Printable View
got my baby today:
http://img399.imageshack.us/img399/4...602xpmw4mw.jpg
I noticed there's a T rather than N in the corner, I looked at FX-60 and Opteron dual cores and all the ones I've seen were T.
however, most recent 3800+'s and stuff are N and a few rare T's.
Any ideas what this means?
In my experience (note this is not LAW) I have found T chips to be voltage loving. I have also found them to be more resillient to BH-5 etc high volts to MMC. I have also found that high clocks with low volts = dead T chip.
N tends to clock higher on lower volts and tends not to like high volts which means it requires fine tuning tweaking and fine voltage increments to get the max. Contrary to T chips they tend not to be as resilient to BH-5 etc high volts to MMC. I've found that high volts with high clocks = dead N chip
These are just my views from my experience. Others may experience different scenarios.
hmm I dunno if that's good or not lol...
so hopefully itll be good with about 1.5v/1.55v 24/7
aslong as me xp-90c can handle it (if not I'll just cash out for something bigger and better)
what is your AMD64 TCaseMax ? http://www.thecoolest.zerobrains.com...topic.php?t=83
I'm still trying to get cash for a DFI atm (waiting for a refund on a dead proc)
so I don't know
well, as far as ive seen, all ES and MS chips have been N's, (one B, in link). so i think it has something to do with where its being built.
http://www.engineering-sample.com/in...8%20S754%20MS/
i was wondering the same myself..............i did mention to chew that my experience with T & N was different.......subscribing to find out more :)
N chip note default volts...
http://img246.imageshack.us/img246/8646/4600b5zg.jpg
interesting i didn't look at that
my Opti165 T chip doesn't have default volts written on it but in windows at auto vcore it shows 1.328v as does the 3800+ X2 N chip
N= 1.30
T= 1.35
Ircc
I think you need to compare them as X2 manchester vs X2 manchester and so on
The chip in the pic above 3.2 at 1.5V IRC
Mine which is a T chip 3.2 at 1.6v
It's not to do with default volts, the default volts are described in the top line (e.g ADA4600DAA5BV) cant remember which on those letters but one describes what the cpu's default vcore is..
maybe it has something to do with cache? all 1MB per core chips have T right?
My sandy 3700+ = N 1.35v where as other chips were 1.40 stock..........Quote:
Originally Posted by lawrywild
My x2 3800 is the same identical top line as the picture I linked above........default volts is 1.35
http://fab51.com/cpu/guide/opn-64-e.html
A = variable apparently..
Yep. :) Look at it like this. I think N is the PCB equivalent to TCCD and i think T PCB is the equivalent to Bh-5.
but I still don't believe the N and T show default vcore
it's something else...
I think N's were the ones binned for dodgy cache and T are just disabled..
yeh maybe but what you said is a load of bull, T's won't just die when using low vcore and trying to run high speeds... they'll just not be stable :stick:Quote:
Originally Posted by chew*
Well that can explain why one likes volts and the other is spotty with voltage. That could also explain the numerous N chip deaths.
except my last optys, all the T chip I owned had concave IHS. I had 2 3200+ "N" with perfect/flat IHS.
some pls put us all out of our misery and explain what the hell those two letters mean hehehehehehhe
Thats the thing people burn in at low volts to become stable and THAT is when they die..........Actually I know someone that just lost one but atm I can't discuss details. Low volts high clocks mind you. Also I'm sure some people that think OCCT stable is stable can tell you stories of dead cpu's. Another case where they think its stable at low volts when it really isn't. Or the people that prime for an hour and thus "Well its stable enough for me"Quote:
Originally Posted by lawrywild
Any input on this subject s7? I don't mind being wrong but I have compiled results from users for quite some time now. This all stemmed from my stepping search which appears to have been a success......
No comment ;)Quote:
Originally Posted by chew*
Heh I kinda figured you'd say that. What we should really ask is the cpu killing master which has died more T or N chips at high volts. O and that FX 55 is a drool worthy stepping If my stepping calculations are accurate.Quote:
Originally Posted by s7e9h3n
BTW ROSETTA!!!!!!!!!!!
