• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

winchester overclock database

curious.. my clawhammer has always gotten 2310mhz out of 1.55v
never more, never less, since day 1

and my 2500+ barton.. always 2.3ghz st stock voltage. 1.8v could get it to 2.45ghz or so, but not with any reliability. burn-in just didn't work for me.
 
Like most things it doesn't scale very far anyway, otherwise I'd have infinately low voltage with infinately high speed. It's amusing sometimes, and if it didn't do you any good you probably just tweaked it pretty well already.
 
maybe these CBBID winchesters are different. Dunno, all I know is that mine liked voltage...but I don't feel real comfortable feeding it 1.65v to get 2.6Ghz from it when I plan on keeping it for a while :p As it stands now, I can run stock speeds at ~1.25v after all this burn-in stuff.
 
there's no such thing as burn in. Total myth. The same bunk is repeated endlessly until it is accepted as fact, kinda like politics.
 
*shrug*

My Winnie 3000+ was crashing pretty fast at 2.5GHz unless it had 2.65V. I left it loaded overnight with that, now I have it up and stable at 1.55V (with a slightly higher clock). What's your explaination? If you can give me one that makes more rational sense, I'd love to adopt it.
 
Selecter said:
there's no such thing as burn in. Total myth. The same bunk is repeated endlessly until it is accepted as fact, kinda like politics.
Click to expand...

care to explain why my cpu can run with .2 less volts than it did before the burn-in then when no other settings were changed? Also care to explain that to the people over at xtremesystems that can clock their ram higher with less errors after burnin periods??
 
My results are still pending with this burn-in method. But it seems something else may be at work here, if people are getting results with the so called "burn-in".
 
Some of it is because the prolonged "burn" time probably helps set the thermal compound somewhat. But you have to do similar with rechargable batteries, so I don't see why it should be too suprising that you can "train" voltage requirements a little.
 
how long does it take most thermal compounds to set? I use ceramique and it's been there for 4 weeks now...that should be long enough to set wouldn't you think?
 
AS5 says 200 days...I think it's closer to a month for ceramic based ones though. Has to go through actual thermal cycles though so for the first bit you're supposed to actually shut down your system at night...
 
200 hours ;)

nst6563 said:
care to explain why my cpu can run with .2 less volts than it did before the burn-in then when no other settings were changed? Also care to explain that to the people over at xtremesystems that can clock their ram higher with less errors after burnin periods??
Click to expand...
i think burn in really does work for ram.
but not for cpu's. .2v is a huge drop in voltage, which would imply that something else is going on.


also, this is getting waaayyyy off topic. winchester oc thread. if you guys wanna debate on burn-in, make a new thread for it. ;)
 
OT, I wanna know why Windows won't boot when Memtest says it's good...:p

Stupid Winnie memory controllers. Here Venice, good boy, c'mere...
 
i sometimes have that problem... well it happened with my k8n neo. now with the dfi, if it won't boot into windows, i'll get a detecting raid screen. at least i guess it saves me from data corruption :D
 
Ahhhhhhhhh i got this rare piece of crap @ MPM
CBBHD 0441 MPM

im gonna try to overclock tonight.. and post my result at the same time.. my friend got the same cpu too, i will also post his result

im afraid.. MPM sound like a crappy CPU.. anyway i will see tonight!

:D
 
Arobas said:
Ahhhhhhhhh i got this rare piece of crap @ MPM
CBBHD 0441 MPM

im gonna try to overclock tonight.. and post my result at the same time.. my friend got the same cpu too, i will also post his result

im afraid.. MPM sound like a crappy CPU.. anyway i will see tonight!

:D
Click to expand...

Bah, nothing wrong with MPM. Its a pretty common stepping, and CBBHD has been doing pretty good. Expect good things.
 
i FINALLY got my neo2 back from MSI today - so i have preliminary max stable OC here

3200+ CBBID 0451EPFW
2x512 TCCD
Neo2 Plat
Zalman CNPS7000B-CU

stability test: superpi 1m/2m (did both) and 3dmark05

ive got 2.5ghz, 250x10. 1.65v (10% Overvolt in BIOS) 2.6 was unstable, havent tried anything inbetween

memory: 2T (1T was unstable) 2.5-3-3-6 1:1 250mhz 2.6v
 
just out of curiosity, how far can you get at 1t, and with one stick in?

that's decently good for a week 51 though. i'm looking forward to final stability reports :D
 
final stability report:

