.13 micro AMD processors

[scamtrOn]

is there something out there that will show what kind of speed the .13 micron cpu will hit?

i want to know what the max speed will be. the real speed of the xp or any other.

<A HREF="http://www.geocities.com/scamtron2000/Lochel.html" target="_new"> <font color=red>go to my site</font color=red> </A>

 

[Crashman]

I think it's in their roadmap as to what speeds will be released, something like the XP2400+ or whatnot. Nobody knows what the ultimate speed for the core will be until it's released.

What's the frequency, Kenneth?

 

[OldBear]

Don't know if its true or not.

<A HREF="http://www.athlonxp.com/news_xp2100.html" target="_new">XP2100+</A>

<font color=blue>You get what you pay for.... All advice here is free.</font color=blue>

 

[peteb]

.13 micron AthlonXP running at 1.75V - I doubt it somehow.

-* <font color=red> !! S O L D !! </font color=red> *-
To the gentleman in the pink Tutu

 

[girish]

heard it would run at 1.4 or 1.45V, good enough to generate lesser heat so that there will be lesser number of fried Tbreds! I think they will start with XP2200, the AXP still has a couple of speeds ratings left.

<font color=red>Nothing is fool-proof. Fools are Ingenious!</font color=red>

 

[eden]

The max should be as scalable as the P4, not by speed but by % like Matisaro said. It means it would top out at around 3 GHZ if possible, since it switched at 1.66GHZ and 50% is often attained by micron change. Look at Athlon 0.25 to 0.18 and P3 to Coppermine, scale 50% average!

--
For the first time, Hookers are hooked on Phonics!!

 

[IIB]

50% is only from 1.73 to 2.46...
you mean 100% - I say unlikly
you can also look at P3 from 0.25 at ~700 (I think) to 1Ghz...
I dont think the T-bred will be as scalable on 0.13 as the NW (in %)
I think the northwood is designed to scale better at 0.13 and 0.9...

This post is best viewed with common sense enabled

 

[Matisaro]

50% is only from 1.73 to 2.46...
you mean 100% - I say unlikly
you can also look at P3 from 0.25 at ~700 (I think) to 1Ghz...
I dont think the T-bred will be as scalable on 0.13 as the NW (in %)
I think the northwood is designed to scale better at 0.13 and 0.9...

The core design has a minimal effect on the % of clockspeed gains, it is purely a matter of process size.

I dont think the tbred will his 3ghz though, the barton should, but the tbred most likely will top out around 2.6ghz.

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!

 

[FatBurger]

so that there will be lesser number of fried Tbreds!

It will generate less heat but be more sensitive to heat. I doubt that the number of fried CPUs will change.

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>

 

[texas_techie]

To understand scalability, you have to understand the inherent barriers at the .13 micron level.

1. Quantum tunneling - the electrons at the .13 level actually quantum tunnel through the oxide layer and cause signal degredation.

2. Gate leakage - even though the process is .13, the actual gates are less than .13. The ends of the gates being that close together causes charges from one side of the gate to "leak" to the other side. Obviously, this can cause serious problems.

3. Inadequate doping - the process of puting charged particles near a gate = doping. AT the .13 level (and even .18) it is VERY difficult to get enough charged particles, get them in the right place, and get them to behave as they should. Intel and AMD both are having issues with substrates that allow for adequate doping.

NOW- It depends HEAVILY on how AMD solves the above 3 issues on how far the t-bred and Hammer will scale. Obviously they have gone to SOI. But thats only part of the solution. I would LOVE to know what dopants and substrates they are using - then again, so would Intel -
Id say hold the speculation till the chips come out. There is no telling what they will be able to do.

Benchmarks are like sex, everybody loves doing it, everybody thinks they are good at it.

 

[texas_techie]

To understand scalability, you have to understand the inherent barriers at the .13 micron level.

1. Quantum tunneling - the electrons at the .13 level actually quantum tunnel through the oxide layer and cause signal degredation.

2. Gate leakage - even though the process is .13, the actual gates are less than .13. The ends of the gates being that close together causes charges from one side of the gate to "leak" to the other side. Obviously, this can cause serious problems.

3. Inadequate doping - the process of puting charged particles near a gate = doping. AT the .13 level (and even .18) it is VERY difficult to get enough charged particles, get them in the right place, and get them to behave as they should. Intel and AMD both are having issues with substrates that allow for adequate doping.

NOW- It depends HEAVILY on how AMD solves the above 3 issues on how far the t-bred and Hammer will scale. Obviously they have gone to SOI. But thats only part of the solution. I would LOVE to know what dopants and substrates they are using - then again, so would Intel -
Id say hold the speculation till the chips come out. There is no telling what they will be able to do.

Benchmarks are like sex, everybody loves doing it, everybody thinks they are good at it.

 

[girish]

well, Thoroughbreds are not SOI, next generation Bartons will be.

<font color=red>Nothing is fool-proof. Fools are Ingenious!</font color=red>