w00t, we've passed 2000 posts!
S939 forever and ever ;)
w00t, we've passed 2000 posts!
S939 forever and ever ;)
aaaaaaaand 2000 posts!!!
damn one was faster. :horse:
i think that i'm going to buy s.939 again...
s.939 in heart :D
congrats for keeping that thread alive guys... :clap:
This Opteron 185 runs with stock voltage @ 3200 mhz prime stabil :banana: it should do 3.4 ghz i hope....that will be awesome !
lol....its on single stage...
look at this:
http://www.abload.de/thumb/opteron185-0628bpmw229c.jpg
core temp reads out mess, the second core had -7c the first is right.
well seeing what you guys do made me want to have some fun with s939 :) I have a kit of TCCD-GSkills on the way, so I can start with CPUs at home, but what batches/steppings/markings should I look for to have the best of it? S939 CPUs are dirt cheap now so I can afford a few :) Any hints?
Is he seeking after dual core or single core ?
I had Venice 3000 stepping BB and it couldn't pass 2.7G no matter what I tried. Upped voltage to 1,7 but it didn't help anyway.
Right now I'm watching some SD 3700+:
E4 KACAE 0606EPBW
E6 LCBBE 0640CPMW
E6 LCBIE 0625FPMW (this one runs 2,65 @1,3v stable)
E4 KAB1E 0631APMW (this one does 2970 @1,48 on Epox 9NPA, so maybe DFI will clock it better)
and one 4000+:
E6 LCB9E 0641CPMW
Right now I could test them only under water, but is any of those worth playing with?
0631APMW should be good, but they normally do 3g @ 1.3x volts , or perhaps even lower in some cases. The E6 chips could be nice compared to other E6's, but they're usually not as good as the best E4's. KACAE chips aren't that great afaik. SOme suck, others are OK.
Sheesh guy's, You make me wanna upgrade my oldschool Socket A rig to a S939 system. :D
hmm.. i wouldn't buy no cpus anymore... but so cheap. should i?
http://images.qxlricardo.com/ImgUser...9790_2_Big.jpg
btw: vote for ripping.org :D thanks... i know it's cheap but i'm wondering if we could win.
http://imgplace.com/image_bin/2509/1...d95.JPG.th.jpg
Ive got a 3500+ that has seen 3280MHz on air before (max cpu-z verification) and benchmarked at 3055MHz at one time, code is CACAE 0611GPMW.
I guess this is different to Kacae.
hey ben get a better cooling :D
lol
i dont even attempt what you guys attempt, crazy stuff.
im only having air cooling - i did try and make it good air though by using a 80mm delta she fan on well, a ac cooling freezer 64, maybe i could have had a better hs but the fan on it sounded like a vacuum cleaner, just to get 3280MHz.
if I had a vapochill maybe id have seen much better results but I dont have much money.
now ive sold my dfi ultra d and my other socket 939 motherboard is an epox 9nda3i which goes upto 1.66v max so that is useless to get more out of the chip obviously.
ben
Look at this:
http://www.abload.de/thumb/8xdfilanpartygmo.jpg
:D
@ gianni
What do you want with an 3500+ newcastle ? playing with ln2 ?
hmm... http://valid.x86-secret.com/show_oc.php?id=371240 @ 1.62V @ SingleStage
Should there be more possible with more vcore ? I think so....because 1.62 volts isnt real sick. The key is 1.8 volts for the san diegos.
I see you're talking about our processor.
But it's my first post here, so i want to say Hello!
This san diego is: KAB1E 0619XPMW.
Just tested on our new mobo - Dfi lp ultra-d (till now we made it on Dfi rdx200), but now we have some problems with cooling...
Our singlestage is weak, and doing idle -22*
tell me, is it normal, ihs are straight:
- venice 3000+ e3 @ 3303mhz @ 1.82V we have in core temp ~ -7*
- san diego 3700+ e4 @ 36XXmhz @ 1.62V and +14*
?
btw: best spi 1M (on RDX): 24.578s @ 3365mhz, mem @ 280mhz 2-2-2-5:
http://images34.fotosik.pl/255/a55c194ec65b9d8em.png
gianni-gt, stunned_guy and knopflerbruce respect for great job which You are doing with s939 :)
regards Martin & Lukas
Thanks guy ;)
I also have one single stage that will do some times -43c...normal is still -36c.
