New Thoroughbred CPU's. Your thoughts...

[Darksword]

Well, from the early reviews it looks like the T-bred's aren't going to be the killer overclockers we were all hoping for. To be fair though, the chips are still new and yields are sure to improve. I would like to see 2.5Ghz become a common overclocking mark with these chips, provided that they don't run too hot at that speed.

[DexterHolland91]

DK - Tom's hardware don't really know there @$$ from a hole in the ground, and they are Intel biased. A better indicator would be the one we get from www.amdmb.com , where the guy actually knows what he's doing. ;) I would think for an extreme overclock people like OPP could hit 2.5ghz but that would be the top-top end? More pratically 2.1-2.3ghz for the 2200+? Who knows... should get interesting. :D

[NorthernYankee]

I would have to agree with Dexter here I don't frequent Tom's site anymore as he has shown over and over again his ineptness with things.

--NY

[lechumbl]

Hi DarkSword,

Have to agree with Dexter and NY.

Tom's site has never given a fair anything, as far as AMD is concerned.
The reports are very biased toward Intel.
Certainly check out the AMDMB.COM site.
The range of 2.1 to 2.3 is very realistic for the normal O/C person.
Those that have "state of the art" systems should be able to hit 2.5, but not most folks

Take care......

[Player0]

Agree with all of the above. I didnt read toms but I read hardocp's review.

I find I ALWAYS accomplish much higher overclocks than most review sites. XP1800 places only got to 1700-1800mhz...i got it over 2000. KX7 some places only got it to 170-180mhz range...I hit 205.

So when HardOCP claims 2000mhz out of the XP2200, well i think 2200mhz-2300mhz :)

I think these CPUs will definately clock just as well as the Palamino counterparts. Probably in the 300-400mhz overclock range still.

If motherboards support them, and depending on how hard it is to unlock them.

[Farabomb]

I know I'll be getting one :D

Just have to sell one of my old boxes...

[BigBen]

I think AMD are using the T'Bred as practice for the Hammer series of chips...getting the .13 sorted in advance

Regards

[mdzcpa]

Originally posted by BigBen
I think AMD are using the T'Bred as practice for the Hammer series of chips...getting the .13 sorted in advance

Regards

Absolutely correct.

There is no doubt that AMD had pulled all R&D budget to the Hammer side of things...but who can blame them. AMD's long term future relies on real penetration of the Corporate sector...and that means Hammer. So we'll see the desktop stuff go soft for a while.

But, the TBred will do ok, and likely limp along a little ways behind the P4 for a while. Again, this isn't much different than when the P3 began to trail the TBird, and was eventually crushed by the Palomino. The K7 is in for the same fate eventually. Intel and AMD are just playing leap frog with their respective technologies. Hammer will makes it's jump by year's end or so.

Right now the pendulum has swung in Intels favor as the K7 has grown a bit long in the tooth, while the P4 continues to ramp up.

So, quite frankly, AMD fans need to get used to the idea of being a bit behind the P4 for a while now in sheer performance. The bang for the buck contest still goes to AMD hands down, and will be even further in AMD's favor with the move to the .13 micron process with lower costs involved (and hence even better prices). But for pure performance...it may be the P4 on top for quite some time now.

But I'll stick with AMD and get some super cheap break neck performance until the Hammer arrives. And when it does, everything changes. AMD knows it....and Intel knows it. With MS ready to go with a 64 bit (32 bit leagacy) OS already, it will not be long before 64 bit apps start popping up all over...it has to...it is the future. And the awesome 32 bit legacy support of the Hammer will circumvent the only real downside to going with a 64 bit processor (not to mention an affordable 64 bit CPU too).

Anyway, I plan on playin' around with the TBred, and I'll likely try the Barton too when (and if) it sees the light of day. And in the meantime, as AMD works up it's yields on the .13 process, I would hazard to guess that we will see more substantial overclocks in the future out of the TBred too.

My ramble ends...
/sounds of Mike's feet decending soap box/

[lorddrakul]

I dunno guys,

I have seen a few things that do not look good for the overclocking of the new T-Bred core.