That FX55 is my "FX56" in my sig ;)Quote:
Originally Posted by chew*
:confused:Quote:
Originally Posted by s7e9h3n
Early batch of 55BN 'T' chip...eat volts for lunch:)
http://img236.imageshack.us/img236/8117/x18ry.jpg
cpuz read it as claw..
http://img65.imageshack.us/img65/7040/x21tb.jpg
And this 'N' chip with default volt....
http://img112.imageshack.us/img112/9746/x30ld.jpg
ive had two N chips that have seen 2v and more vcore. and ran 1.8v and higher 24/7
Yep screenies with some 24/7 prime then............ Or i could just say yah i have 2 T chips here that im running 4.0 volts through 24/7
Also should note winchester and older doesn't count................Nor does boiling water according to your sig...
"So far, in my humble experience, "N" chips are cold bugged to hell and "T" chips aren't. Anyone want to contradict that assertion? :)"
Sorry guys, I was talking cobblers. The "N" I have under phase at the moment turns out to be ok.
my cabye 0540 is an N chip and it can certainly manage cold
-35C 330-340Mhz HTT idle
Interesting.... I have a 4400 'N' rock solid at 2.6 with 1.41vcore.. anything above requires at least 1.5v.
2.8 was possible with 1.5v but generated a lot of heat which my Typhoon can't really handle as my tcasemax is only 55°. :(
More than similar experience here (same stepping, no signs of coldbug at the same temps at 355HTT)....and my "N" 4400+ loves volts (1.67V is its sweetspot on shoddy water).Quote:
Originally Posted by dinos22
I really don't know what they mean, but if s7eph3n has all Ts, I think I should get some :)
http://img372.imageshack.us/img372/7839/vap3my.jpg
http://img372.imageshack.us/img372/2526/10x2801m3bq.jpg
How's this 4400+ on plain water? :D BTW - IHS STILL INTACT ;)
Yes, sorry Dinos, I talked rubbish. I'm dual SP2004 this 0607 3800 "N" at 345x9, evap -34C and core -9C under load.Quote:
Originally Posted by dinos22
Hey s7e9h3n, that must be excellent internal contact on that 4400. Like I've got a 25C differential between coolant and core here, that's a lot (although 1.616v).
It may be more due to the fact that it's an ADV4400. My first review on one of these cpu's had it running dual 8m's at default speed @ 1.0Vcore on phase :p: Here's a quick run @ 1.63. (Don't mind the voltages as I was trying something else with these settings ;))Quote:
Originally Posted by stone_cold_Jimi
http://img512.imageshack.us/img512/3...2601mbh4bl.jpg
Oh yeah, I remember! Never seen one since though :(Quote:
Originally Posted by s7e9h3n
Here's an "N" that crashed at default because 1.328v core is too high subzero. It needed 1.23v on phase to be happy.
ok guys we are talking high volts though........... T and N mean nothing with coldbug trust me. I didnt start that coldbug thread for nothing. It has to do with how resilient they are to voltage. Etc i've run t chips with alot higher voltage 1.8.........and they love large voltage bumps. N chips seem to like low voltage bumps and stop scaling with high 1.75 to 1.9 voltage in most cases, they also usually die if kept there to long.......
Just to solidify this more S7 was asked to Obtain me a 3.2 24/7 low voltage chip( I have an aversion to running over 1.6v 24/7). Guess what letter it has.............
still doesn't really prove anything..Quote:
Originally Posted by chew*
maybe we should ask AMD
Or (and i'm not trying to be an ass) I suggest you do a days worth of searching on this forum for cpu deaths and max voltage to chips + max OC's. I did it for 2 weeks straight till I had them all compiled however i will not submit my results do to the fact that when you saw them there would be a mass buying spree for a particular stepping. Lets just say based on my analysis you could tell me what you were looking for in a chip and i could tell you the exact stepping that would give you the results you wanted.
I think this is the same reason stephen has "No Comment"
:stick: : I have really never liked the "I know but I'm not telling." Why tell us you know something if your not going to tell us? If it's a secrect, why not keep it to yourself instead of I know a secrect but am not telling? I'm not trying to be harsh with you, just asking questions. If you feel it is harsh, just go ahead and :slap: me .
lol yeh, if you know something you should tell really, that's what this forum is all about and why I created this thread..Quote:
Originally Posted by dogsx2
Ok well here's part of the conflicting issue. Those testing the chip on air water are not going to see the results I'm talking about. Being that T chips like higher voltage they run hotter. Once you eliminate heat from the equasion you will see what I'm trying to explain.
As far as steppings Go and my research you can ask anyone here that I am more than helpfull if you PM me. I'm just not going to make a thread about it as even with all my research it is still just theory. I can not gurantee clock speeds as each chip has a max, I can just lead you in the right direction of what your looking for.The answers are in this forum you just have to look for them.
Also the other reason is preety clear in this thread alone why i don't make a thread about it......You don't belive it and have questioned my knowledge on this subject through the whole thread...........
Dog's You of all people can verify that I have no problem leading you in the right direction via PM. Etc who approved your recent X2 buy........ I also take into account peoples cooling when making suggestions or approvals. Dog's I knew you had phase.......
Hey, I give as much info about cpu's as I possibly can (and in fact, I reveal probably more than I should). I'm not about to get myself or anybody I know in a comprimising position with AMD. There are reasons for NDA's and I'm not about to directly violate any of those just because I have that info....:rolleyes:Quote:
Originally Posted by lawrywild
my recent one is 72C my other is 67CQuote:
Originally Posted by Willis
I agree. It's a total load of crap. "T" and "N" obviously mean something. If it's signifigant or not I don't know. Most of my chips have been "T" and everything I touch loves voltage. Then again gocchin has my "T" 146 at 3ghz+ prime stable at like 1.45v. I also needed 1.9v for 3.6ghz on dice so who knows.Quote:
Originally Posted by dogsx2
I don't see how explaining the difference between a letter on the substrate could hurt AMD. But even worse is the "NDA" on the 7800GTX vddq mod and saying that they can't talk about certain drive strength setting info on the EXPERT. It pisses me off that they would hype something up like that that then say "sorry can't tell you" when it doesn't work.
well that comment wasn't aimed at you... even though you refused to release info :rolleyes:Quote:
Originally Posted by s7e9h3n
tbh guys whats the point of doing research if you don't release your results.. the way we get through science discoveries etc is by publishing our results and for others to come along and continue the work and experiments etc, it's the same concept here is it not?
Be pissed all you want, it still won't get anywhere with people who do know that info. It's not like I started any thread saying "I know XXXX and can't say". All I 'm implying is that the info is out there, it's just not supposed to be revealed for one reason or another. If, say, Jason was to tell you something and ask that you not tell anyone else, would you come on the forums and announce this info??Quote:
Originally Posted by afireinside
And my comment wasn't directed at you ;) I'm just trying to explain my stance on the whole issue. Some info I reveal is a product of my own testing and can be proven by my work and the work of fellow forum members. Other info isn't and therefore I can't produce thorough enough results to back up these claims.....Quote:
Originally Posted by lawrywild
Of course I wouldn't. Two things though:
1:He doesn't hide silly things (I'm sure the meaning of "T" and "N" won't cause a revelation)
2:I wouldn't go around blabbering "I KNOW ALL OF JASONS SECRETS AND TWEAKS FOR RECORDS AND JUST ABOUT EVERYTHING ELSE BUT I CAN'T TELL YOU SORRY HAHA" even if you're not making a thread about it you're still pissing people off by saying you know what it means but you can't say. And I'm not the only one that doesn't like it.
Ok you want info heres info. From my gathered results this is what i have come up with.
*EDIT* wanted to add that in my search I found that bottom line and top line chips fared best. Mid line chips did OK but not spectacular.