2.51ghz, 251x10, 1.65v

HTT @ 1004mhz (251x4), 2.5-3-3-5, DC, 2T, 1:1 251mhz

1m/2m/4m/8m stable.

anything over this would crash out of 4m

ill get results with one stick and 1T later, but with two sticks, 1T wasnt even stable at stock... although that couldve been my board dying, as i got that result before i RMA'd the board... i doubt that though

2.51small.jpg
 
Do you by chance know what TCCD is under the heatspreader? Your results with your TCCD are terrable, there is a problem somewhere. At 2.5-3-3-7 1t, TCCD should be good anywhere from 250-300mhz. I dont know how PQI memory is, never used it, but it cant be that bad. If you have to run 2t at 2.5-3-3-7 250mhz, something is not working. Since you said that was before, try it again, but if it doesnt work, its problably time to start RMAing (assuming your hardware has not been abused and the warrenty has not been viod). Either the ram, or your CPUs memory controller are seriously underperforming.
 
lithium726 said:
ill get results with one stick and 1T later, but with two sticks, 1T wasnt even stable at stock... although that couldve been my board dying, as i got that result before i RMA'd the board... i doubt that though
Click to expand...
probably the memory controller. the neo2 is a decent board, so it shouldn't be giving you the problems.
 
Viper87227 said:
Do you by chance know what TCCD is under the heatspreader? Your results with your TCCD are terrable, there is a problem somewhere. At 2.5-3-3-7 1t, TCCD should be good anywhere from 250-300mhz. I dont know how PQI memory is, never used it, but it cant be that bad. If you have to run 2t at 2.5-3-3-7 250mhz, something is not working. Since you said that was before, try it again, but if it doesnt work, its problably time to start RMAing (assuming your hardware has not been abused and the warrenty has not been viod). Either the ram, or your CPUs memory controller are seriously underperforming.
Click to expand...
its not my memory. its my memory controller, and it is a known issue with winchesters - im gonna get venice as soon as i can for MSRP or lower to rectify this. the memory had no trouble running 1T in my aXP(to 240 - max of board) so said CBID
 
A64 3200 Winchester CBBHD 0450 SPAW
speed 2.75 ghz 1.595 vcore XP-90 with medium speed panaflo load 38-39c
DFI NF4 sli-dr mobo
I ran Prime95 for 36hours with no errors. I tried 2.8ghz with same vcore but got an error with prime95 after 3 min.
 
10 min. ago, the FedEx lady brought me 2 boxes. One was from Monarch with my hundred-twenty-seven dollar A64 3000+ s939:

CBBID 0507 FPAW - OPN: DIK4BI

This is a slouch, right?
 
the 2nd post is there for a reason. you can make conclusions just as well as i can ;)

however, the 3000+'s have seemed to have sucked lately.. so i'm not expecting anything great. there's one way to check though :D
 
You have valid points. It's too hard to say b looking at a chart of other peoples' PCs... Two different stepping code confuses me just as much.

Either way it doesn't really matter. I only paid $127 for a brand new chip, so I'm happy. It'll do just fine to get my DFI nF3 Ultra up when they finally release the damn thing...heh It does fine as a functional place holder for the enthusiast Dual Core when it comes out in s939.
 
You're right, either way it was a good price.

Server & EE dual core are coming out pretty much NOW, I think the best I've seen on Toledo so far said Juneish.
 
How long did it take for you to burn in Ashmedai? I have burned in 3 times with 10 hours each time, but my stability at 1.586 volts isn't improving. It did improve once, and I got 20 minutes Prime stable, then it went back down to being stable for only 4 or 5 minutes with no improvements.

Windows has also been corrupted I think, is there a good way to prevent against corruption while burning in with non stable voltages?
 
dmonkey said:
Windows has also been corrupted I think, is there a good way to prevent against corruption while burning in with non stable voltages?
Click to expand...


use a Windows PE cd and put the burnin program on there. Boot from the CD, run the burnin. When you want to test, just reboot without the CD.

if you're not familiar with WinPE, google up Bart PE and it should be the first or second hit.


I'm gonna start a seperate thread for this burnin stuff. After all, it IS off-topic (and I apologize if I'm the one that started it :p ).
 
Thanks, good idea about the new thread too, i don't think Eclipse appreciates all this talk about burn-in in his database thread.
 
yeah, thanks. i actually suggested that the tread would be made a few days ago. ;)
and i'll try to get the pic updated sometime soon. i was having fun with my ram earlier though :D
 
Firstly, Bart PE != Windows PE. If you look at his site he goes to quite some lengths on the topic.

Bart PE or some incarnation of Live CD style Linux strikes me as a very good idea for overclocking work, however I haven't gotten around to doing anything with it myself. Bart PE wouldn't be redistributable, however the Linux angle would be, therefore I'd think that would be prefered. Overclockix was suggested to me as an existant solution at one point but IIRC I had trouble with their webpage and never got back to it.

I mostly did 4-hour runs with this program when I was going to go off and do something else for a while, plus a couple times overnight when I didn't need it to be doing memtest. You can also use Prime to do it. Don't ever use Folding @ Home for burn-in / stability testing!! If you do everyone within 500 miles will come over and kick your ass. It messes up the work units. Prime's got stuff designed to let it be used this way and has instructions on how to go about it on their site.

I can't say much either way as to whether burn-in is valid or not, there are too many other variables that could be involved and the amount of testing + time that would be involved is much more than I want to undertake. However CPU load testing is a good thing of itself, and tuning a system after the initial overclock is a good thing too especially after the thermal compound has been through a few more cycles. Maybe it's just that it helps the thermal compound set thus giving you a bit more margin than you had before. No clue. There's so many variables involved that it can be hard to point at what change(s) are at work. But burn-in resembles good practices that you should already follow, so I don't see why it matters much either way.
 
Yah...somewhat OT I know. *sheepish* You can tell the part of it I care about though. Burn-in, boring, pointless, over-argued. F@H for stability testing, happens too often. Started the post before the other thread got mentioned or I'da hopped anyway, too late now.

I'll get ya the info on mine before I send it back, I just don't feel like unmounting my XP-90's clamps & breaking out the rubbing alcohol when I'm going to do it in a couple days already anyway...
 
gotcha. and i'm glad to know that i'm not the only one who doesn't see the merits in burn in with cpu's.

and i'll be waiting for that cpu of yours :D
(i know it has a decent memory controller. hehehe)
 
CBBID 0508 APHW

Max clock I got it to was 2592 with 1.6V on air. It's a 3000+ so a gain of 792.

Memory topped out at about 255MHz stable in XP.

Curiosity won out. ^_^
 
And of course now that I got it apart, I started moving everything over to my new case, and packed up the CPU to get shipped out. I just downed my main rig for about a week over idle curiosity. Heh. And I feel good about it.
 
Easiest to post my Signature from DFI Street

AMD 3000+ Winnie ADA3000DIK4B1 CBBID0502APMW
@ 2250 (250MHz HHT) @ 1.55V (110% setting)
Thermalright XP-90 w/ Vantec Stealth 92mm
DFI Lanparty NF4 Ultra-D
2x512Mb Crucial PC3200 CL3 (CT6464Z40B) @ 205 2.5-3-3-7 @ 2.8V
Gigabyte NX66T128VP (6600GT VIVO -500/1120 default clock)
Creative Audigy 2 Platinum
WD 1200JD 120Gb SATA HD (8Mb cache)
2x 160Gb Seagate 7200.7 SATA HD (8Mb cache)
Liteon 166S DVD (Pri Slave)
NEC 2500A DVDRW (Sec Master)
Liteon 52246S CDRW (Sec Slave)

Hiper Type-R 480W True PSU (24 Pin)
LianLi Case
Dell 1901FP TFT
Trust 5.1 Headphones
MS Intellmouse + Internet Keyboard

Originally I had a ASUS A8N-SLI with the set up and the processor wouldn't do more than 230MHz HTT and that was a struggle. I haven't tried o/c much with the DFI so I think there is more to come :)
 
hehe, sorry guys, i've been getting a bit behind. updating database now.

lithium: 2510 turned out to be the max speed at that voltage?
 
Well eclipse....I will have more results for you soon. I got a 3200+ whinny today to play with. I will have the stepping, as well as its results monday or tuesday. Cross your fingers
 
(cf)Eclipse said:
hehe, sorry guys, i've been getting a bit behind. updating database now.

lithium: 2510 turned out to be the max speed at that voltage?
Click to expand...
thats what pi said, but i got a random hard lock so i took it back down to 2500 and i havent had a problem yet.
 
You must log in or register to reply here.
Back
Top