don't worry bout it crapping out at 2950 on water/air. Mine did same thing up to 1.55v. Loves DI and voltage though :)
Printable View
don't worry bout it crapping out at 2950 on water/air. Mine did same thing up to 1.55v. Loves DI and voltage though :)
1.72v BSOD'd big time. Loaded up 16m/16m and took a shower. Came back BSOD. I have a feeling it's the IHS. My 170 did the exact same thing until I took it off.
Trust me it's not AFI. Mine is just as bad. Put it under DI with IHS on and you'll see.
Alright :) Last time I used an IHS I froze half the board through and had to shut down to prevent condensation damages. Better insulate better.
Dielectic grease FTW. manta is bathed in it now.Quote:
Originally Posted by afireinside
Nah I don't want to make a mess. I'll probably just put more nail polish around the outer socket area and some stick foam on the back around back plate. My insulation sucks now.
http://afireinside.evilspork.net/AFI...n/IMG_0308.JPG
I've added another layer since.
Old insulation:
http://afireinside.evilspork.net/AFI/OPTERON/mpot1.JPG
http://afireinside.evilspork.net/AFI/OPTERON/mpot2.JPG
Sorry for the OT
would you believe I use no insulation on the cpu side of the board save the armacell booty on my tube :) and dielectic. I use a small armacell pad on backside also coated with dielectic.
http://img.photobucket.com/albums/v4...insulation.jpg
:D
Grease is so messy I don't want to use it. I'm fine no insulation on CPU but IHS freezes mobo pcb over.
wont be at home Tomorrow either i could test my 0601 TPMW..
but only with air:D
in 2 days we'll see :D
quick prime95 shot...food...bbl :D
http://simon.webideal.ca/Temp/3.0%20prime.JPG
found another chip in the dumpster. will also have 4 dfi ultra-d boards + 1 a8n32sli to test out tonight.
hopefully, i can do 2.9ghz+ with this chip. last chip i was able to do 3d at 2.95ghz.
1 hour dual SP2k4 on LDBFE 0607UPMW, "T", 3GHz phase cooled. vcore at 1.472v, load temp -8C. Shots tomorrow.
So i got an 0601TPMW "N" chip, seems to top our at 2.9 Ghz with 1.55v, watercooling 43c load. Any higher and it wants more volts, and then hits high 40s c and errors.
EDIT failed prime after 5 minutes :(
they run really hot
"N"s, or LDBHEs?
Absolute, Phase usually wakes up the beast in these cpu's I suggest you try it first before dumping it.
my 2 0601 lbdhe are "N" chip and they run really hot...except on my new Ultra-D AD0 (I think I had like 10 ultra-d lol)
39c loaded @ 1.5v...impossible
I'm getting 39c loaded with 1.42v on good watercooling.
45c loaded with 1.56v
49c loaded with 1.6v
I just had an Opteron 165... really miss that chip lol, loaded at 38c with 1.53v. Idle temp was 26c. Phenomenally cold CPU.
3800+ is doing 2.8 with 1.42 now, seems happy, definitely below average for all the 0601 chips i've seen though, as i've got the thing on good watercooling, there's not much more i can do for it. The phase unit requires a bit of work and isn't something i'll be getting around to doing for a few days at least.
2.8 GHz with 1.42v failed... at 38c load too. Damn core0 really blows.
my 0601SPMW sucks bad, won't continue testing it since it cannot pass 2.9ghz 8m spi at voltages 1.4-1.6v
back to rigorous testing with 0601TPMW...
modded my Big TT typhoon with a delta fan and shroud. I'll report back with more results later.
86f outside, room temp at 25c.
edit: finally getting up some pictures.
http://www.xtremesystems.org/forums/...1&d=1143162561
http://www.xtremesystems.org/forums/...1&d=1143162561
Will the 0601SPMW run 2800 on 1.4v ?Quote:
Originally Posted by Alex08
no idea, once it didnt run 2.9ghz, i just gave up on it.Quote:
Originally Posted by Blue078
at the moment i'm running spi2004 on the other chip so i'll have to wait to try out the other chip again.
ahhhh, both my processors got delayed in shipment from california. They shipped separately, and one will be here monday....the other tuesday. mother@%$#!!! :mad: :mad: :mad: :mad: :mad: :mad: \
;) oh well.
