Doesn't support newer boards now, the author stopped further development. You could try systool instead.
EDIT: looks like i was wrong :p:
Printable View
Doesn't support newer boards now, the author stopped further development. You could try systool instead.
EDIT: looks like i was wrong :p:
Thank you :toast:Quote:
Originally Posted by TEDY
2906MHz 1.5Vcore ST2004 stable for +1h, CPU -12ºC under Small FFTs stress.
Hmm, I think this CPU is more stable now with 5.10-2 BIOS :toast:
so i should change my 5/10-3 bios for -2 ?
i have bh-5 pc3500 twinmos
Give 5.10-2 a try :banana:Quote:
Originally Posted by TEDY
Yep better change it.Quote:
Originally Posted by TEDY
I can do 262MHZ(BH5) single chan with the 510-2 bios on 3.18V.
Unfortunely my other stick does barely 243MHZ single chan :(. (orange slots)
http://www.benchmarkhq.ru/english.html?/be_cpu.html
For cpu tools :)
CBBLE 0512 DBPW
found S&M , it does really stress CPU to max......anyone knows how to loop it ??
Really don't know and i would be glad if someone knew because then i can throw the other 4 stress test in my recycle bin :DQuote:
Originally Posted by TEDY
I ran some superpi and it went great ;)
http://www.ocxtreme.org/forumenus/at...tachmentid=110
Nice :)Quote:
Originally Posted by NiCKE^
3006MHz 1.5V stable now for 22min SP2004 :stick:
Sounds very good Dani!
Measure for yourself and get the truth.Quote:
Originally Posted by TEDY
Check it w/ a DMM - find out what it REALLY is :fact: (Prolly better than you think). If not at least you know for sure.
Smartguardian usually reports my 12V as 11.77. MBM shows 12.05 and so does my DMM :)
3GHz under StressPrime2004 full load:
Smartguardian: 12.16V
MBM5: 12.47V
DMM: 12.47V
My rails are always under what they should be, wonder why...
Play'd some more with superpi now and got better results :D
NiCKE^: I'm guessing your "OCZ 520W" is a Modstream? If not just fix the rails ;)
0517BPDW 3004MHz 1.5V :D
http://koti.mbnet.fi/dani69/DFI%20nF...re%201h11m.gif
Yepp it's a modstream.Quote:
Originally Posted by exscape
Love my Powerstream :) A voltmeter is light years better than any motherboard sensor. According to SmartGuardian the 12V is now 11,83V - according to the voltmeter its 12,03 under load :)Quote:
Originally Posted by NiCKE^
Yeah mine is also over 12v with a multimeter ;)
0516 CPBW is very bad ,is it max????? :soap:
http://wowgame.free.fr/3000+%20939/L...AW%20decap.jpg
for the moment prime 1.504v 2906Mhz full load 32°C :banana:
before IHS removed prime 1.52v same speed 43-44°C full load :D and can't pass over 2H30 at 1.504v...
I hope no one minds I have copied the database over to ExtremeOverclockers and encouraged someone with time to go through their Venice overclock thread to complie more data.
overcrash...holly glory!!!!!...watercolling without HS ?
2900 at only 1.5v wow...that's 1.55v in bios :)....only 32°c ? wow....tried S&M at 100& ? :=)
lol load temps lower than my idle by 6 lol..
maybe i should take it off as well when i get my venice :D
I'd be quite pleased with that Aeoplus! I paid for a 3800+ CPBW and run at your speed! You bought a 3000! + CPBW appears to be a sub 3ghz, but well see about that when my gskill LA arrives :woot: good jobQuote:
Originally Posted by Aoeplus
yes Watercool without IHS and 1.55v selected into BIOS, i've tested 2950 1.52v isn't stable more than 1H ,now testing 1.552v (cpu-z)Quote:
Originally Posted by TEDY
I wanna take the ihs off with my wc cooling but i am bit scared that i crush the core. :|
What a *ell is wrong with my 3200+ 0517BPDW :confused: :confused:
http://koti.mbnet.fi/dani69/DFI%20nF...re%201h25m.gif
Batch number is in high 700s :(
So far 240*10 at default. Probably the worst 0517 there is :(
cooling box :confused:Quote:
Originally Posted by Juliette
My 3200+:
ADA3200DAA4BP
LBBLE 0517GPPW
Z644446E50709
240*10 at default vcore and no more.
