A64 OC Database & Overclock Guide

SpoogeMonkey said:
2.73ghz w/ram @ 200mhz, just isn't as fast as current ram settings.
Click to expand...

I guess I need to validate this statement : Running 2.72 @ 195 (not 200 as said above) results in a loss of 4.6 sec./frame when crunching the same 10 frames in FAH. (4.6sec x 400 = ~30min/project) Not THAT big of a deal...just a lil slower.

Now I did go and start from scratch, since this a new mobo, and previously went by what I KNEW this cpu could do....and came up with the following.....

2.73ghz 1.47vcore (342 x 8) 3 x htl ram 228mhz 1t 2 3 3 6 133 div.
This actually makes it ~1sec faster now...woohoo :rolleyes:


And thanks again to eclipse and baron for taking the time to make the guides, and do the benchies, and offer all the advice.
 
Also, if you have to drop the frequency down because of dividers, couldn't you tighten up the timings from 242mhz 1t 2-3-3-6, to 2-2-2-6 1t at 200mhz?
Click to expand...

Theoretically yes.....once my comp is actually up and running and I've got my PSU back I'm gonna be curious to see what my chip'll do for some quick superpi 1M runs at 200MHz DDR 2-2-2-5...just for shits and giggles. :p
 
this is getting a little rediculous guys. not to be an ass, but go read rule #1 again :p
 
Just hit 2.7 @ 1.425v with my 3000+ venice stepping in my sig. All the other stuff liek mobo is wrong. Running a DFI non sli board now.
 
TehDuffman said:
Just hit 2.7 @ 1.425v with my 3000+ venice stepping in my sig. All the other stuff liek mobo is wrong. Running a DFI non sli board now.
Click to expand...

By "just hit" you mean you got that far, or it's Prime stable for 6 hours or more? If it's actually Prime stable for 6+ hours then that's one hell of a chip.....seems there are more and more of 'em around the capability of mine popping up all over. ;)
 
TehDuffman said:
Just hit 2.7 @ 1.425v with my 3000+ venice stepping in my sig. All the other stuff liek mobo is wrong. Running a DFI non sli board now.
Click to expand...
STEPPING! (like LBBLE or CBBLE, the letters before the week)
and cooling too.. :p

and corn, it's cool.. i have the same tendancy. i think there's a reason we're junkies :p
 
3700+ San Diego

ADA3700DAA5BN
CABGE 0516MPMW

Max Stable so far
250X10 = 2.5Ghz
1.47Vcore
Coolermaster Hyper 6 (Air)

I have a new OCZ Powerstream 520W coming to replace my generic one. Im praying that this will help my crappy OC results. I seem to have gotten one of the worst chips so far, which is just my luck... This is my first Overclock though, so for a first timer im not too disappointed. I have asked for help but could never find much other than "sounds like your chip is a dud"

Stepping.jpg
 
AMD X2 4400

ADA4400DAA6CD
CCBWE 0522TPMW

My max stable OC so far is
220 x 11 @ 2.4Ghz
Stock Voltage I think its 1.3V
With an XP - 90

Im limited by my ram cant go over 225 fsb.

OC.JPG
 
Scoob () said:
Im limited by my ram cant go over 225 fsb.
Click to expand...
DROP THE DIVIDER AND GET BUSY! :p

budwise, i appreciate you posting. you're the first one with a 'dud' 3700+ to actually follow through with posting ;)


i'd update tonight, but i have no time. silly life :p :eek:
 
budwise, if you go back to page 8 and look 2 posts down you will see info on a "burn in" method. If you give it a try, you might see an improvement. Also, the vcore is still pretty low, does increasing it not give you any headroom?
 
Venice 3000 @ 2.52ghz (280x9)
+720mhz
Air Cooling @ 1.55v

Stepping LBBLE
Week 0519
Code XXXX - Very Sorry, will get it ASAP, Stepping and Week stuck out most in my mind I guess, LOL

Setup:
DFI-SLI DR
2x512 OCZ PC3500 (217mhz) Cas 2.5 DDR
Coolmaster Hyper6 HSF w/ 80mm Masscool Sleeved-baring fan
Antec P-160 (2x120mm Fan)

Bios Settings:
VID - 1.55v (Reports as 1.5 with Smartguardian)
150mhz Base mem speed
HTT x 4

Idle - 38c
Load - 250mhz HTT - 46c
Load - 280mgz HTT - 50c
Chipset - 47c
PWM - 47c

Booted Windows @ Base clock w/ 150 memory bus. Generated Load using Prime95 (small FFT). Used Clockgen during testing in 10mhz increments, taking temp samples after 5 minutes of load.

Board Stable to 330mhz. Some Ram instability beyond 220mhz, although I was stubborn with lowering my ram timings, finally settling for Cas 2.5/4/4/10/1T.

I was able to Prime for 30+ minutes at 287, but 290 was a no go at 1.55. Primed stable at 292 using 1,7v - my Load temp shot up to 60c and I decided it was not worth the headache for another 90ish megahertz. Not bad for a sub $150 chip, IMO.

Feel free to PM me with any questions or if you see room for improvement by all means let me know. I will pop the HSF and get you the CPU code number if you require it for accurate record keeping, though I personally feel the week number is the most relevant number after the stepping.

To the OP - thanks for compiling this list - It made my decision to go from Winny to Venny (along with PCI-ex, ATX2.0, etc) an easy one to make. It is appreciated.