Got an LDBHE 0606 EPMW 3800+ X2 today, T in the corner. Dual Prime95'd for 5 hours at 2.6 and 1.3v, load got up to about 46-47C according to the A8N-SLI I have it in, cooled by an Arctic Cooling Freezer 64. Currently at 1.44v and 2.8ghz with Dual Prime95 tasks going again, its at 58C now so a step up in the cooling appears in order, perhaps something a little less silent. Going to leave this on overnight and see how it did in the morning.
It's really important with these to confirm that it is one particular core and not either core falling over. If it's either core or won't finish SuperPi, look to RAM tweaking. Drive strength is critical to stability, especially with N chips.
If you get 2.8 stable on stock air, 3.1 is in the bag on phase.
2.8 @ 1.44 fails, but 1.46v is looking good so far. Core0 appears to be the weaker of the 2 for me.
The new good steppings (LDBHE 060x), are they E6 or E4?
What should I try then? I've been running 3/3 like OPBs UTT/BH5 timings say.Quote:
Originally Posted by stone_cold_Jimi
Pretty sure their all E4Quote:
Originally Posted by ManagHead
Yep, they're E4s. ALL LDBHE are E4s IIRC.
burn in work :cool:
http://simon.webideal.ca/Temp/x2%203...95%20prime.JPG
^^ Nice frigging chip. Beats the crap out of mine, 1.45v and 2.8 Ghz fails repeatedly.
defintely nice chip...ur chip seems to be identical to mine 0601 LDBHE XPDW (not sure about the last 4 - have it written at home). hits 2.95 GHz relatively easily, but needs MUCH more voltage at 3.0GHz due to the heat rise...
Sorry, been out all day.Quote:
Originally Posted by afireinside
Well, my CDBHE 4600 found 3/1 or 3/auto best with my G.Skill HZ, whereas my 3800 BSODs on 3/anything but likes 7/2 or 7/3.... same sticks. Where that list of Oskar Wu's wacky DS numbering scheme? It's completely counterintuitive.
On a completely different issue, I like a good mount on the Vapo head. Too good. Look what I did to my N 0607. Someone check the pins for me, I can't bear to look.
Hi all, i have a noob question here.
Ive never overclocked anything before in my life and these new Manchesters seem like an ideal opportunity to try it out for the first time.
I would be happy with an overclock from 2.0 to 2.5, i understand from the thread creator that this was possible with the retail HS and fan, can anyone confirm that this is indeed possible?
Also could someone please recommend a motherboard for me which will help me achieve this overclock without hassles for a noob 1st timer like me?
Many thanks.
yes, KingRoLo it is very possible and almost definate that you wil achieve 2.5ghz with the stock heatsink and fan and a little reading in these forums. It will depend however on which motherboard you choose. Are you using pci-express or an agp graphics card.
Welcome to Xtreme Systems, anyone in here is very welcome to help!
Thanks for the welcome fullup3 and good to hear i can achieve a 2.5 with the Retial HS/Fan :)
Graphic card wise it will be PCI-E. With that in mind could you recommend a motherboard?
I was thinking about this one.
http://www2.abit.com.tw/page/uk/moth...939&DEFTITLE=Y
TBH the SLi doesnt interest me, i only want to know if this will be a good board for me to overclock the Manchesters without much hassle.
Any other mobo recommendations are appreciated :)
I, and alot of others around here, would recommend the dfi ultra-d. It is very good for overclocking, and offers more tweaking options than most boards. It is however, not the most friendly board for someone new to overclocking an a64. the asus boards are pretty easy to use, and overclock well also. I'm not too sure about the abit you mentioned....I have no experience with it, and the only abit I ever owned pretty much sucked!
i finally got around to using the dfi ultra-d today, it's a revision AD0. I'm not sure why, but I definitely like it over the asus a8n32sli even though the asus didn't give me any problems. Anyways today was very productive.
my chip is now 3.0ghz superpi 32m stable @ 1.55+104% with typical southern california weather. I haven't tested the max of this chip yet, but i'm definitely satisfied with it so far. :woot:
here's some pictures i took while running superpi32m. my new setup is found in my signature. my 2gigs of ram running @ 250 @ 2.5-3-2-5 @ ddr500.
http://www.xtremesystems.org/forums/...1&d=1143288120
http://www.xtremesystems.org/forums/...1&d=1143288120
http://www.xtremesystems.org/forums/...1&d=1143288120
some undervolting fun...
