The Official Opteron 939 Overclocking Thread

Printable View

Show 100 post(s) from this thread on one page

10-19-2005, 11:37 PM
ben805

Quote:

Originally Posted by Nettwerk

Got my CABNE 0528GPMW from compumusic today :) I think mine would benifit the ihs removal but it doesnt break 48c load on good watercooling.

http://img381.imageshack.us/my.php?i...rown1488nr.jpg

Congrats on that awesome chip bro, just be very careful if and when you remove the IHS....$300 isn't exactly cheap, I still haven't decide what to do with mine when I receive them, the TT Big Typhoon will be here tomorrow as well so I"m not sure if I should stuck the massive heatsink on it to test it on air first, or put it straight under the PCICE phase change :D :D
10-20-2005, 12:14 AM
NYCTeeN82

Yes all the people that just recently got Opteron 148's from Ewiz are getting the CABYE 0528GPMW (and myself). Everyone who has this one, post your results. Your max stable results. Not 1M SUperPi, but Prime 12+ hours.

Mines 2805 is stable @ 1.4V. I can't seem to be able to do anything above this. 2850 @ 1.5V or 1.55V errors out in Prime. I hope 2805 isn't the wall. Temps are under 50C on Full load.
10-20-2005, 12:20 AM
eva2000

Quote:

Originally Posted by NYCTeeN82

Yes all the people that just recently got Opteron 148's from Ewiz are getting the CABYE 0528GPMW (and myself). Everyone who has this one, post your results. Your max stable results. Not 1M SUperPi, but Prime 12+ hours.

Mines 2805 is stable @ 1.4V. I can't seem to be able to do anything above this. 2850 @ 1.5V or 1.55V errors out in Prime. I hope 2805 isn't the wall. Temps are under 50C on Full load.

that's a weird one... what prime you running small/large ffts or blend ? try prime95 v24.14
10-20-2005, 12:28 AM
Rufus7

Prime Custom FFT´s In place every 15Minutes changing. Thats maximum Stress for the CPU!! Prime Blend only tests the Ram!!!! And Large is testing the Ram and the CPU!!
Best for the CPU and the hardest Test is Prime95 Version 24.11 and higher.
Ultimate CPU Test.
10-20-2005, 12:57 AM
mbm

I would like to hear from others who have the 146 CABGE. How high they can OC.
10-20-2005, 04:02 AM
D4nnyOcean

If someone were to order one today from ewiz what do you think the best chances of getting a good stepping are and what flavors 144,146,148?

And OEM or Retail? Anyone gotten a CABNE OEM?
10-20-2005, 04:43 AM
F|apDro|`

1 Attachment(s)

yay just got my opteron installed ...

Its CABYE :( but it was the best I could get !

Lets hope for some Mhz pwnage! I want it to clock better as my 3500+ that stopped at 2.6Ghz 1.75V :stick: ..

For now the opteron is like 15 C lower in temp idle then my NC wow :confused:
10-20-2005, 05:30 AM
eR1k

Quote:

Originally Posted by F|apDro|`

yay just got my opteron installed ...

Its CABYE :( but it was the best I could get !

Lets hope for some Mhz pwnage! I want it to clock better as my 3500+ that stopped at 2.6Ghz 1.75V :stick: ..

For now the opteron is like 15 C lower in temp idle then my NC wow :confused:

Where did you buy yours flapdrol? alternate.nl ?
10-20-2005, 05:32 AM
F|apDro|`

yep :) ordered it tuesday

still have to format my bench HD so I can tweak a little :slap:
10-20-2005, 06:36 AM
eva2000

testing 32M at 3007mhz and weirdess experience so far with 2x 512MB Gskill PC4400LE on 180 divider 32M would clonk out 2-3 loops in at 334HTT/300mhz 2.5-4-3-6 at 2.85-2.95v

so decided to try 2x 512MB Patriot PC3200XBL TCCD at 180 divider 32M and would clock out 1-2 loops at 334HTT/300 2.5-4-3-6 at 2.69v

so thought why not try 166 divider with mem at 273mhz 2.5-3-3-6 7-17-2223 with Patriot PC3200XBL

32M
at 2.69v = clonks out 2 loops in
at 2.76v = clonks out 4 loops in
at 2.79v = clonks out 5 loops in
at 2.85v = clonks out 6 loops in

so change from 7-17-2223 to 8-18-2223 loosening Trc and Trfc
at 2.85v = passed in 24m 29s!

http://www.fileshosts.com/DFI/NF4_SL...rpi-32m_tn.jpg

Prior to this revelation, i was upping vcore with no effect and it turns out it was vdimm that was lacking! So now i need to lower vcore and see what happens :)

Strange is at 300-312mhz 2.5-4-3-6 only 2.69v is optimal for this Patriot, but at 2.5-3-3-6 it handles volts fine!

