Every chip is different, 5/1 might not be too weakQuote:
Originally Posted by ben805
7/3 and 7/2 are much less stable than 5/1 (just happens to be the best one for me too) on my chip
Printable View
Every chip is different, 5/1 might not be too weakQuote:
Originally Posted by ben805
7/3 and 7/2 are much less stable than 5/1 (just happens to be the best one for me too) on my chip
My Lvl II's run best on 5/1 or 3/1 on the Expert (with very tight timings). Now I'm wondering if I'm doing something wrong...
Hey LuckMan,Quote:
Originally Posted by LuckMan212
I just purchased a 170 CCBWE 0550UPMW from Dave at Flickerdown Data Systems. You can PM him through this forum at dave_graham
Also his prices are the best I've seen - I got mine for $382 + shipping. I was going to buy a 0546XPMW through eBay for $450 but when I found out I could get an 0550UPMW from him I jumped all over it.
I'm getting myself 5× 170 ccbwe 0550 upmw..i hopping that one of them will do 3ghz on water (with ihs on ofcourse) and keeping the winer..3 already sold, one will go ebay..can't wait tho
where the hell did you order 5 from...
some german online trailer
You simply have too much money...Quote:
Originally Posted by no_n|ck
Yeah I bought him a 165 0551 UPMW that does 2.8 with Big Typhoon (still tweaking). Great guy to deal that dave :) Fast and cheap!Quote:
Originally Posted by SkewdLogix
Testing an 170 CCBWE 0550UPMW now..
Can't get it past 2850 with less than 1.550 vcore on water...seems my chip is not as good as Magic's and Pet@rs .... :( it's a "434" (core0 is the one failing on me...core1 runs fine - maybe I should try to burnin core0?)
Or maybe I am doing something wrong.....
RAM is at 100Mhz to rule that out, HTT is at 3
My core0 is also much weaker, how do I burn it without removing the IHS? Thanks :)Quote:
Originally Posted by -.-PhanTom-.-
just ordered 170 0550 UPMW today :banana:
also ordered the 2048 MB Team Xtreem DDR500 UCCC kit
maybe they are here tomorrow:toast:
lets see what they are able to do on my MSI h3h3
greetz
heh, not at all..i will make a good profit here + mine 170 opty will be for free..Quote:
Originally Posted by -.-PhanTom-.-
guys, the best part is that all five of them are 1387434L50193 :banana: :banana:
they should be here in friday, will report how they will be doing..
Got my 0550 opteron 170 finnaly ;)
Did around 2700mhz on stockvolt.
Prime failed at 2800mhz at stockvolt within seconds.
running at 1.440vcore atm 2800mhz did 47min and then I stopped prime.
Edit: Now trying 2910mhz and had to step up vcore to 1.552 and priming.
I am using single phase and low settings atm just -20.
Will try to go lower but not sure if there is a coldbug yet with this one.
didnt boot up at the settings I had with the x2 so I will try to go slow down in temp.
What are common vcore to get to 3ghz btw?
using single phase?
3160 for dual 32M 1.5x1.04 in BIOS.......Quote:
Originally Posted by flopper
http://img71.imageshack.us/img71/856...31604rk.th.jpg
got my 170 0550 in today, will be installing after class (dam you studio!)
doesnt ben have a batch 408? is it the general consensus that batch 408 is better than 434? or what? just luck?
just got my 170 0550 batch 408 and i am priming at 2700 @ 1.37v
so far so good
Well, it isnt bad and it isnt good.
3ghz fail prime95 at core 0 almost directly.
2960mhz and priming at 1.6v
Just curious, does the core do better if you lower the temp a bit?
Since I am somewhat a noob on what voltage and stability and temp and prime95 does when it is regarded stable.
Does temp affect more or is it voltage that makes it stable?
It just would be so hugh silly to not get 30 or 50mhz more stable and to have a 3ghz 24/7 running cpu.
