Thanks Stasio :up:
What you after the Pi time using XMP or what the board sets like below? Bit strange btw I got a few more apps to read SPD & they all slightly different lol
http://i106.photobucket.com/albums/m...3C8SPDXMP1.png
Printable View
Thanks Stasio :up:
What you after the Pi time using XMP or what the board sets like below? Bit strange btw I got a few more apps to read SPD & they all slightly different lol
http://i106.photobucket.com/albums/m...3C8SPDXMP1.png
Seems I should reinstall lost ~0.4s, or.... 0.2, but made a few improvements (tertiary timings)
Tried that asus memtweakit, which did not work, and then installed the ASRock tool, which didnt allow adjusting timings.... Removed both, but seems OS is a little hurt...
Anyhow, my effeciency in that ASUS tool is listed as 39000 for ram, not sure if it blows a little, or much :)
Also, perhaps I should read up on what the stuff I change actually means, no clue what C2 and C6 package actually is, only that C6 seems to be a little more effecient..
EDIT: Had to bump the maxmem alot, just to get this tiny improvement...
http://img411.imageshack.us/img411/9456/screen031.jpg
EXperience update; I get alot more consistant results when disabling hyperthreading...
But in my case, I tend to get crashes more rapidly, had to increase vdimm a tiny bit to get stability..
I showed this one to hicookie earlier, but think some of you might find it interesting...
No CW
http://img829.imageshack.us/img829/2636/screen032p.jpg
CW
http://img100.imageshack.us/img100/9417/screen033u.jpg
Minor change in my setting, tras to 26, and maxmem to 1024 (yes, I need to reinstall OS^^)
M.Beier - 6:20:797 - 48*104.1 - 1250mhz - 8-11-8-26-78 1T - ASRock Z77 Extreme6 - Corsair Dom GTX3 PSC - 1.835v - WC on CPU, 120mm on ram.
http://img535.imageshack.us/img535/4384/screen034w.jpg
lol 0.001s behind the person I was trying to catch :D
EDIT;
Now the secret is WHAT did I improve here? :p:
M.Beier - 6:20:579 - 48*104.1 - 1250mhz - 8-11-8-26-78 1T - ASRock Z77 Extreme6 - Corsair Dom GTX3 PSC - 1.835v - WC on CPU, 120mm on ram.
http://img41.imageshack.us/img41/24/screen035l.jpg
I found only the latest version of Tweakit worked for me.
The Tweakit efficency thing imo is nothing to tune for. I can make over 100000 & run 32M by lovering DRAM refresh down to 1250 however the run is crazy slow.
Dunno but keep doing it :D
some crazy nice runs here, i didnt even see Cookies/Dinos latest runs, nice work guys
also a great run from Salske,
i got so much happening right now and ive promised a few people a few things first, but still hoping to do an assault on this challenge before it ends!
plays with some psc chip now, I can not get them 32m stable at 1400mhz C7, but here's a little video they boot up and run 1m :)
http://www.youtube.com/watch?v=HEpBT...ature=youtu.be
CKE#3 finally stable... Puuuh...
Rosty, jeg er næsten ikke misundelig på at kunne køre 1400!!! :)
Trans am, I am breaking OS soon, want to do my 'FUGGER' run, I am sure I've nailed how he made the score.... I need sound driver and WMP or internet access to keep up with him!
1400 cl7. mother effffffffffff!
Failed badly, cant get this CPU 5ghz stable on my win7 :(
Had it running in winxp, but without sound, as there isnt installed drivers, nor a sound card ;)
But; 6m 28.x sec while doing NDA-tweak during 32M :p: I am pretty sure that is what FUGGER did
EDIT;
Covered up the mighty tweak, sorry splave!
please do not discuss trololo tweak here, keep in private
Man am I happy for me this is cranking :D
C-N - 6:19:875 - 47x106.4 - 1276MHz - 8-11-7-22-70-1 - Gigabyte UD5H - GSkill C8 2133mhz 2x2GB/BBSE - 1.90v - Stock Air
http://i106.photobucket.com/albums/m...2M/6m19875.png
http://i106.photobucket.com/albums/m...9875AIDA64.png
ive had the tfaw argument before. im inclined to agree with you, it does give a boost, though theoretically it should just default to 16 as its lowest actual value.
I honestly wouldnt know but if so, it could be like CAS 1.5 in the days, improved stability in some cases :)
- And I am squeezing the maximum frequency out of the setup, as for RAM.... It can be that I have been unlucky with CPUs, but I think this particular sample of board is not the strongest out there, atleast.... DDR3-2600 is mission impossible, not even Ethan Hawk can do it...
But you know, on a second thought, I am doing it from BIOS, it might have impact on other subbies, though I've pretty much set everything manually... DRAM refresh interval and the 4. timings, are the only I've not set from BIOS manually - they might be be affected by other timings^^
Nice run Dumo!
Trying a new kit, 46.00 from Sears lol, didnt know they even sold memory. Had a slightly better run prior to this, but screwed up the screen.
OC Nub - 6:19.907 - 102x49 - 2448mhz 8-10-7-21-1T 1.89v Air - M5G - G.Skill Ripjaws F3-1600cl9 - CPU on runmc SS.
nice tommy, Dumo and CN. I'm gonna post my lastest milestone.
trans am - 6:26:203 - 50x100 - 2400.10 MHz - 9-11-9-24 1t - Gigabyte UDH3 – Gskill F3-17600cl7D-4GBPIS - 2200mhz 2x1gb/PSC - 1.85v – H20
forgot to take the AIDA. sorry. Honda, just cut me some slack i just want to post something in the meantime.
CN I see you using memtweakit on the giga. you got a link?
^ Thanks!
Try tightening your sub-timings.
Latency Is lower/better than my 2450 runs, I've seen way better just cant make it stable..... maybe its something to do with the 3570K :shrug:
I have played with tFAW a fair bit & believe lower works better for me..... on the above run it was set to 5 in BIOS & Memtweakit shows 10 as lowest
All the good stuff is here thanks to Stasio http://forums.tweaktown.com/gigabyte...ity-tools.html
Probably a misreport between software and BIOS for that. tFAW is the spacing between 4 ACT to the same rank and each ACT is spaced by tRRD which has a real minimum of 4 due to bus width (you can check the Intel public datasheet to confirm if you wish).
I did look up the Intel datasheet amongst other things the last time Raja mentioned tRRD/tFAW 4/16 earlier in this thread & although it could be a BIOS/Software report issue its less likely as the apparent performance increase has been found on both my Gigabyte & M.Beier's Asrock so I'm inclined to say there may be more to it.
I may re-test just to be double sure I am getting a boost but not that bothered as whatever works ..... works ay.
I did note tFAW too low @ 2600+ hurt performance so maybe it has a limit of effectiveness :shrug: