Best Barton 2500 stepping

[Duneadan]

What is the best Barton 2500 stepping for overclocking? I'm thinking of getting a new one. The 2500 I have now just can't make it stable over 2174 Mhz.

[c627627]

It may not be the CPU. It would be smart to list your entire specs so someone can say a thing or two about the brand of your power supply, etc...

[Akira283-IGN]

There are no "good" overclockers or "bad" ones. The newer the CPU, the more overclocking potential it has, so find a CPU with the most recent "birthdate" and you'll be set.

[Mr B]

Originally posted by Akira283-IGN
There are no "good" overclockers or "bad" ones. The newer the CPU, the more overclocking potential it has, so find a CPU with the most recent "birthdate" and you'll be set.

I have to disagree with this one. I've seen lots of chips from specific "batchs" mentioned around here, that *usually* overclock well, and chips of the same PR rating and type (IE 1700+ T-Bred "B") with a newer date that have bad "batchs".

Each chip is unique, and all of them will overclock differently. But it's definitely known that some steppings are better than others. Certain dates within certain steppings even...

"Newer" does not always equate to "better", but I will agree to the point that as time goes by, the process is usually refined, and the chips do yield higher numbers. Also, as that happens, sometimes those 1700+'s get labeled as 1800+, 1900+, etc....

There's too many variables to just say newer = better.

B.

[Akira283-IGN]

Sorry, but it's a fact that newer CPUs have more overclocking potential than older ones. Overclocking "databases" are terribly unorganized, and they overlook fundamental principles that make inferences about overclocking-quality useless. The differences between each CPU from every batch are not known to anyone, and never will be. Very few CPUs are actually overclocked, and do you have any clue how many factors actually influence overclocking? Voltage amount and quality, cooling, motherboard quality, user knowledge - these are just a few.

A handful of overclockers post their specific results in some online database or on a forum such as this one, and that's supposed to be evidence? There is no method for control since every result contains variables which are rarely accounted for. Everyone uses a unique cooling setup, motherboard, power supply and RAM to achieve their results. Even their own knowledge and ability as an overclocker factors into the result. If someone wanted to compare the overclocking ability of two or more CPUs, they would need to eliminate these variables by testing all the CPUs in the same exact setup.

[L337 M33P]

To me that above post looks almost exactly like teh one in this thread tbh: http://forum.oc-forums.com/showthread.php?s=&threadid=243381

[Akira283-IGN]

You're right, I copied an earlier post I made and placed it here. It's the exact same topic though.

[c627627]

To: Mr. B

From: c627627

Re: Banning threat


I was threatened with a ban (triple exclamation point), I stopped posting and was on the verge of leaving the forums altogether because a moderator in your absence deemed my post to be too condescending to Akira283-IGN regarding this very issue:



This is where your sense of rationality will display itself or not. Facts only facts, no egos. None of this personal stuff. We compare facts here.

Here we go, straight up from AMD FAQ:


Originally posted by dippy_skoodlez
***New stepping*** Just bought a 1700+ from newegg and its an axda AIWGA 0324 VPMW

DLT3c BTW... Can't wait to see how it OC's!


...and right after that

Originally posted by c627627
Poor dippy_skoodlez got himself a DLT3C T-Bred A:

His code was AIWG--->A<---

http://forum.oc-forums.com/vb/showthread.php?s=&threadid=222237

Do not buy 1700+ from newegg.
They only guarantee 2100+ as T-Bred B's.


Fact: It was a Revision A without the extra layer of copper interconnects to reduce interference. This is why A's scale to 2200+ and B's scale to 2800+ as you can see here
http://forum.oc-forums.com/vb/showthread.php?s=&threadid=240564

The fact that this was a brand new 0324 directly contradicts your statements:

quote:

Originally posted by Akira283-IGN
All you need to do to get the most overclocking potential is to simply buy a new CPU. If you can find a place to get a guaranteed week, great, that way you can simply choose the newest CPU possible. Just don't be fooled by myths that older weeks are better.


The entire content of the quote was disputed using facts without any reference to your person.