I use phasechange single and can get a tad more out of it, dunno if the cpu has a cold bug tho ;)
sweet, what production number did you get? you only feed 1.37? hehe....stop messing around, give it more juice and shoot for 3.0Ghz!! :D :D :toast:Quote:
Originally Posted by cronic
Been thinking the same...it seems that after a certain point, it gets really hard finding the optimal balance between vcore and max Mhz.... increasing vcore might give the CPU more stability for doing calculations, but also give an increase in heat output....how to find out when increasing vcore simple isn't worth it anymore?Quote:
Originally Posted by flopper
One thing I don't understand is how people on phase can get higher overclock on lower voltages, compared to people on water?
I mean, lower voltage=less heat, right? ... so why do people who are not on phase have to increase voltage to increase overclock (and heat will rise as well)
My critical point is around 2850 @ 1.550 vcore...after that, it is really a struggle...but core0 is the sinner every time..maybe bad IHS fit?
Maybe once AC5 is settled in, my CPU will be able to clock at lower voltage....
All this tuff about CPU burnin, isn't it merely a question of allowing the thermal compound to settle in (or burnin)?
i am typing this at 2925 @ 1.63v. runs a little warm idle at 42 c with room temp 21.5c or 70.7f. this is with a sorry as+ 133 divider as 166 and 183 don't like to boot to windows.
Well, it seems as a first attempt, it does around 2960mhz at 1,6volt.
Depends on how far prime95 will go, got to test it out for a longer period but it is anying not to get 3ghz stable when your close.
Not sure of temp tho, it gets at 27c doing prime even at phase so heat is a big thing here.
Might try for a 2900mhz at 1.55 instead for a 24/7 running.
I dont want to pop the IHS since that is to much of an hassle.
i have no idea how much vcore i can give my 170
if it arrives tomorrow.
MSI in sig + 170 cooled by Scythe Ninja and a 50cmf fan.
give me an advice pls:rolleyes:
i suggest 1,4-1,45?
thx:toast:
I'm running at 2800 (dual primed for 13 and a half hrs) with 1.36v on DFI Expert and Big Typhoon cooler with a PanaFlo M1A fan.....went up to 2850 stable with 1.425v, not tried for more yet !!
Waiting to watercool this Opty....they are VERY fussy with temps though!!
what batch #?Quote:
Originally Posted by bito
Batch 408...almost the same as yours in fact :)
nice :D
EDIT: what are your air temps at 2.85?
well it appears that anything over 279 with 200,183, or 166 divider does not work at any cpu multiplier or hypertransport setting. has anyone else run into this with the a8n32? i am on the 1009 bios and i am going to flash back to the 703 now and see if that makes a difference. it will boot to windows at 2790 @ 1.4v
Received mine over the weekend. Straight in 2.8ghz (see screenshot)
http://i5.photobucket.com/albums/y15.../32mPI2800.jpg
PWMIC temp never exceeded 45 Degrees throughout. But, if I'm priming @ 2.9ghz and take the voltage to 1.5v + PWMIC temps hit 74 by loop 15 !!! Something to do with the ampage going through the board me thinks. Damn shame really, even with a 80mm fan cooling the MOSFET, temps climb too high for my likings.
I also applied applied the WIN XP hotfix for dual core processors. Machine boots up and responds a bit quicker. Also applied a throttle registry tweak and agin, this seems to have helped alot too. Happy with 2.8ghz at 1.425v at the moment. I'm using Arctic Cooler 64
I alos have an DFI Expert coming next week, so hope to take the Opty further. Now its come out that these and the FX60 share the same core, I am a happy man.... :)
could someone email me the stupid superpi v1.4 zip file? the mirror site appears to be down at radeon.
cdelong2@rochester.rr.com
CYMQuote:
Originally Posted by cdelong
i didn't get anything man?