Will try water later, this stock isn't doing me much good. Just found out my winchester is suffering from the concave syndrome, hope that will beat my venice once i get it lapped :mad:
First test with my new 3000+ Venice.
Bought it last week in a local store (it was the only 3000+ I could find in 3 stores) but had to wait for my DFI to return from RMA.
LBBLE 0517BPEW
Zxxxxxxx0589
I had some initial problems with the cpu running stable, using the 510-1 bios on my SLI-D.
Flashed back to 310P bios and seems running fine now.
Had to change some memory timings (2*512MB Corsair PC3500) though, but still searching and tweaking.
Anyway,
Managed to run SuperPi 1M at 2900Mhz, 1.61-1.63V with watercooling.
Right now, it's running Prime at 2700Mhz, 1.47V. Not bad for a start, I'd say.
http://users.skynet.be/MetalHead/300...ce/2900Mhz.jpg
http://users.skynet.be/MetalHead/300.../2700Prime.jpg
My batch number is 892. :slapass:Quote:
Originally Posted by Juliette
But yours does 2600mhz. Mine won't :( Can't seem to get over 2400mhz stable, even eat 1,6V
Bah! I've had enough, this is one terrible CPU.
It's being RMA'd and has been agreed to by the company I bought it at :)
has anybody an 0519???
and whats better?
an high or low batchnummber?
i guess low batchnumbers.
holly :banana::banana::banana::banana:
CPU BURN-IN and running S&M really helped.....
At 2800 I started with 1.5v, now i'm at 1.485v.....will try 1.475 tomorrow.....that's settings in bios :>
Quote:
Originally Posted by TEDY
what cooling do you use?
i will get a 0517 dpmw too :D
2808 @ 1,5v seems very good :slobber:
Hmm, maybe I will use CPU Burn-In and S&M
Hope I can get my BPDW 3.1GHz stable at 1.5V :woot:
Im stuckt at 3056MHz 1.5V StressPrime2004 stable :slapass:
Mine is 0589.Quote:
Originally Posted by Mr.President
I know a guy who has 2 3200+. One has a batch number 5xx, the other 9xx.
The second one (9xx) clocks better than the 5xx one. Pure luck?
lower batch number is just plain BS
Perhaps. From my own experience a lower batch number was always better - I'm giving my own venice another un, I hope it's only a ram issue.
A sucky Zalman 7000B-AlCuQuote:
Originally Posted by Mr.President
Hey, you've wasted two thread spots with this and haven't yet explained your reasoning. I made a statement and then offered an explanation of why I had come to my conclusion. You basically are just saying it to say it. Please, let us know why you think the way you do.....Quote:
Originally Posted by ReelMonza
Can you show me some OFFICIAL AMD info regarding these numbers ?
I had a 0019 and a 0243 JIUHB XPMW, the 0243 OC a LOT better than the "lower batch" one.
Lots of people already shown that lower batch numbers doesn't always translate into higher OC's, so if I were you I would stop telling ppl to get "lower batch" ones cause they OC better...