2320 MHz @ 1.09V
2465 MHz @ 1.15V
2610 MHz @ 1.20V
2755 MHz @ 1.26V
Nice one Alex08. Good to see your temps are still manageable with those volts on water.
I just grabbed a LDBFE 0607 XPAW with a low batch number from Tankguys. Hopefully it will be a "N" and then I'll have it made. It should be here mid next week, I'll keep you guys updated on how it does.
why you want a "T" chip ?Quote:
Originally Posted by EnJoY
"N" chip clock probably better with low volts
Um...yea, that's what I meant... :rolleyes:Quote:
Originally Posted by Supertim0r
air ?
phase soon ? :p:
Where do I find the "N"? Do I have to get the cpu naked? I will go choose a 3800 next monday and I am confused about that "N".
Anything else I should look for to get a good ocer?
It's down here - this is an N.
Honestly i have found no difference between N and T.
My current LDBHE 0601TPMW "N" loves volts... I need to increase voltage just to get it to run 2.7, and it takes 1.45v to run 2.8. I give it 1.6v to prime at 2.9. It's a major voltage whore; i've used a lot of CPUs and I know what it's like when they go far on low volts and then top out due to temp constraints. My 0515APAW and of course my 170 VPMW come to mind.
But this 3800+ "N" chip is definitely NOT a low voltage chip nor is it a great clocker.
Another "N" chip i have is an Opteron 148 and it likes volts just as much as anything elese.
i've tried many chips and the 3 "N" chips loved low vcore (high clock too)
i'm on the same boat with Absolute, my "N" chip scales with higher voltages.
1.5v for 2.9ghz
1.55v for 2.95
1.6v for 3.0ghz
i havent tried past this yet as i'm doubting my chip would do anybetter than 3ghz on air.
hello guy's :D
LDBFE 0607UPMW
Z894869B60598
Is it a good stepping ?
You tell us man. The 060x are all pretty good. I dont know how 0607 is versus 0601 though. There's only a handful of 0607 results around. Looking forward to having another one ;)
http://img.photobucket.com/albums/v4...N/IMG_0444.jpg
85% humidity this is going to suck but whatever
LMAO AFI. I'll be joining you tommorow. Hopefully i can crack 360 htt cold this time :) and take down the record if you don't break it now.
Maybe I should start SSing :D
Do it to it :)
f'n cold bug on a "BH" can my luck get any worse?
Yeah it's gonna bug of course, but where does it occur?
I can boot 3ghz-ish at 1.55v. 3.2 doesn't boot at 1.7v
load sp2004 and clockgen to 3.2 at 1.55 and it runs
any voltage up to 1.7 loaded in windows I can't get 3.25 even with 200mhz mem
Your kidding me.......You know what AFI i had to run a really weak drive on phase with mine.........Have you played with drive strengths while cold?Quote:
Originally Posted by afireinside
To much of a bump AFI try 1.6 and if its an N chip do very very small increments.....They hate large bumps in voltage.......
Playing with it now. What should I try? I was running 3/3 and tried 3/1 3/auto 7/3 and 7/2 like jimi suggested.
I run 0/0 in the ati board on phase where normally with my 2 gig set i run 6/6 I hit the wall at 330 with 6/6 dropped to 0/0 and it went to 350.
I'm on an expert 7/2 7/4 7/3 auto/auto 1/1 3/3 3/1 3/auto all did nothing.
