I am in love!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Dadofamunky

Platinum Member
Jan 4, 2005
2,184
0
0
Man, that new RealTemp utility is pretty nice. Appears to be more accurate with the Wolfdales and Penryns too.
 

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
20,882
3,230
126
Originally posted by: Zap
Originally posted by: aigomorla
and no one believed me when i praised the 9650 so much. :T

seriously, the 9650 has to be one of the best chips made for the LGA775, only to be beaten by a new X3380.

So, 9650 at new price of $340 or so is recommended?

no zap you must go i7
 

n7

Elite Member
Jan 4, 2004
21,303
4
81
I'm fighting with 9x475 right now & losing.

With my Q9550, i couldn't things stable over 465ish, as i'd get freezing during P95 or even just at idle occasionally.

I'm not sure if freezing of the OS is NB or VTT related or both, but i'm not getting anywhere @ 475 right now, other than freezing up within 10ish minutes...

Damn VTT/NB + GTL tweaking...
 

n7

Elite Member
Jan 4, 2004
21,303
4
81
Originally posted by: aigomorla
Originally posted by: Big Lar
Could be the ram...

+1

Not likely considering i can do all 8 GB @ DDR2-1050 & higher stably.
Old pic, but shows what i'm saying...

I just have to find the right mix of VTT + GTL & NB + GTLs...or back down to a lower FSB & give up
I could never get anything over 465 stable with the Q9550, even at a low multi where the CPU wasn't an issue, so i'm already doing a bit better with the Q9650.


...argh, just crapped out @ my latest attempt, 11 minutes into Large FFTs.
I'm not really getting anywhere it seems.


This is what worked for 9x470.

Q9650 @ 9x470 1.30000v 1.34vtt 0.68-0.68GTLs 1.58pll 1.36vnb 0.68NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Blend 4 hrs stable


Then i jumped to 475, & here's the notes of fail...maybe someone else can make sense of them, because i sure cannot.

Q9650 @ 9x475 1.30000v 1.34vtt 0.68-0.68GTLs 1.58pll 1.36vnb 0.68NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Large 7 mins then fail

Q9650 @ 9x475 1.30000v 1.34vtt 0.68-0.68GTLs 1.58pll 1.38vnb 0.67NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Large 31 & 34 mins FAIL workers 1&2

Q9650 @ 9x475 1.30000v 1.34vtt 0.68-0.68GTLs 1.58pll 1.38vnb 0.675NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Large - froze system 2 mins in

Q9650 @ 9x475 1.30000v 1.36vtt 0.67-0.67GTLs 1.58pll 1.38vnb 0.67NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Large - error 3 mins in worker1

Q9650 @ 9x475 1.30000v 1.36vtt 0.675-0.675GTLs 1.58pll 1.38vnb 0.67NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Large - froze system ~8 mins in

Q9650 @ 9x475 1.30000v 1.36vtt 0.68-0.68GTLs 1.58pll 1.38vnb 0.67NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Large - 37 mins error in worker1

Q9650 @ 9x475 1.30000v 1.36vtt 0.68-0.68GTLs 1.58pll 1.38vnb 0.675NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Large - 11 mins then error in worker1



This is what worked for 9x470; then i jumped to the below & am not getting anywhere.
 

Big Lar

Diamond Member
Oct 16, 1999
6,330
0
76
"Q9650 @ 9x475 1.30000v 1.36vtt 0.68-0.68GTLs 1.58pll 1.38vnb 0.67NBGTL 1.6PCI-e 1.2SB 1.96vDIMM Large - 37 mins error in worker1 "

Just something to try,Pci-e and SB on Auto.

NB GTL seems a bit high.

What Mode are you in, performance/optimized/compatible?

Also the PLL is a bit high as well as the NB. I found that less volts gets me further on this board/shrug.

You can try Clock Skews when you get closer, mine runs -100 on the Cpu and -200 on the NB in Performance mode. Just some stuff to try.

Larry
 

Rafael

Senior member
May 11, 2001
868
0
0
I was going to suggest the same thing as Big Lar, the clock skews. Maybe a first try with -100 on CPU and leave NB on auto. That helped some people to boot with Q9X50 with high FSB.

BTW, nice CPU you got there!
I hope I get lucky in a few weeks when I will buy mine!
 

n7

Elite Member
Jan 4, 2004
21,303
4
81
Wtf...thanx for mentioning the RAM guys...i wrongly assumed it would be fine...

I ran Memtest86+ while i was at work today, & it found a single error in the first pass in test 6, but nothing in the sucessive 9 passes.

I know it did not fail the last time i ran Memtest with my Q9550 @ 8.5x465 1:1 (DDR2-930), so either it's gone slightly bad, or it doesn't like running at 9x475 1:1 (DDR2-950).

I'm running it at 465 (DDR2-930) now to see what happens...

Thanx for all the suggestions guys

I'm already happy with getting at least 4 GHz as i mentioned, but the overclocker in me doesn't want to stop till i've maxed everything out as much as possible


Edit: At 465 (DDR2-930) already passed test 6 without error...i have a funny feeling i won't be able to reproduce this single RAM error, especially since it only did it once out of 10 passes...
 

n7

Elite Member
Jan 4, 2004
21,303
4
81
Gahhhhhhh. Exactly as i figured, i cannot reproduce any RAM errors.

465 1:1 (DDR2-930) = pass
475 1:1 (DDR2-950) again same settings = pass...lolwtf?
442 5:6 (DDR2-1051) = pass

Figures, phamtom RAM errors. :roll:
I'm pretty sure it's not really a RAM error though; it's like caused by FSB or NB settings, since otherwise, it makes no sense than i can fly thru just fine at a far higher speed like DDR2-1050 w/o issues.

edit:
I'm crossing my fingers for this to not error out a couple hours in

9x475 1 hour Large so far...
 