EDIT: 0243 could do 2.55GHz @ 1.9v and 0019 would do the same @ 2.1v
I've got an LBBLE 0518APFW 3200+.......however I won't be able to use it until DFI releases their nf3 socket939 AGP board, or if they don't do that, then I'll have to sell my current GT to go PCI-E
Yep, my old FX55 had 0055 batch number and it did'nt even do 3GHz SP8M stable at any Vcore with VapoChill LS and 3 different mobo's :fact:Quote:
Originally Posted by ReelMonza
So that's it with the low batch nro :fact:
First off, if you would care to read what I wrote, you would know that I said that a low batch number doesn't always translate into a better oc'er it also depends on how good the wafer the cpu was cut from. And second of all, I never told anybody to "go get lower batch ones" as you say :rolleyes: Just because it wasn't the case with your 2 cpu's, it doesn't mean that applies for all other cpu's produced. Also, what's it to you if people end up looking for a lower batch number cpu when they decide to buy? Hey, if it can *maybe* increase the chance that the cpu was build on some solid silicon, then why not? Sure as hell, it wouldn't hurt.......BTW, are you sure that your cpu's are from the same wafer? i.e. They share the exact same serial number besides the last 4?Quote:
Originally Posted by ReelMonza
didn't the early winnie batches OC a lot better than the later ones? Of course you'd always find one that'd still OC like mad, but when compared in the masses didn't their OC's become a little worse?
In the cases of venice there was such a high demand for them especially in the beginning it seems they were just dumbing down otherwise superior chips, but this is just my speculation
@s7e9h3n, I asked you some official AMD information on what those numbers really mean and you didn't provide me, without such info it's just speculation for me and it'll still be no matter what ppl say so I don't care about those numbers mean anyway.
What you're referring to is the manuf. date and stepping. The batch code is in the serial number of the cpu.Quote:
Originally Posted by Nubius
Just a note: People, don't pay soooo much attention to that batch number. (Don't get me wrong, if I had a choice between a lower numbered cpu and a higher numbered one, personally I'd chose the lower just because.) If you're gonna buy a cpu based on the numbers on the IHS, then look first at the cpu's stepping and then go from there. Even with lowered numbered cpu's, as with every other cpu in existence, it's overclockability isn't guaranteed. Some will be studs, and some will be duds. Here's my example of a high-batch numbered stud: My Venice 3200+ LBBLE APAW batch code in the 5xx......
http://img215.echo.cx/img215/6171/cp...erified0qy.jpg
My 0515APAW is batch code 0340 and my max stable is 2760, 10 Mhz above my 24/7 speed; i hit an instant wall there.
The explanation offered earlier in a thread was that the lower batch number signified a core closer to the center of the wafer, so being closer to the center means it's better
awesome results overcrash! I should call you OC for short, it would be fitting ;)
EPAW seems much better than EPEW*
There are some things AMD doesn't document "officially" in there public records. If don't believe anything that isn't "officially documented" by AMD, then I'm sure you don't think that the .13 FX55 is built on strained silicon either.....Quote:
Originally Posted by ReelMonza
@s7e9h3n, I didn't said "I don't believe anything that isn't "officially documented" by AMD", I just don't believe that the "lower batch number" theory is actually true cause it has been proved many times that the "higher batch" ones can clock higher than lower ones.
You know, I think everybody may be misinterpreting what I was trying to say. The idea that a lower number batch code guarantees you a better OC'n cpu is not what I wanted everyone to take from my statement. It doesn't necessarily hold true for all instances, but in my experiences, it has many times. OK, think about it this way......If a wafer is a circle, and the cpu#0001 is stamped from the center, then how do you think they'll arrive at the numbers 2-????. My guess is that they'll stamp in a spiral pattern of some sorts so if a certain area of the wafer contains better silicon than another, the batch numbers from the resulting cpu's will vary. I just tend to lean towards the lowest batch numbers since, chances are, the silicon is better towards the center of the wafer. I'll try to get OPB over here....he had an awesome FX53 with an 0002 batch number......Quote:
Originally Posted by Absolute_0
Just Google it and you'll find out what the numbers mean......Quote:
Originally Posted by ReelMonza
It sounds like a good theory to me, and it's certainly better than no theory. Of course i recognize that a lower batch number won't always OC better than a higher batch, but if it seems to be a general trend, then it supports the theory. Seems like lower is better to me...
I just found SPECULATION about it.Quote:
Originally Posted by s7e9h3n
So can you please point me to the document that states that the .13 FX55 contains strained silicon? Or if I google it, will I only find speculation? :rolleyes:Quote:
Originally Posted by ReelMonza
Or maybe there's a document or two about the A64 3400+ 939 pin cpu...or is that also just a myth?