:( I can't even begin to guess what the dfi expert eqivalent would be to this board if even weakest settings didn't work
3200 is completely stable and runs 8m with a normal time. My 3dmark scores are trash though. I'm about to sell CPU and expert :(
my 3dmark score was horrible too :(Quote:
Originally Posted by afireinside
3ghz 1.45v, ballistix @ DDR600 3-3-3-8 2.6v :stick: 7800GT 500/1200...really bad score (maybe it's the new 84.21 ?)
yeah 84.21s kinda suck. I kicked my ram speed up and everything look ok, waiting for it to end. 554/1340 on a 7800GT isn't supposed to score 9k when you've done 9.5k same gpu clocks and 2.85 cpu.
bummer that chip is bugged for you man. You'll find a workaround for it if there is one though.
Sorry to hear that AFI.
Windows starts crashing around 3.23 :( Kinda odd that my 170 crapped out at the same place. Though I could run half of 3dmark with that and it crashed at the exact clock my ram walls at. This I get a windows error for one instance of pi over 3.23ghz
Gosh this is really extreme...
Most of you are getting 3GHz out of the chips, and you're far from happy with them.
I wish i could get one x2 3800+ wich could do 3gigs.
Currently 3000+@ 2.8GHz and pretty satisfied, it doesn't reach 3Ghz, but i don't care got a pretty good overclock and performance here.
Be happy people, when i eventually buy one of those x2 i'll be jumping up and down if i get one that's 3GHz stable. No matter how many volts.
Mine currently is taking 1.65v (damn chip won't even do 2.9 with 1.8v), cause this way i can get from the normal 42ºC to 60 without crashing in the middle, getting ready for summer temps :)
Eversor thats on DI.........And you should expect alot from DI. I ran mine up to 3.5gig...........Which is why he's bummed out.
Man, my LDBHE 0601TPMW is HARD to figure out. This chip's behavior is completely sporadic. Sometimes it will prime for hours and fail as soon as i open a song with WMP. And then i can close WMP, reset SP2004, and it runs fine.
Sometimes it will prime for like 5 minutes, fail, and then i reset it and it can't go more than 30 seconds no matter what i do.
I used to need 1.58 for 2.9 Ghz and 1.60 for 2.95 Ghz, and the latter would fail.
Now i am doing 1.53 for 2.9 Ghz and 1.56 for 2.95 Ghz, but i can't get it to prime at 2.96 Ghz even with 1.6v.
So.. the paste is setting in or it's responding to burn in or something, because it's working @ lower voltage, however, i have a major wall past 2.95 Ghz and it will not prime with more volts.
i've actually got 3 LDBHE 0604 CPMWs sitting around here...anyone have any data on those?
dave
Nope Dave, i've googled it a few times before, can't find anything on 0604.
i guess i'll have to eat one outta stock and give it a test run.
lordy. need to dump them to get more 170s... :)
dave
if i were to buy a random new CPU ...X2 3800 or opteron 165? these X2 results are nice (but so are the opteron ones :rolleyes: )
this somewhat seems like what i encountered with the ultra-d. Once I tweaked the mem timings, I had no problem with 3.0ghz. However, before I tweaked the memory, I was hitting a wall at 2.95 and no matter what volts I used I couldn't hit 3.0ghz.Quote:
Originally Posted by Absolute_0
Dave,
I have an LDBHE 0604 in a build I am putting together for a friend right now. It is super stable at 2.6 GHz on 1.4v in a DFI NF4 Ultra-SLI with a Zalman CNPS 7000B HSF on it. Runs 35-37 degrees with both cores under full load.
I like this CPU I would be interested in one of yours for myself.
rockin'...just the data i needed to know.
bjonz, let me know by PM what you want to do.
dave
my memory controller rox with my new AD0
my ballistix passed 4h47mins of memtest @ DDR600 2.6v :eek:
the only thing bad...it's not windows memtest stable :(
any suggestions to get my memory stable ? already tried 2.7v and 1.47v cpu (instead of 1.45v)
Raise MAL/RP and/or adjust drive stengths.
Well I just started playing with my new ldbhe 0601 xpmw and it is not the greatest so far. Prime failed after about 20 minutes at 2600 at 1.32. I bumped vcore to 1.36 and prime passed 45 minutes until I stopped it. Now priming at 2650 1.39v. Will report in a few.
it should be a nice clocker on phase
I just got today a LDBHE 0606EPMW I bought here in Portugal. I hope I can reach 3ghz with my TT.