LDBHE = cold bug free
N on wafer= Low voltage higher clocks/lower temps to a certain point/possible death clocked high w/high volts (24/7 use) high volts = 1.6-1.65
T on wafer= High voltage higher clocks/higher temps/possible death clocked high w/low volts (24/7 use). low volts = 1.4-1.45
Week
Ive found the first weeks to be best
for instance 0501, 0601, and so on. 0551 works also but it appears that the actual first week is the creme of the crop.
Highest OC steppings are higher last 4 2nd line.
RPMW low
SPMW moderate
TPMW medium
UPMW normal
VPMW high
XPMW Highest
Batch # in the 0030 and lower would be better yields but not necessarily in the sense of HIGH OC, the difference can also be a stronger mem controller or higher over all HTT.
Once again this is all in theory and FYI has been compiled with over 1000 cpu results from users here.
Now with all this compiled I went on a stepping hunt. I ended up with this stepping based on the equipment i have in my possesion.
E4 revision X2 3800
LDBHE 0601 XPMW
batch # 60017
T chip
and the results are here.....This isn't maxxed yet either as it was more of a test run for my DI tube and manta versus anything else.
http://valid.x86-secret.com/show_oc.php?id=79315
What makes you so sure the meaning of T vs. N is silly? And FYI I don't go around "blabbering" that I know info and can't tell. I was asked directly and I responeded accordingly. Why should anyone be pissed? I guess you should be mad at the world since EVERYONE probably has some info which you don't :rolleyes: . When you get trusted with info that's NDA, maybe that's when you'll know what I mean (or maybe not).Quote:
Originally Posted by afireinside
Where did you see an 0501 LDBHE????? :eek:Quote:
Originally Posted by chew*
About 3 months ago when i stated searching. I can't remember the chip. I know the poster claimed no coldbug.Quote:
Originally Posted by s7e9h3n
If you could, please try to get the cpu info for me :toast:Quote:
Originally Posted by chew*
If I did know something under NDA I wouldn't publicly anounce that I knew it. If someone personally asked me I'd tell them I'm not allowed to talk about it. Simple as that. I'm done.Quote:
Originally Posted by s7e9h3n
I guess when I said 'No Comment', that must have meant "I KNOW ABOUT THIS SECRET AND I'M NOT TELLING" :rolleyes:Quote:
Originally Posted by afireinside
YES!!!!! and that's why i like you :toast: :DQuote:
Originally Posted by dogsx2
either don't say a word or say what it is..........hinting just doesn't do it for me either
@s7e9h3n,
How about passing that information to someone that is not under NDA? :D I don't think that only one person at AMD knows the meaning of "T" and "N" placed on the CPU package (substrate) - I would even think it has to be some sort of common knowledge among their engineers - so If "your friend" knows, then ask him to tell you that and then you relay it to someone that has nothing to do with AMD or any NDA. So in the end, neither you nor anyone else close to you can be linked to "leaking" :rolleyes: this information... :)
Quote:
Originally Posted by chew*
Thank you :)
Interesting though. We already knew that low batch numbers were best and that "BH" didn't bug. First week also makes sense since my 3000+ DTR 0501 chip was an absolute monster doing 2.95ghz on ice water.
My 170 is an "N" XPMW though and it can't even do 3ghz on ice water. It also needs 1.75v for high clocks. I've had a 3700+ "N" chip that required 1.8v to reach it's max clocks.
This seems like a good guide line but I still feel it's a lot of luck in the end.
Just wow at that overclock. 350 HTT is rock.Quote:
Originally Posted by chew*
I got 356x9 but not 360x8 on SS, so that's a better chip I think.
Sorry to say, but that's just an excuse to be lazy. Some info is out there, you just have to gather it and do the research yourself. When I decoded the stepping on the A64, no help was given to me by any "insiders". It was basically a tedious process which involved anylizing the info on hundreds upon hundreds of cpu's. Obviously, Chew looks to be doing the same for his own purposes. You're entitled to your own opinion, but please don't expect information to be just handed to you just because you requested it.Quote:
Originally Posted by dinos22
As i said this is not LAW this is just compiled results from over 1000 cpu's.