Quote:
Originally Posted by cdelong
ok..I tried again
LMK
Neal
I received my Zippy 700W on Monday to replace the PCP&C 510 SLI, the Zippy kicks butt and I love it!! Not sure if this has anything to do with it but I can now overclock with lower vcore. :slobber: The Gskill 2GB HZ also arrived today, haven't play around with it much but so far it's running 266 3D stable and rock solid, no longer getting stuck at stupid 100(1/2) divider, 133 also work now!! :D :D
3.2Ghz @1.53v
http://www.xtremesystems.org/forums/...chmentid=42992
dual SPI32M
http://www.xtremesystems.org/forums/...chmentid=42993
Sandra
http://www.xtremesystems.org/forums/...chmentid=42994
PCmark04
http://www.xtremesystems.org/forums/...chmentid=42995
3Dmark06
http://www.xtremesystems.org/forums/...chmentid=42996
It has everything to do with it.....but what's odd is that your psu is showing fluctuations on the rails. Mine barely budged at all, even on load - MAYBE .01-.02 at absolute max. Yours is showing different voltages during idle.....:confused:Quote:
Originally Posted by ben805
I'd also noticed my 7800GT with stock cooler now can overclock higher as well, with the 510SLI it maxed out 485/1180 but with the Zippy it can now do 497/1200 artifact-free!!Quote:
Originally Posted by s7e9h3n
hehe....at first I too was a bit concern about the fluctuation so I use my handy DMM to measured all rails, they're rock solid :D during prime they don't move at all but occationally shift either up or down 0.01v that's about it, DFI sensors have a mind of its own, same thing happen with the PCP&C 510 SLI as well ;) I have loads of crap (including two integrated amplifiers, my file server, laptop, AND the phase change) connected to the APC smartUPS 1400, during dual prime the UPS load was at 76~78% so I believe this Zippy still have plenty of headroom left :D :D
Check this out: I'm priming at 3.25Ghz @1.55v now, Smartguardian and MBM gives me different voltage readings, all 12v, 3.3v, and 5v are off!! my DMM shows 12.08v and didn't move one bit when I took this screenshot, so I think the MBM is more accurate than the Smartguardian :D
http://www.xtremesystems.org/forums/...chmentid=42997
Took this picture when the pc was idle, different 12v reading on smartguardian again :D :D
http://www.xtremesystems.org/forums/...chmentid=42998
I have one on the way now as well! I so can't wait to clock it!:banana:
Oh boy.. im feeling the need to get one of these.. !
Zippy Powa :woot:Quote:
Originally Posted by ben805
http://img254.imageshack.us/img254/9...pysmall7ks.jpg
ok so i just plugged in the 0550, try it at 2.6@1.4v and it fails within a minute
do i need to reinstall anything? reflash my bios? or can i go from one processor to another of the same type?
more voltage!!
its running at 2.7@1.5v now
it fails prime within seconds, but i can run superpi and i can bench it (3dmark01, 3dmark03, aquamark 3
any clue?
It takes time. Thermal paste needs to set in, and you may nee to tweak your memory settings. these Opty are memory sensitive.
On a good note.... I now have AS Ceramique sitting on my 170 0550 UPMW and it's now priming @ 2.8Ghz w/ 1.44v!! I was usng AS5 before. I seem to get my best OC's using Ceramique.
core 1 fails within a minute running at 2.7@1.55.. ..
could it be possible that i'm putting too many volts in it? people say it likes low volts, but would it be possible that it is more stable @ under 1.5v than at over 1.5v?
Quote:
Originally Posted by terrordemon
terror i have been playing with mine all afternoon. i have booted into windows at 2924 but it took 1.63v to do it. mmine seems to reaslly like the volts above 2790
no but like mine isnt even stable with 1.55v in bios @ 2700
what do you suggest?
Zippy is a beast! Where'd you get yours ben?:stick:
There is lots of possibilities, could be memory setting, could be bad heatsink mount or too much thermal compound (how hot is it running?), could be too much vcore? mine run best with just enough to get through.Quote:
Originally Posted by terrordemon
I got it from servercase.com, look for PSL-6701P and I did a write up on this beast at PSU section last night :DQuote:
Originally Posted by mitftw
ben805... you ever try that 170 on air?? I'm happy with mine on air, but just wonder what phase would do for it!!