Say what you say, I'm saying it's a myth and that's what it is, again it has been PROVED many times that so called "lower batch" ones won't clock better than "higher batch" ones, it's so hard to you to admit ?Quote:
Originally Posted by s7e9h3n
Hey, I have no problem admitting that there are higher batch numbered cpu's which OC better than lower numbered ones, all I'm saying is that the majority of higher clocking cpu's, in my experiences, tend to be from lower batch numbers.....Quote:
Originally Posted by ReelMonza
Please guys, there is no need to take this in the thread, start a new discussion thread about it and it will be very on topic ;) I believe s7e9h3n's theory because it sounds logic.
Alright, if you have kept that to yourself we wouldn't have to go through this , someone else's theory would make sense but in reality it doesn't work that way.Quote:
Originally Posted by s7e9h3n
I must confirm s7e9h3n's theory/speculation since most of my cpu's with a lower batchnumber did better then my higher ones.
But then again I had a few that did the opposit.
But I still think the chance of a higher oc is with lower batchnumber (or just a low one)
Ok, I have no idea what you just said, but if you're speaking from personal experience, let me ask, how many cpu's have you been able to test, say...in the last 3 months? If your answer is your 2 dlt3c's, then you're basing your theory on either your personal experience and/or what others have said. So basically, you're speculating. If I say that I've tested probably at least 10x the number of cpu's you have in the past 3 months and give my PERSONAL experiences, then you tell me - how can you possibly make your statemnet statement and yet have still not offered any evidence to support it?Quote:
Originally Posted by ReelMonza
Hey dude, it's not me, just browse the web, visit the forums or google and you'll find at least a thousand "higher batch" cpu's that perform better than "lower batch" ones. you know that's not true. if you really tested hundreds of cpu's you know that wouldn't always hold true. So how do you explain lower batch ones not being quite as good as higher ones based on someone else's theory ?Quote:
Originally Posted by s7e9h3n
If you read what I wrote I NEVER said every lower batch number is better than every higher number. All I said was that chances are that the lower numbers will be better since they're closer to the center of the wafer and I did explain my theory on why lower numbered ones may not be as good as some higher numbered ones in a few posts back.....Quote:
Originally Posted by ReelMonza
There is other factors that can limit the OC like with CBBiD and Winchester, then it almost doesn't matter if you got a dream week like 0447SPAW 0001 and then the CBBiD will limit you :D But please link me two three examples where 2 cpus of the same sort are compared, high batch vs low batch so we can try to figure out this.Quote:
Originally Posted by ReelMonza
I'll tell you right off the bat that the x2 4600+ I sent OPB with a batch #0029 is definately better than the current #0053 x2 4600+ that I'm testing now. It's got a stronger memory controller and is more stable with less vcore.....I'll have pics up of my current cpu later.... ;) Just a note: this comparison shouldn't mean too much since the cpu's ARE so close in number.......Quote:
Originally Posted by NiCKE^
Calm down guys, no need to start discussing this here.
Use a PM, or MSN, AIM, whatever if you wanna argue about it.
But not in this topic, so let's keep it clean and not to ruin it for the others.
After all, finding a good clocking chip is still a crapshoot, isn't it?
I don't doubt it at all. http://home.no/nickeh/cpu3.jpg this was the dream stepping in Sweden and those who were first to order from the Swedish webshops got this stepping. And look, 0019 as batch number ;)
*EDIT* AS Jupiler says, it's most luck when you get a good cpu but what I think everyone can confirm is that if you get one of the first cpus ever made of a model the chances of getting a high clockers is much higher, right? *EDIT*
It always is, but sometimes you gotta know how to play to help your chances of winning ;)Quote:
Originally Posted by Jupiler
Geez Nick, you think you could make the picture a little bigger? I can't quite make out the numbers :rolleyes: :p:Quote:
Originally Posted by NiCKE^
Hehehe, you can't complain that it isn't readable :)
OK, but what if those numbers have nothing to do with position of the wafer ? that's why I say it's plain speculation. If you can just point me some reliable info on it that'd be nice. unsubstantiated rumors can lead to this kind of talk, for me you are spreading rumors but have shown no facts.Quote:
Originally Posted by s7e9h3n
were those two "from the same wafer" ?Quote:
Originally Posted by s7e9h3n
:rolleyes:Quote:
Originally Posted by s7e9h3n
How many batches do you suppose there's been for the 4600+? :rolleyes:Quote:
Originally Posted by ReelMonza
you tell me, post some pics.
3200+ LBBLE 0518 APAW @ 2800MHz Watercooling
Which vcore furyfax?
They're the same stepping. I'll post pics of the new one when I have time. You'll find pics of the current one if you do a search.Quote:
Originally Posted by ReelMonza
This is really starting to get out of hand......Let's finish this up once and for all. You make the claim that that it doesn't matter what the batch code on the cpu is in terms of how well it overclocks and I say that the lower numbered batch codes tend to do better than the higher ones. Now for anybody to take your claim serious, why don't you just post a few links that state exactly what you claim and then maybe it can be considered......until then, it is you my friend that is speculating.......
Here are just a few that state exactly what I claim:
http://www.lowyat.net/guides/athlonxp/
http://www.xbitlabs.com/news/cpu/dis...322034857.html
http://www.aoaforums.com/forum/showt...t=16134&page=4
http://ocprices.com/forums/showthrea...oto=nextnewest
http://www.idhw.com/textual/guide/no...d_k7label.html
This link is for you......read the middle of the page or so. It talks about DL3TC batch code #'s.
Why don't you just do it now ?Quote:
They're the same stepping. I'll post pics of the new one when I have time.
I guess you had plenty of time to put all that together.
Those were the same exactly links I found too, btw I read those a long time ago.
EDIT: This theory first appeared on the web back in 2003 at the Lowyat forums and all those links are nothing but news/guides on the same Lowyat article.
There is nothing you can tell me buddy, this is just speculation and always will be.
Nice links :rolleyes: ...end of discussion.
... and X2 cut from the "same wafer" pics... and yet you can't tell those numbers mean what they're supposed to mean...
Can someone write an mail to S&M programmer to say him that S&M endless loop would be great :banana:
BTW, 3005MHz @ 1.5V on BPEW now S&M finished and StressPrime2004 is on now
lucky you :mad: cooling with icewater? i want also 2.9g primestable. vut u cant get 3ghx primestable ?
no ice, just water outdoor, but in night :DQuote:
Originally Posted by lehmin
i think 3Ghz possible with ice, and 3.2-3.3 with chiller...
1.65Quote:
Originally Posted by NiCKE^
i might be able to choose a venice core...
can any of u perhaps give me a resume of what to look/ask for?
after studying the list on the first page, these are the ones that come above:
3200+ LBBLE 0515APAW
3000+ CBBLE 0512DPAW
3000+ LBBLE 0517EPAW
3200+ LBBLE 0517BPEW
3000+ CBBLE 0512DPBW
3200+ LBBLE 0515GPBW
anyone who knows more then this? plz lemme know...
anything else i should look for?
I'd sure like to be sitting with an 0517EPAW right now, but I guess my 0516EPKW will have to do. :rolleyes:
3000+ LBBLE 0517EPAW is the easy choice ;) Seach for LBBLE 0517EPAW and see what results you're getting.
Personally, I like the 3200+'s 10x multiplier:
mem:fsb 1:1 ON AIR LBBLE 3200+ 0515APAW
http://img31.echo.cx/img31/5254/10x295spi1m6tf.jpg
Damn man my won't even do close to that :(
No worries Nicke, not many will ;)Btw, I'm sure it'll make you feel even better when I tell you that it was with a stock fx heatsink :p:Quote:
Originally Posted by NiCKE^