Anandtech I think is a very respectible and AMD receptive site. Below is their experience:

Anandtech XP2200+ OC Results (http://www.anandtech.com/cpu/showdoc.html?i=1635&p=14)

While Hexus.co.uk do not go indepth, they come back with a nocer spion but much the same numbers:

Hexus.co.uk XP2200+ OC results (http://www.hexus.co.uk/review.php?review=376&page=9)

It seems that the ceiling for OC on these XP chips has remained much the same over the transition from the 0.18 to 0.13micron process.

I agree that it appears as if AMD are simply testing the processes for the Barton, but at the same time, the ground that was lost by the XPs to the P4 has not been regained. -There are widespread reports of the Northwood based P4 1.6A hitting 2.6Ghz with little problem at 1.65Vcore.

I think that AMD need to accelerate the process with the Barton lest they loose the Kudos gained with the entire XP line to the burgeoning P4 following.

[4.6POWER]

that anandtech review had the 2100 overclock farther then the 2200... I think that show it is inaccurate. The core of the 2200 had to be either better than the 2100 or the same as the 2100. On those grounds... the chips should overclock evenly. Meanwhile anand only got ~20mhz more out of the 2200? That isn't an overclock at all... if anything that chip won't even operate 110% at it's rated speed, something was wierd there. Don't forget... they also use the factory AMD heatsinks :o

[Player0]

Anandtech has never had a decent overclock ever ;)

Let me at one of those puppies...ill give it a proper showing :)

[DexterHolland91]

Go get em P0. :D

[4.6POWER]

Player0's cube is like a fate worse than death for those little chips.... it's a scientifically designed chip torture prison :D complete w/cryogenic freezing & high voltage electricity :p poor little guys :drinky:

[lorddrakul]

Update:

Anyone seen This (http://www.amdmb.com/article-display.php?ArticleID=188&PageID=1)

LD

[NorthernYankee]

Hmmmm very, very interesting.

--NY

[dicki]

yeah we allways see this when a new core comes out... give it a couple of weeks and we'll se a better spread of results and be able to draw some real conclusions

dicki

[lorddrakul]

I think the guys have answered their critics though. When this emerged there were those who doubted what was being reported. I think that they have answered all questions in a reasonable manner whilew verifying what they produced.

I think I will be hanging onto my Palamino for a while yet. ;)

LD

[Old sptw]

This guy from Overclockers Australia (http://www.overclockers.com.au/techstuff/a_amd_tbred/) got almost 450 Mhz with a 1900+ and Vapochill case!!!!

[mdzcpa]

Here's a very interesting blurb from Van's Hardware which sheds more light on the TBred situation:

AMD Launches Thoroughbred

AMD's Athlon XP 2200+ debuts today. Codenamed "Thoroughbred," this chip brings AMD's 0.13-micron process to the desktop for the first time.

Over the last few weeks, I have been overseeing verification of a critical project and have had no spare time for our review. I apologize and will finish the article at first opportunity.

However, I have noticed a very large number of misconceptions in reviews today regarding the "low potential headroom" of AMD's 0.13-micron Athlon XP. It is clear that AMD is currently using a process skew emphasizing higher yields and lower power dissipation, apparently to produce large numbers of mobile chips. This is logical since mobile parts command higher ASPs. Such a mobile-friendly process skew would result in a slower transistor and reduced maximum clock-speeds.

Faster transistors have shorter gate lengths and are therefore leakier (consuming power when doing nothing at all). That AMD is using a slower transistor is demonstrated by typical current in Stop Grant, which is less than one-third of that leaked by the Intel Pentium 4 Northwood core in this sleep mode (the P4-Northwood wastes an amazing 18A when the chip is doing nothing!). This makes the Thoroughbred a much better mobile part, but higher clock-speeds will likely have to wait until AMD remixes its recipe for faster transistors.

I had heard this before as a rumor, but Van seems to beleive there is some substance too it. I hope he is right:)