Nice results Ben! Damn, you make me want a zippy.
nope, but I might someday, will have to yank the mobo out and undo the seal strings on both side and everything, don't feel like messing with it yet :lol:Quote:
Originally Posted by cdelong
Quote:
Originally Posted by Absolute_0
thanks man, I'm glad I went with this instead of the PCP&C 850, much prefer one beefy rail than multiple splits, this server grade psu doesn't cost an arm and leg either :D
:fact:
http://www.xtremesystems.org/forums/...ewpost&t=88153
temps arent the problem, load temps (for the short time i can get it to that) is around 40ish
EDIT: if not using 704-2bta, what bios should i use? specs in sig
for HZ, maybe 623-3?Quote:
Originally Posted by terrordemon
Hey ben I noticed your HZ timings @ 266 are 3-4-4-7
You 'maybe' should be able to get 3-3-3-7 @ 266MHz and at worst 3-4-3-7 as thats what my HZ pair (2x1GB) do with 2.68Vdimm set in bios :)
Temp running 2 x Prime are around 38-40c on DFI Expert (all of these read low)Quote:
Originally Posted by terrordemon
Try running yours at lower voltages.....say 1.375-1.4v. Mine seems to run with great stability at these volts.....if i up to 1.55v then prime will error straight away !!
Give it a go and let us know how you get on....:)
i'll try it with lower voltages... i'll be really pissed if this chip is even worse than my dud 0530
why is it that when i actually start overclocking all i get is duds.... :(
EDIT; tried 2.6 @ 1.4 and 1.375, 1.4 almost instantly fails in prime, 1.375 instantly freezes comp when i start priming
just arrived ;)
now i'm waiting for my ram and my cooler...maybe tomorrow.
xxxxxxxxxxxxx
CCBW 0550UPMW
1387434L50398
Attachment 43017
hope it will do fine :toast:
greetz
you can edit your sig since you have it now;)
not stable at 2650 @ 1.5...
would a clean install of windows do anything?
i left the dividers alone and went back to 1:1 for some testing. my core one failed prime after 1 hour and seven minutes at 1.45v. i then tried 1.425v and it was running for 2 hours when i left this morning. i will report back. i think i am nearing the max of my memory at 277-279. i will know more this afternoon
well it faild after 5 hours at 1.425v
so i just made it a cabriolet. popping the lid cut 5 degrees off load and idle.
i am dual priming now toppless and with ambient at 20.9c my load temp is 35 c at 2770@1.41v with all fans at there lowest settings.
hmmm....
can boot @ 2700 1,4V but prime fails...after 2 min. lol....
is it the heat or is it my motherboard?
btw...
MSI K8n neo2 + UCCC ram = crap?
greetz
I've been lucky enough to get my hands on one of these great 170-0550-UPMW Opterons as well.
Here's screen grab from a few days ago...
2.8Ghz @ 1.4v
*** Click to enlarge! ***
http://www.coldfudge.com/members/tim...o_0550_28g.jpg
*** Click to enlarge! ***
Since then I've put it through daily use including gaming, xvid encoding, photoshopping amongst other things and everything has been rock solid.
I'll try and go for 2.9Ghz and above later this week :)
@Timko
Batch?
434/408? ???
just got mine in aswell, priming now at 2750@1.45v but thats just what i put it at for initial testing. first time i booted i forgot to reset my old settings and it booted up fine at 2840 1.52v, just got a little hot (40c idle).
ill report back later with more :)
/edit
2850 at 1.45v crapped out pretty much instantly, so at 2800 now.. stable for 45mins so far.
mine's from 385 batch btw..
It's a 434 :)Quote:
Originally Posted by tHe_jOkeR
and the batch?
are u using wc? or air?