Big Lar

Diamond Member
Oct 16, 1999
6,330
0
76
The errant ram error can point to the NB GTL. Could just be the chipset being flakey due to being hot too.

With my board if I run prime blend at say 490fsb for a coupla hours, then reboot and run prime again, I'll error out fairly quick. I have excellent cooling, so I doubt thats the problem, I think its just the nature of the beast. I'd run Blend tho.

Larry
 

n7

Elite Member
Jan 4, 2004
21,303
4
81
Originally posted by: Big Lar
The errant ram error can point to the NB GTL. Could just be the chipset being flakey due to being hot too.

With my board if I run prime blend at say 490fsb for a coupla hours, then reboot and run prime again, I'll error out fairly quick. I have excellent cooling, so I doubt thats the problem, I think its just the nature of the beast. I'd run Blend tho.

Larry

I'm pretty much about to give up on fighting with this much more.

I've spent all week & last night all evening & early this morning rebooting trying things, & whenever i find a combination that works, it'll go for 2, 3, or 4+ hours but then it goddamn fails.

And it's sometimes on a larger FFT, sometimes a medium one like 160k, & even on 20k, & i know it's not CPU or RAM holding things back. So heat it may well be...

I've tried a zillion combinations of VTT + GTLs & vNB + GTL, & i basically always have to come back to settings around this area, as they at least get me a couple hours before errors.

CPU Voltage: 1.3
CPU GTL Voltage Reference(0/2): 0.680x
CPU GTL Voltage Reference(1/3): 0.680x
CPU PLL Voltage: 1.58
FSB Termination Voltage: 1.36
DRAM Voltage: 1.96
NB Voltage: 1.38
NB GTL Reference: 0.670x
SB Voltage: Auto
PCIE SATA Voltage: Auto

Load-Line Calibration: Enabled
CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
CPU Clock Skew: -100
NB Clock Skew: -200
CPU Margin Enhancement: Optimized


I've tried 1.32-1.36 range VTT along with GTLs from 0.63 to 0.71, & basically i have to stick w/ 1.36VTT + 0.68 & 0.68.
Any lower VTT is no good, & @ 1.36VTT, if i try lower GTLs or higher, or mix them up like 0.675 & 0.680, not good...has to be the same value, & seems to like being right around 0.925 (1.36 x 0.68).

For NB, increasing from 1.38 to 1.4 hasn't helped, & i need to recheck my notes to see if going from 1.36 to 1.38 was needed or not to get from ~465 to~475.

I do seem to need the higher NB GTL; lower doesn't work.
My guess would be due to my 8 GB.

It's been 1.38vNB + 0.67 GTL that get me a couple hours or more before errors, which is right around 0.925 too...both VTT & NB GTLs like that area, & whenever i try to lower that or increase that, doesn't work, so that much i've figured out.

I've tried upping vCore & vDIMM, but doesn't help, as those aren't issues.
I've loosened RAM clock twister from Moderate to Light, but not helping.
Loosened tRD from 10 to 11 or 12; not helping.
400 strap vs. 333 strap; not helping.
I forgot to try lowering PLL, but increasing to 1.6 didn't help. I really doubt lowering it will do anything, but i'll try that next i suppose.

Latest thing i decided to mess with was CPU & NB clock skews.
I've actually been using them @ CPU Skew -100 & NB Skew -200 for so long now i can't even remember why i started...it might have been your suggestion back with my Q9550, Big Lar

But i tried a whole bunch of combinations of them from AUTO up to -400.
Turns out the only way to actually get into the OS & then get more than seconds into P95 is at my usual CPU -100/NB -200, or edit: nope, system locked up 20 mins in.
So only CPU Clock Skew: -100 & NB Clock Skew: -200 are stable for at least over an hour.

Auto doesn't work!
Neither does higher combinations.
I'm guessing this is because i've always had them at -100/-200, & my VTT/NB + GTL settings have been set around them to work that way?

Anyway, thanx again for all the suggestions.

After wasting hours & hours trying to get over 465 truely stable, it appears i may very well be stuck at the same FSB i was back with my Q9550, which well, isn't too surprising really.

Just was hoping i'd have gotten a bit smarter by now & figured out how to outsmart my mobo...



Originally posted by: theprodigalrebel
I salute your jpeg naming convention.

Hah, glad you like it.
I keep track of settings that way in notes & the screenshots i take.
If i didn't note the exact settings i use i'd never get anywhere when OCing
 

Big Lar

Diamond Member
Oct 16, 1999
6,330
0
76
With PLL if I run 1.58 I error out, 1.56v is where it prefers it. Seems all these chips are individuals with a grudge on thier shoulders! Nice clock tho :beer:
 

Thor86

Diamond Member
May 3, 2001
7,886
7
81
Hmm, I'd say your VPLL and VNB are too low. Try 1.8 and 1.6 respectively to see if your P95 LargeFFT testing passes more than 4 hours.
 

Zapper48

Member
Oct 7, 2007
167
0
0
Ran 25 hours of Prim95 Blend with 8 gb of OCZ PC2 6400 at 471x8.5
LLc enable runs 1.268v LLC off I use 1.325v I must be lucky I guess.

EO stepping
Batch# L835B078
 

n7

Elite Member
Jan 4, 2004
21,303
4
81
4.2 GHz 8+ hours Blend stable

I used RAM @ auto everything, which puts tRD @ a ridiculous 13.
I then tried it at auto everything RAM except tRD @ 10 (what i could run for the Q9550), & it errored an hour in
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |