mhz.. so now i used automatic SPD info and found out what's the cause, well it failed even at auto with these settings 5-7-7-25 advanced: 5-5-9-70-5- static tRead 7 - auto- 2T it fails by either -plus LRAD or -Stride6 (cache enabled) that's those 23-28% by windows memory diagnostic, im saying this because it passed with mem test86+ 4.2 even at 1090mhz, while in windows memory diagnostic tool failed at 1st round 23%. I decided to use 450mhz fsb and 1080mhz ram with +0.35v and mch +0.300v and now it finally works with these settings 5-5-5-15 adv: 5-6-9-70-6- static tRead 7 -auto-2T mem latency http://i144.photobucket.com/albums/r163/ToxicfilteR/everest%20mem%20benchmark/cachememoc36ghz.png although those 1087mhz would be nice too but it fails if i use memory enhance - turbo, same settings but memory enhance- standard its ok up to 1092mz, although copy speed suffers and is allot slower ~8400mb/s compared to 1080mhz 9600mb/s.. anyway im happy with this :sunglasse Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.