WC in the sig. I'd guess at that voltage and clock speed, it's on water.Quote:
Originally Posted by tHe_jOkeR
My TDX is just begging for one of these 0550's.. hehe..
Hopefully in the next week I'll have one !!
ok mine is a 170 o550 batch 408 *naked*
2880 @ 1.57v ddr522
prime stable for over 17 hours
http://i1.tinypic.com/nbx741.jpg
priming at 2898 at the moment and will report back
Just getting started with AMD, i'll be needing advice soon :) On water. loose ram timings.
Prime 10x275
[IMG]http://img404.imageshack.us/img404/3...dmedium.th.jpg[/IMG]
32 mil. 10x280
[IMG]http://img324.imageshack.us/img324/4...dium1gp.th.jpg[/IMG]
Prime temps running 10x275
[IMG]http://img402.imageshack.us/img402/4...dium3ls.th.jpg[/IMG]
I am just wondering about my opteron 170 0550UPMW Oc. Its the 434 variant and the batch is 667. Alright, how many of you would say that an enermax noisetaker 600w is too less of a powersupply for a insanely clocked 7800gtx and also a high clocked dual-core? I am saying this with good reason. The below are my experiences and all at stock volts.
Initial: the cpu would fail 2.6 instantly
After bios flash to fx60test2: the cpu would last just under 5 mins before erroring
After IHS removal: the cpu would error after 9 mins and load had dropped from 38C to 28C, idle temps havent changed.
Now this is the fun part,
After putting my gtx back to stock and removing two system fans, the cpu would remain error free for just over 15mins!
Is this a sign the powersupply is holding me back?
Also, after two 3 hour burn-in periods the chip is prime stable for just over 31 minutes. What I am doing this week is priming it everyday until friday night after I come back from school.
I am looking into getting an OCZ 520 powerstream to run the motherboard + cpu and using the enermax to run the w/c loop, fans, hds, cd-roms, and the gtx.
Advice?
I'm not sure if it's the PSU causing you problems but can you not borrow a different PSU from a friend to test that theory out before you spend some cash?Quote:
Originally Posted by madgravity34
I wish I could, I live in oklahoma and there arent a lot of computer nerds....lol.
I was thinking of getting a modstream 520 from a local compUSA and using it to test if its the powersupply or not. The main thing I need is more amps than the 18 the enermax puts out. Do you think the modstream is good enough to at least be good for testing purposes?
18 amps is pretty weak, One sign of a over worked psu is random reboots. Fans dont draw alot of juice, but im sure that gtx does.
a 520 powerstream should be just fine, mines powering all the stuff in my sig.
In DFI Street the PowerStream 520W is the noted as the best with 1xGPU. I had (it died) a 420w and no problems with a venice and a 7800gt (488/1241) so 520w should do the work for you.
Got mine today...trying 3k. Looks like maybe with the IHS on it's gonna take some decent voltage (1.55v-1.59v) to get it stable...
I dumped my 0550UPMW after i tested my new 170 0550VPMW ->
My 0550VPMW is blowing smoke so far ... latest is :woot: 10x315@1,504 under Water (IHS is still untouched !) :woot:
Screeny from last night:
10x315@1,504V:
http://img85.imageshack.us/img85/438...02236tc.th.jpg
Other Results on h2o:
10x310@1,456V
http://img303.imageshack.us/img303/6...02230rw.th.jpg
10x300@1,376V
http://img428.imageshack.us/img428/9...602231j.th.jpg
Guess it`s time for a chiller :toast:
Pet@x those are awesome results
Thanks - Yes, they are .... prolly the best 170 on real h2o these days. 10x330 is also bootable :woot: ok, SPi 1m Ok ... for Prime the temps get ugly ... so a chiller will help there .... as i want to keep the IHS as long as possible :nono:Quote:
Originally Posted by cronic
Really looks like these 0550VPMWs blow smoke :woot:
Well not all of them, looks like you just got an amazing chip. According to your sig the other one you had neede more volts to hit 3ghz than your UPMW did. Anyways, I love the clocks on the good VPMW you have. What powersupply are you using?