Also anyone ever remember the story about the hen that plowed the field,planted seeds,watered the field , harvested the corn, husked the corn, ground the corn, made flour then baked the bread and no one helped her do any work but when the bread was done everyone wanted some........
This may be why some people don't like to share info.
Cmon michal, you can't be serious, can you? It's kinda like if you're not caught stealing, then it's OK? :p:Quote:
Originally Posted by bachus_anonym
yeap you got me i'm just lazy :rolleyes:
chew* that`s A LOT of work you done combining every possible info and just telling it to all ppl here in short clear statements. :up:
Appreciate it man :toast:
thanks for the results chew*, good to see members supporting my own opinions about others too.
If you got info share it, or if you cant, dont say a word about it, it just entices people to spam a thread, pestering for that info.
Thank you for re-inforcing the reason why I didn't want to post here anymore. I'll leave you all to all your theories and such. This is where my XS stint comes to an end.......Quote:
Originally Posted by shuRe
oh really? so if you can afford to spend your time to do such things you assume everyone else can too?Quote:
Originally Posted by s7e9h3n
:up: for chew*.
I also want to thank all those that helped me in my 3 month long analysis. O wait there is no one to thank but myself............. actually I can thank stephen and FCG for there work on the cold bug.
Just for the record I DON'T have the TIME. I MADE the TIME.
right, tell me about that, when someone steal the theory of AMD stepping and no cold bug or n&T stuff and sounds he is the first one to figure out that? :mad:
Give me a break.:stick:
T or N..that 's what I told initially, and oh ya, there are lot more I bet you don't know. No matter Steve call me and said
He is done for XS.
And I will see who gets AM2 first.
Actually i didn't steal it OPB. In fact i helped steve on it and compiled results in a thread for him to prove his theory........Alot of the other stuff i deduced on my own. And once again this is just theory I could well be wrong as there are alot of factors that i can't calculate in such as the user behind the PC. We all know that Overclocker A might not be able to run a CPU as high as Overclocker B.
If I take credit on anything it's for the fact that i compiled this for 3 months and thats all.
oh ya? at least you should have my permission to post the pic of "my" CPU.
http://www.xtremesystems.org/forums/...11&postcount=8
permission :lol: you guys have a serious problem
YOU ARE POSTING ON A PUBLIC FORUM....................
Well thats not your cpu....Didn't you kill it........ That is your cpu's little brother from what I was told and a cpu that i am attempting to aquire...............through stephen and he sent me the pic......... I don't think steven is mad at me OPB.........Maybe he is if so have him PM me......Quote:
Originally Posted by Onepagebook
i think some people just get off having this info and not letting others know.
heck, if i get any kind of info i feel could be benificial to others i try to let them know. i dont hide it away for myself so i can think at night how cool i am with this secret info.
and not everyone here is rich, not everyone here has access to high end and ES chips. so not everyone can do this research themselves.
For you guys that are coming down on Steve for not sharing his info, that he might or might not be able to share. I'm pretty sure that the reasons that he posts these things is to let us know that even though he can't come out and tell us what he knows, that we're on the right track, or ont he wrong track, and is trying to urge to keep on looking for info, because parts of what you're looking for is out there, you just have to find those parts and put it together for yourself.
maaaaaaaaaan how about OPB getting the :banana::banana::banana::banana:s with Andre or someone because they didn't share info :rolleyes: looks a bit two faced from where i'm sitting
Everyone here has had acess to this info for a long time.........You don't need to be rich you don't even need a cpu. You need to get off your ass and READ............
And not for nothing but stephen Does not have to offer any info whatsoever being in the position he is in........In fact I think he's offered to much as without some little tidbit or hint he has said here and there I would'nt have been able to come up with all this.
As for you attacking me OPB i have no clue why..........And dino don't egg the guy on please......