I can't test right now cause I also sold my graphic card... I hope I can get a 7900GT tomorrow to replace my old.
I will keep you informed
Ok, so Last night I did some testing on the LDBHE 0601 XPMW that I received and it took 1 voltage bump to be stable at 2600. Then 1.39 for 2650, and 1.44 for 2700. When I would try 2750 or 2800 with up to 1.5v ( the most I tried) either one core or the other would fail randomly..probably due to heat load showing above 42c on my ultra d. this chip was a "T" chip and would probably be suited to higher end water cooling or phase. Now I have to decide whether to ebay her, or find someone with phase that wants a good cold chip.
Now on to today's adventures....
I just unpacked my new LDBHE 0601 TPMW which is a "N" chip. I started with a quick 20 minute prime at stock at 2600, and all was good.
Then I did 10 minutes at 2700 at stock (1.32v) and impatiently stopped it to try more.
Now as I type this I am 1.32v @ 2800 dual prime!!! This baby is a keeper fo sho!
I'm about to have to go back to work for 2 hours, and I'll leave it running at this and report back in a few.
Quote:
Originally Posted by fullup3
Nothing like a good old "N" chip.:toast:
i have the same chip as fullup3 also with a "N"
but i cant figure out the temps on my Expert yet:D
Looks like the 3800's we've got right now are all "T" chips. I wonder if that actually makes a huge difference.
Someday, it sure would be interesting to test 100 chips or so, so that we could do a statistical analysis to see if stepping really matters, T/N, and if burning in actually helps.
Quote:
Originally Posted by TankGuys
http://img281.imageshack.us/img281/1901/can7ag.th.jpg
it sure made a difference in the ones I just got.
the "T" wanted alot of volts to scale, and this "N" is testing dual prime right now at 1.45v 2950mhz.
same...2950mhz 1.45v
No comment.
Don't get me wrong, I don't doubt what people are seeing. My point was that it would be very interesting to have *many* chips all benched by the same person, on the same equipment, under the same conditions.
So, all I meant was that it would be very interesting to be able to remove all those variables, so that we could get actual "scientific" data to prove (or disprove) our "anecdotal" findings.
I agree completely, TankGuys.
Well, the wife just got home and made me turn on the heat in the house, so looks like no more good pushing of the chip tonight. I'll just try burning in at stock volts and keep nudging fro 2800 as high as it'll go. at 2820 now @ 1.32v.
:eek:Quote:
Originally Posted by fullup3
i got the exact same stepping, with watercooling i need 1.42v or so to get 2.8 stable and 1.6v to get 2.95 stable. Nice chip man...
thanks, I'm excited about it. I am gonna try to prime all night at 2.8 stock volts, and let my thermal paste settle in, or whatever, and see if that helps.
Absolute, didn't yours improve with burnin?
Yeah, i used to need 1.45v for 2.8 Ghz, and 1.6v on 2.95 Ghz would fail in a few minutes. Now 2.8 needs 1.42 and 2.95 runs fo rhours with 1.6v. I used to need 1.58 for 2.9, now i can do 1.56, or 1.53 if the room is cold. It's kinda weird. I'll keep trying to burn in, but no matter what i do or the volts, i can't do 3 Ghz.
Hey guy's, Iam new to this forum, I wondering I just bought a X2 3800
stepping is a LDBHE 0606 wondering if anyone has hit 3.0GHz or close to it.
and one more thing which DFI mobo should i get the ULTRA-D or SLi
thanks
not sure about the 0606's....not too many around. You will be very lucky to get to 3.0ghz without phase, but that doesn't mean that you can't.
and the mobo depends on if you want to use 1 or 2 video cards. 1 = ultra-d
2= sli....also, the ultra-d can be easily modded to the sli.
good luck with your new toy!
Well, I've run a T and an N from the same week. They clock about the same, just have different ways of getting there. Very different thermally.Quote:
Originally Posted by TankGuys