PS. How much of an oc improvement did your expert give you with these DC Opterons than your regular nf4's?
Well ... the 2nd VPMW is also on the way down on VCore ... right now 1,52 for 10x305 ... but i 1st was stuck on the other one as it performed that well ;) . Expert ... well ... let´s put it that way ... it handles better the power needs from the duals ... on the "older" nf4´s (SLI-DR etc) the PWM Temp was very high compared to the kinda "low" temps on the 4Phase Expert ... that´s basically why i`m stuck with the Expert ...Quote:
Originally Posted by madgravity34
If you are lucky you`ll find VPMWs that can do 10x300 ~ 1,45 and much lower ... :woot:
PSU is an OCZ 600 ...
cya
Just got my CCBWE 0550 UPMW and it's not lookign like a stellar chip. Trying to get stable @ 2900 now but it failed 2950 & 3000 even up to 1.6v:( It's week 0286. Anyone else have this week?
Great looking chip Pete. You guys in EU complain how we in the US get cheap hardware but you guys seem to get some really great stepping cpu's.Quote:
Originally Posted by Pete@X
What was the room temp on your ss?
i hate you PeteX :p: :eek: :toast: :toast: :toast:
A big part of the reason why the Expert has a much lower PWMIC temp than the older layout DFI boards is due to the location of the PWMIC probe. On the older boards I get anywhere from 22-32C PWMIC readings while one the Expert, I've seen temps as low as 9C:Quote:
Originally Posted by Pete@X
http://img95.imageshack.us/img95/48/pwmic3ex.jpg
Its cpu socket location also can help cool the memory.....when using really cold phase (same as RDX200).....
http://img239.imageshack.us/img239/2421/mvc196f7ug.jpg
.....Or kill your pwmic as it did on my Expert :p:Quote:
Originally Posted by Dumo
Yep, RDX200 mortality rate is still high:)Quote:
Originally Posted by s7e9h3n
Doh...not even 2880...this isn't looking good. My 0546xmpw is prolly going to be a better chip.Quote:
Originally Posted by FZ1
U sure it's not your modstream holding u back?Quote:
Originally Posted by FZ1
it's weird dudes .. i have 175 0550 VPMW but i couldn't even hit 2.8 becoz of da vcore .. @2.7Ghz i need 1.55v @2.75 i need 1.65v to stable .. 1.65v on air da cpu produces a lot of heat so i couldn't continue to push higher
Buying a 175 is a LOT of money, especially if you cant get it to run 100% stable at only 2.75GHz :slapass:Quote:
Originally Posted by ericware
scary fact :DQuote:
Originally Posted by Dumo
that's just crazy!Quote:
Originally Posted by Pete@X
I´d rather say insane for dual core.:) Btw if it´s not a secret then what´s the batch.;)
... insane, crazy ... oh well i heard so many words in the past days :toast: It`s a 505xx, but ... i have 2 others here (507xx, 502xx) - they do also 3Ghz, just with more VCore :woot:Quote:
Originally Posted by Evil
cya
Pete
Im having no luck with these 170 0550 UMPW's . I bought a 434L one from trend4pc after people where getting good results and mine wouldnt do more than 2.65Ghz. I then got a 408L from the same place as loads of people hitting 2.9+ and just 2.7Ghz for me.
I noticed both these chips had a low number from the batch, both were well under 501** . I notice Pete@X's chips seem to be much higher.
Maybe Im doing something wrong, but then again my 165 seems to do well 2.75 @ 1.5v.
Pretty sure. I checked the rails and they are very stable - well within spec and measured the vcore under load and stable as well. There's really no other way I could tell without sticking another PSU in there I guess. I haven't heard of anyone having issues with the modstreams.Quote:
Originally Posted by fareastgq
just a heads up...
i got: 596K, 888K, 786L, and 678L substrates this time around on box 170s.
dave