You shall now
1.) Ignore this post
or
2.) Talk about other things you posted instead of addressing the quoted content
or
3.) Retract the quoted statements

Your game.


It ain't right to ban me after 1500+ posts of helping people. What am I supposed to do, not reply now?

I tried my very best not to personally attack Akira283-IGN and although my post was condescending it certainly is no reason to close the previous thread on this subject. Closing threads, along with banning should be treated like holiest of holies by moderators.

[felinusz]

*yikes* This will go nowhere good.

c627627 - the treatment you received (the warning) was not warranted, and should be apologized for, in my personal opinion - but neither should you have added that last bit in your original post:

You shall now
1.) Ignore this post
or
2.) Talk about other things you posted instead of addressing the quoted content
or
3.) Retract the quoted statements

Your game.

Which certainly seems to be unrelated to the discussion, and very, very, Ad Hominim. How would you feel if, say, I were to use similar wording at the end of a "counter-point" directed at you about something you believed in and were posting with the intention of helping others/sharing your opinion? No offense intended.

I was wondering why I hadn't seen you for awhile - I guess this explains it - it is good to see you back.

.

.

We're onto this topic again. With a Barton 2500+ you are most likely to receive best results with the newest processor you can find.

With a T-Bred you are most likely to receive best results with the newest T-Bred "B" (has a stepping ending with B - I.E. JIUHB) that you can find. Other than that, and stock Vcore indicators in steppings, steppings are pretty much meaningless when it comes to overclocking.

With any given type of processor (the DLT3C T-Bred "A" being a type, the DLT3C T-Bred "B" being a type, the Barton being a type), newer usually means better for overclocking.

[c627627]

There's too many variables to just say newer = better.

Prime example: a batch of CPUs with a newer date may have weak spots not discovered during manufacturing stress tests. With increased voltage, those weak spots may lead to permanent failure sooner than earlier manufactured batch regardless of your equipment.

That's why we used to say don't overclock more then 1.8 Vcore and that's why hitechjb1 said statistically 30% increase in Vcore, that's well over 2 volts will give you "only" 59% reduction in expected CPU expected life span. 6 years intead of 10.

If it goes out sooner and heat was not the cause, those weak spots because of that particular batch, regardless of your equipment, caused early death.

And that's why "there's too many variables to just say newer = better."

[Mr B]

wow....what a can of worms I've opened here. :eek:

Ok...I'm not even going to re-enter the subject of old/new/stepping/date/color/smell/taste, and I almost wish I hadn't in the first place. I'll agree to disagree, and go my own way. I know what I know, and that's all I need to know...

c627627, I cannot comment on the subject you've brought up, as I'm not familiar with it. I can look into it, if you'd be so kind as to PM me with more specifics that I would appreciate not posted in the open Forums. Let's keep it in PM's, please.

I'm going to go play w/ my 2100+...the one that outclocks some newer ones..... ;) :cool:

[felinusz]

:) A very good point there.

Well - lets just settle this and say that each chip will perform differently - and you cannot know how well it will perform in your rig untill you have played with it yourself.

For all we know there are little faeries that decide which chips will do well
8-D

"there's too many variables to just say newer = better."

After a very brief amount of thought, I agree. But newer=better is a very *weak*, but still present "trend" that has made itself apparent in *some* cases, for *some* people, whom *some* of which are willing to use a higher Vcore than others, with *varied* cooling, with *some* combinations of hardware...... etc. I hope that came across as intended to both of you! ;) ;)

- felinusz

[c627627]

You're of course right. Generally, unless you know a particular stepping performs above average, it's best to gamble on a new Thoroughbred B or Barton chip.

I'm sorry about the way I worded things. If I could go back I wouldn't post in such a way.

Discussion in previous threads was along the lines: "just go to newegg and buy the newest chip." newegg at the time was shipping massive amounts of brand new T-Bred A's which perform 300 MHz below T-Bred B's on average. Only 2100+ was guaranteed.

Discussion in previous threads was also along the lines "every newer chip will be better than every older chip and if it isn't, you're doing something wrong."

This of course is not true. It is a fact that older batch can be better than a new batch of CPUs. It usually isn't but it can be.