Perc can confirm this weird experience regarding vdimm as he uses same ram!

edit: Can confirm so far, lowered vcore by 0.025v in bios and tad da!

http://www.fileshosts.com/DFI/NF4_SL...68vcore_tn.jpg

further proof

http://www.fileshosts.com/DFI/NF4_SL...pi-32m_tn2.jpg
10-20-2005, 06:40 AM
foofreaknfighte

Quote:

Originally Posted by ozzimark

i never saw mention about a fan on the top corner of the board. i have a 120mm up there to keep all the important mosfets and ram cool.. pwm is idling at 28c for me

Ok I have a theory I am going to throw at you guys, let me know if it sticks. I am wondering if my not being able to get my PC4000 ram above 215 has something to do with them overheating. I know they are in the same area as the PWM and my PWM is reading 60+c at load. Do you think there is a problem there?
10-20-2005, 06:51 AM
ozzimark

there's one way to find out. increase airflow to the point where you're not overheating and try to see if you can go further ;)
sounds like some of your mosfet sinks have bad contact or something :(
10-20-2005, 07:14 AM
foofreaknfighte

I have a another fan on placed on top of my heatsink facing the ram now. Seems to be cooling the PWM about 5c better. I have a feeling it is just the mosfet sinks though. I would reseat those but I heard they are soldered on.
10-20-2005, 07:20 AM
F|apDro|`

1 Attachment(s)

my out of the box result :o

not tweaked or stable :o
10-20-2005, 07:50 AM
foofreaknfighte

Quote:

Originally Posted by F|apDro|`

my out of the box result :o

not tweaked or stable :o

Really pushing that vcore huh. I don't think you will need it that high to get to 3003mhz.

My RAM cooling idea was a no go. The fan over my ram is keepingmy PWM cool though, not sure how important that is. What do you guys think the max temp for PWM is?
10-20-2005, 07:51 AM
ocZZ

not bad, I'm running at 2.95 at 1.55v with my CABYE 148 -- temps are still very high though.
10-20-2005, 08:04 AM
periquitos

Guys , let´s start to post the maxTcase value? already post mine on the tcase thread , but I think will be a good idea to post here (to many diferent cores/steppings on the another...)

I didn´t test my processor yet , all I did was set it to 2700mhz at default vcore and it appears to be very stable (primed by 1 hour and 4 hours rendering a work scene in 3dsmax6)

Please report the complete info (bath, year/week...)

OPTERON 148 CABYE 0528 GPMW

CPU Information (CPU #1):
Standard CPUID: Family: F, Model: 7, Stepping: 1
Extended CPUID: Family: F, Model: 27, Stepping: 1
CPUID String: 20F71
Processor APIC: 0
Processor: AMD Opteron 148 (San Diego)
CPU Speed: 2695.40 (245.04 x 11.0)
Revision: SH8-E4
Platform: Socket 939
Startup VCore: 1.400v
Maximum Case Temperature = 65C
10-20-2005, 08:07 AM
F|apDro|`

Quote:

Originally Posted by foofreaknfighte

Really pushing that vcore huh. I don't think you will need it that high to get to 3003mhz.

My RAM cooling idea was a no go. The fan over my ram is keepingmy PWM cool though, not sure how important that is. What do you guys think the max temp for PWM is?

neh dont think so ...

My mobo sux cus onboard LAN died cus of OC and since then it does some weird things ...

But as I said this is out of the box :)...

I Primed 260x11 @ 1.6V for 2 hours ... then it gave error ... Max Load temp = 47C ...

Now Priming again and letting it burn in and AS5 settle :). I know vCore = high but I need prime stable ;_;

edit:
my MaxTcase = 57C :(

I like that 65C more pera

Stepping:

http://www.xtremesystems.org/forums/...chmentid=38825
10-20-2005, 08:19 AM
el rolio

Quote:

Originally Posted by eva2000

testing 32M at 3007mhz and weirdess experience so far with 2x 512MB Gskill PC4400LE on 180 divider 32M would clonk out 2-3 loops in at 334HTT/300mhz 2.5-4-3-6 at 2.85-2.95v

so decided to try 2x 512MB Patriot PC3200XBL TCCD at 180 divider 32M and would clock out 1-2 loops at 334HTT/300 2.5-4-3-6 at 2.69v

so thought why not try 166 divider with mem at 273mhz 2.5-3-3-6 7-17-2223 with Patriot PC3200XBL

32M
at 2.69v = clonks out 2 loops in
at 2.76v = clonks out 4 loops in
at 2.79v = clonks out 5 loops in
at 2.85v = clonks out 6 loops in

so change from 7-17-2223 to 8-18-2223 loosening Trc and Trfc
at 2.85v = passed in 24m 29s!

[IMG]*snip

Prior to this revelation, i was upping vcore with no effect and it turns out it was vdimm that was lacking! So now i need to lower vcore and see what happens :)

Strange is at 300-312mhz 2.5-4-3-6 only 2.69v is optimal for this Patriot, but at 2.5-3-3-6 it handles volts fine!

