There are only a handful of C2D users who have LinX-stable systems. If I hadn't sold my E8500 I'd have been in this club :D
Printable View
There are only a handful of C2D users who have LinX-stable systems. If I hadn't sold my E8500 I'd have been in this club :D
Oh ya some notes on mine. 12GB ram, and little 2x220 hot rad
Quick OC question:
Once you've reached the point where increasing vcore and v-uncore doesn't help anymore, what else do you guys tweak to get higher?
I'm stuck at 4.2 GHz, which is rock stable with 1.275v LLC on.
But I can't get 4.3 GHz to pass LinPack even with 1.400v LLC on.
It stops crashing if I give it enough vcore, but it will give wrong residuals for the longer runs no matter how much vcore and v-uncore I give it.
I haven't been able to get 4.3 GHz to pass more than 10min. of IBT on very high.
On the other hand, it can take prime95 for at least 10 min.
y-cruncher stress-test gives mixed results, since it runs hot enough to kill off turbo boost and clock down to 4.1.
Thx, I'll try it in the winter when the outside ambient temps are sub-zero Fahrenheit.
I bet its somewhat temp related. get it colder and bet it will be more willing
tried with HT off yesterday, made 4.8Ghz but the Gflops were really low, so don't know if my RAM timings needed to be loosened up a bit, because they are pushed at 4.7Ghz.. so I lowered the OC a bit and the Gflops wen't to a "normal" level.
Rig in sig.
http://i18.photobucket.com/albums/b140/hoss3430/13.png
You have really low temps for such clocks... Well done! :up:
Why HT off if you're not limited temperature wise, though?
What would be the temp difference in that case?
Or no way it can get stable then?
HT off usually gives 100-150 more on max clock, and at lower temps too.
HT off lowered temps a ton, and I had 61F-62F ambiant..:yepp:
I now run 24/7 @ 4.7Ghz well 223 B clock x 21 multi with HT on, but I wanted to see what HT off would do.. and it went 20 passes @ 229 x 21 but the Gflops went way down, so I think the RAM was being pushed too hard, and I didn't want to mess around with my timings much just for a temp OC.
Gotta say I totally agree with raising the passes from 10 to 20.
I was 15 passes in on a LINX run last night, 4.6ghz @ 1.48v. Needless to say my screen suddenly turned orange and the computer locked up lol.
My daily 24/7 settings on water, all fans at 800RPM. it is 24C ambient. :D
http://i645.photobucket.com/albums/u...46G-linx25.jpg
no problem, PCI-Express frequency is 102.
http://i645.photobucket.com/albums/u...5v46G-sett.jpg
this chip can do 1.34v 4.6G in WinXP, but must 1.37v in Win7 :rofl:
http://i645.photobucket.com/albums/u...st/134v46G.jpg
Here's my effort 4.566GHz. HT off- couldn't quite get it stable with HT on, but hopefully I can still join the club;)
[IMG]http://img22.imageshack.us/img22/841...20xlinx.th.jpg[/IMG]
EVGA Classified
Heatkiller 3.0 (Cu) cooled by peltier chilled water (keeps water temp < 25C during LinX run)
http://i921.photobucket.com/albums/a...le_45_1280.jpg
Ram is only at 1286mhz. My ocz kit refuses to be linx stable with the 2:8 multi. Timings are tight however at 6-6-6-18 and 1.5v to vdimm. I've tried dramatically loosening the timings and raising the voltage as high as 1.75, but no go with higher speeds.
VTT is 1.35v, and I've sinced managed to get it stable with 1.354v vcore. 4.6ghz is stable as well, but requires considerably more voltage. I'm stuck at 222 bclk though. I cannot even get bench stability above that.
Not sure if I can update mine, but I got my vcore lowered a decent amount. 1.418v now, instead of 1.44v.
http://members.cox.net/tjgjkt/4.5stable.png
w00t w00t guys, some awesome chips in this thread, congratz to all owners!
My 3848A352 w3520 would like to join, unfortunately I am running a DFI JR at the moment, which cannot cope with the current and shuts down due to PWM overheating even at 4,4Ghz - 1,3V :ROTF:
Will join you guys as soon as I get my E760 back from RMA, here's a teaser though, memtest stable at ~1,38V VTT / 1,72V Vdimm (Samsung HCF0's) ;)
http://database.he-computer.de/Bilde...520_winrar.jpg
My e8400 on high vcore but it passed.
http://img152.imageshack.us/img152/6...hzlinx8400.png