Core temp shows some mess sometimes so dont doubt it all time.
KAB1E is very cold bugged but this 0619 XPMW is very good. I heard about that stepping that will does 3 ghz prime @ default vcore and higher with air cooling. So its no problem to go higher than 3.5 ghz.
It seems that your problem is the heat spreader ---> it has not 100 percent contact to the DIE. Help: you could remove the hs for better temps. Or you could take dry ice for cooling.
Than you can take more votage and 3.7 ghz and higher should no problem ;)
For superpi take 2x512 mb and your time will be better.
But nice for your first try.
another good xpmw... in the past we've had the chance to buy one of these, but it was too expensive. but i think my wr is safe.
We never bought such 0619 XPMW why ever ? I should search after that :D
3800+ Newcastle on way to me.
opty 146 in the house!!:D
CABYE 0540FPAW, #093:)
Hope it's a good one. 3g @ 1.4v would be nice, as breaking 3.8 with 10x-multi is pretty hard...:rolleyes:
hey knut, that opty 242 AAADB was my :(
and this was on ebay, but i have not money for that
http://i8.ebayimg.com/03/i/000/f5/12/26a1_1.JPG
made 02 23 (july 2002 approximately) :clap:
wow. those are with ceramic package :eek:
I have here 1 Opteron 252 with ceramique package....and without hs. I will sell this cpu now.
i think that was the cpu from the system that never runned :D
btw:
http://www.ripping.org/img/cpuz_warning.jpg
How's this Opty 170?
CCBWE 0551UPMW
@ Turrican
Yes its socket 940...but i will sell this system that never runned :(
@ gianni
I saw that warning in our database :D
Sorry knut but my 3000+ venice is still in my hands :D I will never sell this chip.
this cpu isnt good under single stage and prime...only 2.6 ghz with default vcore and prime....
The step of that opteron 242 with ceramique package is CAATC 0420MPMW, its an OSA242CEP5AL....its not an 252 :D My fault.
I also have one OSA252FAA5BL CABHE 0448MPMW here. this is an 252 with green package.
Hehe, I didn't want to buy it (well, I want that too, but it's not the reason I asked:D) - I just wanted to compare it to my 3200+ 0602APAW. I think your chip did 2.8 @ 1.41v (at least a 3000+ 0602APAW at forumdeluxx did that). My chip did about the same. Vcore went from 1.39-1.41v all the time, temps slightly below 40c. Didn't test it for ages, was just a quick air run to see if it worked or not:) But if it's got the same coldbug as your 3000+ it MAY be a gianni-killer:D
nice man, oldschool overclocking at its best
hmm...i do have an a8n-sli deluxe and a few 939s laying around :)
I tested my 3000+ yesterday again but the cpu is dead :( I wonder how this was done ? So no one will break that record I think.
Sorry to hear that. I also had a 3000+ 0602APAW, but it was DOA:mad: So no idea how great it was.
But yeah, that 3000+ Venice WR is perhaps the hardest one to break. Except for the 4000+ SD record, maybe.
I have an SLI-DR on the way, btw. And a chiller:) Maybe I can do some semi-serious benching in a couple of weeks time... Depends on alot of stuff, though, but I'm tired of not being able to max out all the CPUs I keep getting. Must have like 30 CPUs that wait for dryice. With the chips that weren't maxed out under dice, it'll be close to 40 on Ln2:D Not counting the 15 CPUs I expect to get from the US pretty soon:p:
oh stuned that's bad... but that's happend to a lot of mine cpus also... used under subzero, doesn't use it a couple of weeks, remounted => dead... i think the cpus was boored to death :D
I think so also ;)
Knut, you are crazy !!!! :D
My time on socket 939 is also coming to an end....i sell almost all of my hardware to do some new things. but one setup with single stage and sli-d i will not sell to do here and here some good socket 939 action.
how do you go about breaking the 3000+ A64 record??? due to its 9x multi surely the htt needed would be way too much for most mobo's to handle, I know my 939 dfi lp ultra d wouldnt do more than 360.
i have a 3000+ venice that is E6 and has seen 2700 ish MHz on 1.45v but it runs at stock and has done for 3 years now! sometime I may try it on a mates board and see what it can do, on air though.