Perc can confirm this weird experience regarding vdimm as he uses same ram!

edit: Can confirm so far, lowered vcore by 0.025v in bios and tad da!

[IMG]*snip

further proof

[IMG]*snip

WOW. i am so happy someone is experiencing this same crazy :banana::banana::banana::banana:!
quickly, lemme post my confused situation, and lookin ato both your comments and revelation and my situation it seems like the same issue!

Quote:

Originally Posted by elrolio

ok wow. so still tryin to OC this opteron. seems like ive run into a werid wall. with stock vcore i can go all the way up to 277x10. i have the memory on a divider (since no active cooling) at either 183 or 166 with timings of 2.5-4-4-9 and then 14 and 17 for the next two timings... anyhoo... no amount of voltage seems to get me stable at 280x10. it would then seem to me that there is some other issue happening. cuz even with the 183 divider puttin the mem at about 254 which i know its pretty stable at with 2.5-4-4-9 / 2.6v. and of course the 166 divider puts it at about 233. anyhoo. basiclally i can super pi 1M but none of hte larger ones and prime will fail anywhere from half hour to 5 hours. i pretty much stepped up the voce in bits:
ive tried startup +10%
1.375 + 10
1.4 + 10%
1.425 + 10%
1.45 + 10%

1.4 + 5%
1.45 + 3.3%

and note well that with "startup" voltage set and no over % set i went up to 277x10.

one thing i guess i can try is move from the 1.c standard and tccd bioses to a lower bios?
anyone here using an opterOWN with the neo2? which bios been best for you? since i have a CABNE chip right here.

ok. so basically ozzimark was saying something else must be holding me back. all mosfets are sinked, systool and everest shows me that system temps are decent. chipset at like 20°, system temp at like 40° + actually touching the mosfet sinks and that whole area around the socket is pretty cool and has airflow.

but yea, looks like eva had the same experience as me lasst night, only i wasnt able to figure that whole mem thing out since im limited to 2.6 essentially cuz i have no active cooling. hmmmz. any ideas?
oh and i had a buncha screenshots showin the weirdness of what was happening. will post em when i got home as i was still testin all the way till when i left for work.
10-20-2005, 08:34 AM
jahjahbinks

So what's the current XS status on benching those dual core opterons?
10-20-2005, 08:52 AM
foofreaknfighte

Quote:

Originally Posted by F|apDro|`

neh dont think so ...

My mobo sux cus onboard LAN died cus of OC and since then it does some weird things ...

But as I said this is out of the box :)...

I Primed 260x11 @ 1.6V for 2 hours ... then it gave error ... Max Load temp = 47C ...

Now Priming again and letting it burn in and AS5 settle :). I know vCore = high but I need prime stable ;_;

edit:
my MaxTcase = 57C :(

I like that 65C more pera

Stepping:

http://www.xtremesystems.org/forums/...chmentid=38825

I know what you mean. I am priming right now, hoping to be able to prime for 8 hours at 2.8ghz. I think I should be able to do that, and I guess that I could be happy with that...might pick up a CABNE if I see one though.
10-20-2005, 09:24 AM
F|apDro|`

1 Attachment(s)

Quote:

Originally Posted by foofreaknfighte

I know what you mean. I am priming right now, hoping to be able to prime for 8 hours at 2.8ghz. I think I should be able to do that, and I guess that I could be happy with that...might pick up a CABNE if I see one though.

well tried 3Ghz @ 1.6V via clockgen ...

it was picstable just as at 1.65V :stick:
10-20-2005, 09:40 AM
LORD

Here's a 150 quick results. I'll probably play with it some more later. At the moment there seem to be some issues with BH-5, the board or mem controller. Maybe bios and or setting? S&M MEM TEST seem to error 3/4 way through at speeds that would complete before ( when San Diego used) tested at 310x11 166 divider as well. Otherwise it's stable. Plays games etc. So will have to screw around with it later and see what's up. Doesn't seem to have any obvious startup probs with subzero. Can't say for Cascade type temps at the moment though...

I'm on a diff comp now so only have this jpg to show.:D

It also boots at 330+ with 10x multi etc.

http://lord.e-access.com.au/Opt150_Spi1mm_12x284.jpg
10-20-2005, 09:58 AM
foofreaknfighte

Quote:

Originally Posted by F|apDro|`

well tried 3Ghz @ 1.6V via clockgen ...

it was picstable just as at 1.65V :stick:

Try a non-clockgen when you get a chance. I had a lot more trouble keeping pi stable that way...
10-20-2005, 11:36 AM
NYCTeeN82

CPUID: 20F71
APICID: 0
Revision: SH8-E4
Precessor: AMD Opteron 148 (San Diego)
Speed: 2805.00 (255 x 11)
Platform: Socket 939
MaxTCase: 65C
VCore: 1.400V

Show 100 post(s) from this thread on one page