Thoroughbred out on the 20th of March

[hallen]

AMD ramps up the war
By Mike Magee, 05/03/2002 15:01:31 BST


UPDATE
WE NOW HAVE confirmation from sources close to AMD's plans that the first Thoroughbred will be released on the 20th of March.
And AMD will release a speed step, the XP 2100+ for the first day of CeBIT on the 13th of March.

The two are not to be confused. The XP 2100+ is the last in the line of processors in the Palomino family - AMD has been able to clock it one ratch higher.

There are three changes in the Thoroughbred from Palomino, as we mentioned in the AMD roadmaps we published earlier this year.

The CPU ID pin has changed, of course it is now .13 micron processing, and the Vcore is changed to 1.65 volts.

Nevertheless, there are expected to be no difficulties with compatibility. According to our information, Thoroughbred chips will be compatible with KT266x chipsets and KT333 chipsets.

The firm will start to release different speed ratings of the Thoroughbred chips at carefully timed intervals in the months to come.

No doubt a few of those who will review the processor are already testing the chip to see how fast the Thoroughbred gallops. µ

UPDATE II Some say that the 20th of March is not the Thoroughbred date.. We'll see what we come up with further on this front, tomorrow.

http://www.theinquirer.net/05030214.htm

 

[]-[itman]

Is it me or is "The Inquirer" hard to take as a reliable source? I've had many instances in the past where their "reports" are completely off. Just like in the last line of this report. That's basically saying this is what we think it is but incase we're wrong we'll put this in the report so no one can blame us if we're wrong. I'm sorry I just can't trust anything "The Inquirer" says anymore

 

[Bruno]

so who we gonna trust, till they're on the street.

and what board? I'm tired of frying drives.

I can't wait for q 2g that will do 2G, air cooled.

God help us all if they fail like the XP.

Pray brotha's.

Bruno

 

[bobt17]

so who we gonna trust, till they're on the street.

and what board? I'm tired of frying drives.

I can't wait for q 2g that will do 2G, air cooled.

God help us all if they fail like the XP.

Pray brotha's.

Bruno

what do u mean "fail like the XP"?

 

[]-[itman]

so who we gonna trust, till they're on the street.

and what board? I'm tired of frying drives.

I can't wait for q 2g that will do 2G, air cooled.

God help us all if they fail like the XP.

Pray brotha's.

Bruno

I have to agree, I don't think the XP's failed at all! To answer your question, the t-breds will be able to fit in the sockets you have now. They'll probably be used in KT333 boards but i'm not sure. It's not until the hammers that we'll need a new board(not to say the newer boards won't be faster). If I did my calculations correct, then the first t-breds(2200+) will be clocked at 1.76ghz. Now, the new steppings for the pally cores have gotten some pretty good results. So, I'd say, with the new .13 core you will see 2.0+ghz very soon if not with the 2200+. IMHO I'd say it looks very good for the AMD fans!

 

[Bruno]

the generation Xpees, didn't do much more than an air cooled 1.4g bird. Very few flew like the birds, unless you vapoerchilled em. and you know it.

the xpees sucked.

bruno

 

[bobt17]

XP's oc just as good if not better then tbirds. for a tbird to perform equal to an xp1600+ it would have to run at 1.6ghz and a 1700+ the tbird would have to run at 1.7ghz thats how amd made their model number system


PS: please dont show me some stupid sandra score either because sandra cannot show the performance difference between an xp and a tbird unless u look at the PR that sandra gives

PPS: the tbreds will probly perform about the same as the curent xp's based on the palamino cores because as far as i can tell they are just a die shrink and lower vcore so they will probly oc very good probly oc better then palaminos

 

[Sonny]

Real Proof not Opinions

.....for a tbird to perform equal to an xp1600+ it would have to run at 1.6ghz and a 1700+ the tbird would have to run at 1.7ghz thats how amd made their model number system.....

Any proof that what you are saying is true?

 

[bobt17]

here

 

[Sonny]

The CPU ID pin has changed, of course it is now .13 micron processing, and the Vcore is changed to 1.65 volts.

DON'T EXPECT MUCH MORE OUT OF THE THOROUGHBREDS

 

[bobt17]

the best things i see about the tbred is since it uses less vcore then the palaminos that gives us overclockers more room to ajust the vcore higher so we should be able to get higher oc's and since it will be true .13 micron we should have at least a bit less heat to deal with witch would also help in overclocking it higher so to me at least it looks like it will be a better overclocker then what we have now

 

[Sonny]

here

That PDF has great info.

.....for a tbird to perform equal to an xp1600+ it would have to run at 1.6ghz and a 1700+ the tbird would have to run at 1.7ghz thats how amd made their model number system.....

You just proved yourself wrong. TB 1.4GHz vs. XP1600 only a 7% difference at best so the TB DOES NOT have to run at 1.6GHz to equal an XP1600 & so on.

Who said it was true .13 fabrication?

 

[bobt17]

"Desktop processors based on the "Palomino" core will be marketed as the AMD Athlon™ XP processor. As a way of communicating the performance improvements of the new AMD Athlon™ XP processor relative to the performance of the currently available AMD Athlon™ processor, AMD has developed a model numbering convention"


thats what i read and have been referenced to b4 by a few people the way i read that is that they used the normal athlon as the base for the model number system thus 1600+ would equal 1.6ghz tbird

 

[Sonny]

Their benchmarks don't lie. It really depends on what, not who since both information comes from AMD, you want to believe Benchmarks or Advertising

 

[bobt17]

another reason i belive that is here look at number 13 down the list and all the tbirds that it was able to compleat a seti@home work unit faster then..... that is an athlon xp 1500+ that i used to run and it beat plenty of tbirds that were clocked quite a bit higher then it even at default and with seti normaly memory bandwith is the way to go for faster scores but look at number 14 compared to number 13 thats a tbird [email protected] and it got beat by the default 1500+...... and thats not the only tbird it beat thats just the closest that came to it

EDIT: btw they used the same mobo also

 

[bobt17]

ok i cant really explain that...... if u really want to put this to rest will u run your tbird at 1.4 or as close as u can get it and run the seti benchmark? it is explained on the page b4 it how to run it and then we would both have an answere

EDIT:...... where'd it go lol i could have sworn i seen a post above this by sonny

 

[Sonny]

Took it out. Different FSB on both machines so it could not be used as an example since you state that RAM bandwidth is an issue. Find somebody that has DDR like your rig so you can make a true comparison.

 

[bobt17]

lol good i thought i was going crazy there for a sec lol all i know is that #13 is the one i did and it didnt have any interuptions so it should be acurate and the one below it is a tbird with the same mobo higher clock and fsb and mine still beat it hmmmm i wonder how much this subject has been discussed in the debates section of this forum lol seems like both sides have stuff to work with

 

[Sonny]

Here it is the 1MHz difference in FSB. Not a reliable benchmark. Believe what you want. I just like to stick to reality.

 

[bobt17]

ok heres an intresting thing to note tho run sisoft sandra and tell me what pr it gives your cpu at 1.4ghz and i will default mine witch would give it 1.4ghz and i will tell u the pr sandra gives me i bet it will be a couple hundred higher then your bird at the same ghz