Could've been worse, I think you've got a decent chip. Not WR material, but still.;) My DFi Expert's and Venuses have all done way above 400MHz HTT even at stock chipset voltage. No idea how high they can go, always limited by the "coldbug" of the cpus on air...
Hehe, no more 939 familty, then:p:
Btw, just won an NBBWE 0601DPMW 3400+;) Gianni-killer, perhaps? At least it's not worse than the E3 I tried a little while ago:D *praying for a dice-friendly coldbug*
Some new stuff arrived:) 3500+ Clawhammer CAA2C 0506GPMW, 3000+ Sempron NBBWE 0613DPAW and a CAB1E 0622EPMW 3800+. The last one is pretty decent, does 3130 stable at 1.34v and mid 40's load temp.:)
The 3500+ clawhammer will sucks...i had one with 0507bpmw...you need LN2 for high clocks but this one has a terrible cold bug for an clawhammer around 260 mhz htt.
The 3800+ is very good under air and water but will sucks under cold because also of an very low cold bug....this should be 230 mhz or lower.
I'll see what both those chips can do on air as soon as my SLI-DR arrives:) But I don't feel like testing more chips on this crappy A8N-SLI Deluxe.
I don't think the 3800+ has a coldbug that was that bad. I think 320mhz @ ~ 0 - +5C. That's what my other chip did, at least. I had 2 CAB1E 0639APAW's here, and a CAb1E 0620VPMW as well, at least one of those were quite good, too. Hopefully the 0620, as I remember your 0639BPMW had a REALLY nasty bug!:p:
Crazy CB on that Claw, reminds me of my FX53 CAAEC:p: It wasn't as bad as your chip, but awful compared to most others I've seen. Glad I just sold it, hopefulyl some CAAXC magic can outperform it (and your WR chip as well):D
That 0639 BPMW was very bad with cold bug....but under air the hell, you´re right.
Do you like to buy some chips from me ? :D
http://i3.ebayimg.com/01/i/000/f8/be/d8cd_1.JPG
my new AMD sample only for $33 :p:
I already have one of those, that's why I didn't go higher:p:
I got a nice package from the USA today, btw. 2 quadcore ES's. Opterons. Socket F:D FAAHB GA 0720SPBW.
I also got an opty 160 with a core voltage of 1,15v, socket 940 btw:)
i searching barcelona samples - 4AAAB ;) and i canīt find it..
holy cow, this thread is almost as old as my account =O
Ah, old thread, but incredible clocks. And best of all, the thread starter was using desktop wallpaper made by myself. :) That's why he achieved such great clocks.
But this thread is on the way down...most of us are running other systems...
I have only 5 FX processors left, all other cpus are sold. Is anybody here to will have an opteron 154 ?? :D
By the way: ripping.org is down :(
There was one great 3000+ venice postet with more than 3.6 ghz...is there anybody who will kill me ? :D
http://cgi.ebay.com/Lot-Of-25-SKT-75...QQcmdZViewItem
only for knut :D
and younger than my D235 :p
http://cgi.ebay.com/LOT-of-4-AMD-OPT...QQcmdZViewItem
@ Knut
My sticks could do 275 mhz only :( But i also have one pair of geil one tccd that could do 285 mhz @ 2-3-3-5 :D that is great for tccd.