Quote:
Originally Posted by s7e9h3n
what is being stolen? its information. you already admitted to telling info to people you can trust with it, pretty much. so you will tell people as long as you dont get found out?
and yeah, this is public forum. you post something its up for other people to use.
its not like he was making profit from the pic. he was helping the community.
about T & N chips chew*?
or about good steppings
i think that T&N theory you put forward is still inconclusive as it doesn't make sense with the experience i've had..........i know it's minor experience but it was quite the opposite for me in terms of vcore you mentioned
i call it as i see it.............either that or i don't understand the history/difference between my example and what is being said hereQuote:
Originally Posted by chew*
Quote:
Originally Posted by dinos22
You need cold to see that theory in action bottom line.
Yes but you just lower yourself by adding fuel to the fire..........
i will do that today or next few days.....just glued the IHS back on my Opteron 146 which is 12 Hour prime stable with 1.63v at 3.1GHz on AIR cooling and it is a N chipQuote:
Originally Posted by chew*
I think you'll find the clocks degrade over a period of time.
fair enough.......you know me though i don't hang onto my stuff for too long :D
Quote:
Originally Posted by chew*
Quote:
Originally Posted by chew*
assuming you refer to cold as in sub zero cooling.:rolleyes:
and how many users here with cold aka phase DI whatever post info........which you can READ
QFT, I have public exams in a few weeks, I don't have time to research 1000 cpu's myself so I posted a thread asking.Quote:
Originally Posted by high5
Again, QFTQuote:
Originally Posted by WeStSiDePLaYa
Guys, I'm so glad.
I've ACTUALLY found the meaning of life. It's AWESOME being me! :D
It's so simple, but God's NDA means I can't tell you! Sorry :(
Good for you.......If I break NDA with ATI I'm sure you'll send me review samples to test from now on..............Nuff said
when did this thread have anything to do with graphics cards :confused:Quote:
Originally Posted by chew*
Haha, chew, you pretty proved my point.
What's with the name dropping :(
Heh i wasn't referring to graphics cards........they do make other products....And it's not about a product in general it's about violating NDA.
Chew*, no disrespect, but I have two procs in my posession that thoroughly go against what you're saying. They both like volts on water-or-worse-cooling and have not, in the least, degraded clocks (one running at 1.65V for over 9 months). They are both "N"s.
Also, I don't think that 1000s of CPUs have been tested thoroughly enough to make a judgement on their voltage response.
I'm pretty sure chew* is trying to show you guys how immature, and at times ridiculous, you're being about respecting NDA's.
Chew* asked Steve is he knew anything about it. Steve couldn't say anything, so he posted, "no comment ;) ". He probably knew that chew*, being the smart one that he is, would see the wink, and take it to mean something along the lines of, "Yes, young grasshopper, you are on the right track."
At which point a few people started pestering Steve "tell me, tell me. I have an idea that you're trying to push us to do a bit of work, but that's just too much for me. tell me."
Honestly, I'm surprised at people's responses to S7e9h3n. He's been one of the most helpful and knowledgable people around on AMDs (not to say he isn't knowledgable on many other subjects, I just don't know). And then, when he gives an answer that could be taken a few different ways, people jump on him?
I've seen better.
And on water they are no where near maxxed out............ But like I said and i keep saying this is not LAW this is still theory and will be till AMD tells you differently.
This is the sole reason why I didn't wan't to post the results. Because then I would have a bunch of people saying :bs: your wrong my 2 cpu's say differently.
Btw when I say a 1000 cpu's its actually closer to 3000 But i just threw that # in there to give you an idea of how large and how many results i compared.
I would think there are exceptions to the rule. It can't be said for every sigle chip, but for the majority rather?
DragonOrta, personally I'm not too fussed about the NDA issue.
Chew posting how he had the Godlike info about AMDs and how we were all benile idiots, not worthy of his awesome secret.. repeatedly, was probably what annoyed people.
I was just trying to present it in a less serious manner :P
Anyway he did in the end so :toast:
Well, Chew* also said, if you want me help you, PM me and I'll be more than happy to point you in the right direction, give you stepping that will most help you with what you're doing.
DING DING DING we have a winner...........Quote:
Originally Posted by markr