-Woe
 
budwise, if you go back to page 8 and look 2 posts down you will see info on a "burn in" method. If you give it a try, you might see an improvement. Also, the vcore is still pretty low, does increasing it not give you any headroom?
Click to expand...

Just what I was gonna say....it's a sad shame that so many people never end up even TRYING a burnin phase with their Venice/SD chips.....does help a fair bit if you're spending any reasonable amount of time OC'ing it.
 
cornelious0_0 said:
Just what I was gonna say....it's a sad shame that so many people never end up even TRYING a burnin phase with their Venice/SD chips.....does help a fair bit if you're spending any reasonable amount of time OC'ing it.
Click to expand...
yeah, I had to put mine on pause because my insulation still isn't where I'd like it to be, but I had gotten up to [email protected] bios, 1.525 reading, where as before I had run at least 1.7something. I can't say it was all the burn in, as I did add antifreeze, but that only dropped temps ~4-5C, not enough to justify results that big.
 
I've been seeing more results for 3000's than 3200s. Do the 3000s OC higher than the 3200s?
 
Glow said:
I've been seeing more results for 3000's than 3200s. Do the 3000s OC higher than the 3200s?
Click to expand...
oi! go look at the chart and figure it out yourself :rolleyes: :p

though, one thing to know.. the 3700+ and 3000+ data is a bit skewed, the average mhz should be a bit lower, as i've often complained. a lot of people with oc'ing duds aren't reporting in :(


Woebringer, i'm making an exception for you, but get that batch code QUICK! :D

installing excel again so that i can update
 
i have tried to get 2600 Prime stable, but even at 1.6Vcore it got an error in Prime after a half hour or so. Ive been burning it in at 2500 @ 1.47Vcore for now. When my new Powerstream comes in im gonna keep trying. Its been at 2500 for about 2 weeks now, i figure thats prolly enough burn in time. Although i did try to lower the Vcore at 2500 last night down to around 1.44-1.45, but it failed prime in about 10 minutes with that voltage. So far it likes this speed and voltage, but i will try more when the new PSU comes...

As far as a burn in method on page 8, i dont see any. My screen is prolly at a different resolution making a different number of posts per page. What Post # is it?
 
(cf)Eclipse said:
this is getting a little rediculous guys. not to be an ass, but go read rule #1 again :p
Click to expand...
twice in one page :(

i should just give up and get rid of that rule, but it makes it a pain cause i have to dig through everything to get the info i need.
 
UPDATE: My 3700SD is prime stable 2600 @ a whopping 1.66 Vcore. Its stable, but hits 49-50C at full load. This is too much voltage for me, especially on air. What a depressing OC...

260X10 2600
166 Memory Divider


I will try a little more still when the new PSU comes, but i felt bored today so i messed with stuff to pass the time.


Sorry for not posting "results only" :(
 
Running "naked" absolutely the best 2 hours i've ever spent o/cing

Model: 3000 VENICE
Full CPU Code:
ADA3000
AA4BP
LBBLE 0520APMW
Mhz: 2.610 mhz
Gain: 810Mhz Gain
Vcore: 1.50v
FSBxMulti: 290HTTx9 (Divider at 166Mhz)
Stable: 32M SuperPi, All benches completed (3dmarks01, 03, 05)
Temps. 35C idle / 48C load
I'll add pics. as soon as my web space is running again.
 
TehDuffman, you're still missing part of the OPN
Woebringer, you are too :p
Scoob (), i'm waiting for results that aren't ram limited before putting you up.. also, don't be afraid to bump up voltage. stock shouldbe 1.35-1.4v

working on getting the updated DB, and the new X2 part up right now, give me a few min :p
 
TehDuffman said:
LBBBL 05166GPEW
Click to expand...
still not right. based on other steppings, and what the 2nd part should be, i'm guessing it's an LBBLE 0516GPEW, which is what i'm putting you in as ;)

baron, you better produce a longer prime test than that :D
 
Eclipse: just a suggestion, I think it would be very helpful to others if you also posted up what motherboard each overclock is using and even what BIOS revision. It is almost just as important to understanding the overclocks as the CPU info itself. Its just a thought, cuz i know i sure would like to know what mobo's all those overclocks are using...
 
Budwise said:
Eclipse: just a suggestion, I think it would be very helpful to others if you also posted up what motherboard each overclock is using and even what BIOS revision. It is almost just as important to understanding the overclocks as the CPU info itself. Its just a thought, cuz i know i sure would like to know what mobo's all those overclocks are using...
Click to expand...

It's fairly obvious what the two on this page are using (p0temkin's and mine) since SmartGuardian is a DFI-only thing.
 
i would, but it's kinda more effort than it's worth. i'd have to go back and get answers from all the people who have already posted, which would be a pain. maybe next time.
 
Just ordered my 4800 dual core, I'll quickly put it under my vapo and get some data for ya before I ship it off to datacenter. Should have it benchable by friday.... hehe cant wait :D
 
247 X 11 @ 2700 Ghz
1.5 V
Temps at 49C

prime stable for 30 min so far, will continue with prime95. Will post a screen shot if it stays stable for more than 5 hours.
 
What about the 90nm Sempron Palermo cores? I gotta a 3100+ coming in a day and there's nothing on one. X bit labs got one up to 2.68 Ghz and it rolled over the A64 3500 with no trouble.

I'll submit something after the build is done.
 
You must log in or register to reply here.
Back
Top