The 3000+ venice was an E6 and postet 3 days ago on ripping.org but now the side is down, gianni ia searching a new server. i dont know the stepping but its good enough to go on 3. place with single stage.
yes ripping.org crashed the servers of the webhoster... the complete hoster was down. :D that makes me proud. :) i think that problem had youtube also at the beginning. :D
Fx55 in my sig, 1.82v, ln2 cooled to about -40c, 3916mhz
http://valid.x86-secret.com/show_oc.php?id=384713
Pretty good? I'm third on hwbot :D
Between -40c and -90c I struggled to hit 3.6GHz, at -100c and 2v it crashed saving validation at 3.94GHz and wouldn't POST again. I had insane condensation, so I think it might be a dead board due to that. I'll try it at stock in another board and let you know :)
Oh yes, I received your PM, I'll reply later today as I'm currently browsing from my phone.
w00t, keep puhsing it bro :toast:
I still build (high-end once) A64/Opty systems for some people, they have stocks of power for most games and with the prices of 2nd had good clockers and DFI boards RAm etc its a good value for money platform in my eyes!
at 2v, it MAY just be your CPU that died;) I usually don't go over 1.8v unless I REALLY need it (or the chip is worthless crap that deserves to die). At some point above 1.8v, the risk of killing the CPU is quite high. Killed 3 or 4 at 1.9, but at 1.8 - NONE (and I bet about 100 CPUs have been tested at that voltage).
@ phillip - I just bought your FX53:D
Congratulation, but could you imagine that i wont sell the cpu to you ??? :lol: :D ;)
@ OJ
Very nice result with that temps...
nice. how warm was it on that run (the weather)?
btw: ripping.org is back ;)
13'C and raining :up:
Why do so many people use Ultra-D for extreme benches? Can't imagine it can be more stable at extreme voltages/temps/frequencies than an Expert/Venus (or perhaps even SLI-DR?)
I don't get the temp issues (-100C - -40C), perhaps it was the condensation. May be an FX55 SD WR killer if you can get it to run nicely at -100c. ;) If it's not dead, of course...
Thanks for the pic :up:
I have an Expert but it keeps killing random hardware. I can't find a Venus :(
The temperature issue doesn't make sense to anyone, it's like a black hole :?:
I'm going to try the FX at stock again tomorrow.
I believe most prefered the Ultra-D due to its board layout and the PFC's seem to be slightly better.
SLI-D and SLI-DR are better in my eyes. The venus is too expensive for hardcore overclocking.
@ gianni
Knut bought my FX-53 on ebay --> but he will need LN2 for the record so we can hope to hold the record in our hands.
it wasn't only ln2 ;) it was the power of two sandiego brothers :D
So I tried the FX and I got 3 diagnostic LEDs, meaning the CPU checks out but the RAM has failed. Could it be the IMC on the CPU has bailed on me? I'm trying an infamous 24 hour BIOS clear at the moment.
Hey guys!
I'm willing to give these babes away, i don't get money from them and theyre quite unique... best 1m bench were 292 mhz @ 2-2-2-x @ 3.9 with fluctuacing voltage... and ambient were 24 c '
double sided dw 45 & chips on them maded year 02 week 40-50 dunno exactly.
havent benched those mems for over year now, first ill test if they work right and im ready to ship them for free to any of you especially to stunned or knoble
btw those sticks wont go over 240 with some of the cores, dunno why.
http://www.xtremesystems.org/forums/...d.php?t=139584
i really hope someone takes them to good home and tryes to brake last viking record ^^
Will you ship internationally?
Oh YES! My FX-55 WORKS! I did a long BIOS clear and it POSTED and is now happily running memtest! :banana:
I'm not going to back out but I'd say you need them more than me, so if you back out I'll take them :)
I'd love them, but I've got 4 sets of BH-5 and Knut has none so give him first option and if he says no I'll take them :up:
By the way, what would you people offer for my 3.916GHz FX-55?? :D
This are 2x256 mb sticks only so you cant get any good time in super pi. I had also 2x256 mb twinmos that did 292 mhz superpi and 286 mhz prime blend :D
I want some of that:up:
I sold my pair 3 month ago...
If I may ask, who bought them and for how much?
Someone from the hardwareluxx forum bought them at ebay. should be around 20 euro....i dont remember.
damn i used to have a golden 3700 too sold it tho :(
but now i have teh golden 3800 x2 ! non-manchester core LCBIE 0618EPMW (i believe)
had it at 2.8 easy on air, going to see what it can do soon with some w/c+tdx enchancements.
Absolutely ;) Next time with a higher multi :up: