Sign in with
Sign up | Sign in
Your question

NO THOUROUGHBRED UNTIL MAY 02

Last response: in CPUs
Share
March 10, 2002 9:53:48 AM

I am getting sick of people asking when AMD will release the Thouroughbred version of the Athlon XP.

<A HREF="http://www.theinquirer.net/09030201.htm" target="_new">http://www.theinquirer.net/09030201.htm&lt;/A>

From reading that artical Thouroughbred wont be here until May this year <b>the earliest</b>.

In my opinion AMD are slighlty behind schedule which means that thier .13 micron mass production is not up to full speed yet. (Yes I do know that AMD plan to be fully converted to .13micron at the end of 02). We have already seen Thouroughbred running way back in Q4 01 therefore am sure the processor itself is fine its just that the transition to .13micron it taking slightly longer than expected (I doubt this will affect the Hammer line).

<font color=purple>~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)</font color=purple>

More about : thouroughbred

March 10, 2002 10:02:26 AM

<font color=blue>In fact, on the 20th of March, many motherboard manufacturers will receive fully working <b>samples</b> of the Thoroughbred.

Sources a cigarette paper's width away from AMD confirm that's what's happening round about the 20th, and that <b>we won't see fully fledged machines running the Thoroughbred until about eight weeks afterwards.</b>
</font color=blue>

Believe what you want...

<font color=purple>~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)</font color=purple>
March 10, 2002 10:11:53 AM

too bad, i was really looking forward to it. well, must wait for a while longer!

actually, i still havent got a Palomino XP, I have a 1.1 gig tbird, just thought Tbreds are not far away, skipping Pal would be better!

<font color=red>Nothing is fool-proof. Fools are Ingenious!</font color=red>
Related resources
March 10, 2002 3:45:43 PM

crap! I have to wait longer to upgrade. :mad: 

<font color=red><pre>i so good i jealous of me</pre><p></font color=red>
March 10, 2002 6:03:11 PM

Quote:
we won't see fully fledged machines running the Thoroughbred until about eight weeks afterwards.

Just because "fully fledged machines" won't be available, that doesn't mean that OEM chips won't be out sooner. Usually, the chips are available in the channel at least 3-4 weeks prior to the "available" date.

I thought a thought, but the thought I thought wasn't the thought I thought I had thought.
March 10, 2002 8:01:16 PM

I doubt it very much...the chips are probably <i>in the channel</i> as we speak now. OEM's will only have limited supplies...I wouldn't hold your breath if I were you.

We won't see any action until the last week of April- early May time.

<font color=purple>~* K6-2 @ 333MHz *~
I don't need a 'Gigahertz' chip to surf the web just yet ;-)</font color=purple><P ID="edit"><FONT SIZE=-1><EM>Edited by mr_gobbledegook on 03/10/02 05:20 PM.</EM></FONT></P>
a b à CPUs
March 10, 2002 10:08:48 PM

FVCK!!! I can't wait that long, might just have to go with the NW to get the good overclockability :mad: 

What's the frequency, Kenneth?
March 10, 2002 10:15:53 PM

Now now, Crash, hold the bad thoughts and words, you never know if the Inquirer might be lying here. I mean AMD introduces Tbred in March, then 2 months later it is out? I somewhat do not beleive this. If AMD plays that way, they might be outrunned by Intel.
So hold a little more till we hear officially next week! CEBIT IS GETTING CLOSE!!!

--
For the first time, Hookers are hooked on Phonics!!
March 10, 2002 11:21:22 PM

I agree with you Eden. I can't see AMD allowing Intel to release a 2.4ghz P4 wthout firing back with at least an Athlon XP PR2200.

Let's see what they say during Cebit. I suspect we'll get more detailed info on Hammer by then and maybe even a few preliminary specs from actual silicon. Of course, it will also be nice to see a T-bred announcement with a firm release date.

If AMD has converted 50% Dresden to .13micron I don't see where it would take long to ramp up T-bred, especially if the conversion is going as expected like they say.

Mark=

When all else fails, throw your computer out the window!!!
March 10, 2002 11:32:12 PM

I'm excited to see if any Hammer news come up, especially performance and spec wise, because that would show Raystonn that we are right about AMD not screwing up this one.

--
For the first time, Hookers are hooked on Phonics!!
a b à CPUs
March 10, 2002 11:54:51 PM

OK, ok, I'll try not to mention the NW again. But I really need to get this new system together by mid April.

What's the frequency, Kenneth?
March 11, 2002 12:01:31 AM

why the deadline?

<font color=red><pre>i so good i jealous of me</pre><p></font color=red>
March 11, 2002 2:03:55 AM

They were some many delay on Nw.It logical that AMD get the same trouble that intel have.It been near 6 month that the tuatuline was release.

My prediction still true up to date.

cheap, cheap. Think cheap, and you'll always be cheap.AMD version of semi conducteur industrie
March 11, 2002 2:15:46 AM

Even though - bang for $ it is tough to argue that 1.6A woody. Figure the top tbred is going to be over double the cost of the 1.6A, probably around $350-$400.

-* <font color=red> !! S O L D !! </font color=red> *-
To the gentleman in the pink Tutu
March 11, 2002 2:19:17 AM

Hey Grizzley1, it's been a while. This is the FB from BZ/BZ2.

I'm waiting for the Hammer to come out anyways:

1: I have no money
2: I really want to see it crush Intel's P4 at the time.

:) 

if(GetSystemMetrics(SM_PROCESSOR) != AMD_PROCESSOR)
{
SendMessage(hwnd,WM_CLOSE,0,0);
}
March 11, 2002 3:37:55 AM

Eden, I kinda think that's irrelevant. What folks here think about Hammer 6+ months befoore release is irrelevant. It's nice to diuscuss the possibilities, but really, the proof is in the pudding, and there's no pudding until there's actual product on the shelves and reviewed by reliable sources.

Besides, whatever the info that may come from Cebit about Hammer, remember the P4 info months prior to the release of the original retail Willamette P4 and how different the actual shipping product was (estimated performance-wise).

Mark-

When all else fails, throw your computer out the window!!!
March 11, 2002 5:04:07 AM

edited: due to lack of interest...and possible litigation.

Benchmarks are like sex, everybody loves doing it, everybody thinks they are good at it.
March 11, 2002 9:22:01 AM

Quote:
Even though - bang for $ it is tough to argue that 1.6A woody. Figure the top tbred is going to be over double the cost of the 1.6A, probably around $350-$400.


Looks as if there will be a tbred 1800+, which should be cheaper than the 1.6a.

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!
March 11, 2002 10:26:04 AM

Quote:
Looks as if there will be a tbred 1800+, which should be cheaper than the 1.6a

but what we don't know is if it will overclock 50% on air. Maybe it will, maybe it won't. We know the 1.6A does, in most cases, with the stock cooler... As they say, time will tell. I hope it does, but the P4 is here now, the Tbred is still at least 2 months from general retail, probably more.

-* <font color=red> !! S O L D !! </font color=red> *-
To the gentleman in the pink Tutu
March 11, 2002 10:56:11 AM

Actually if the Hammer "isn't doing so hot" right now for you, I'd say that AMD has done a lot in the last 2 weeks since their intervention at IDF 2002. I mean they had a low clock speed running Hammer to show, no AGP and a reference board that caused a conspiracy with Intel's jingle. Now they've pretty much solved the AGP, and are running at near-full speed!
That's some nice work by AMD, and I have a feeling they are on track for now! We'll just see how it is at CeBIT.

--
For the first time, Hookers are hooked on Phonics!!
March 11, 2002 10:59:07 AM

I remember the P4, but then they did show us in the end the latest specs or "updated" ones later on didn't they?
Also, I just can't see how AMD could not pull it right.
Think about it, the Hammer is bigger, but it ain't a P4 with a 20 stage pipeline! Where can the IPC be lost! Also with a bigger core, and the Hyper Transport, the FPU and ALU were already fitting in the smaller Athlon die, so how can it not work here?
Simply put, there are too many reasons why it cannot do like what Intel did with the Willy. I agree we should wait and see, but personally, this wouldn't make sense because they got it all in front of them.

--
For the first time, Hookers are hooked on Phonics!!
a b à CPUs
March 11, 2002 11:36:22 AM

I'm getting this one PAID for!

What's the frequency, Kenneth?
March 11, 2002 11:57:03 AM

I don't disagree with your assessment Eden, BUT let's not go overboard. I personally would prefer to be a bit conservative than expect more than is delivered. So, while your theories may be right on target, since we don't have actual silicon to base the theories on, perhaps it's best to be a little conservative.

Mark-

When all else fails, throw your computer out the window!!!
March 11, 2002 1:59:14 PM

Where did you find out about a 1.8 Tbred? That could be a sweet spot in the new line.

"Put your desk in the corner." - Stephen King
March 11, 2002 2:16:01 PM

Quote:
but what we don't know is if it will overclock 50% on air. Maybe it will, maybe it won't. We know the 1.6A does, in most cases, with the stock cooler... As they say, time will tell. I hope it does, but the P4 is here now, the Tbred is still at least 2 months from general retail, probably more.


As I have said in many previous threads, theres nothing the Northwood has that the tbred wont also have in regards to oc potential.



"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!
March 11, 2002 2:30:57 PM

I don't mean to be rude, but you saying it and me seeing it are not the same thing. Hats off to AMD if they _do_ it though.

-* <font color=red> !! S O L D !! </font color=red> *-
To the gentleman in the pink Tutu
March 11, 2002 2:57:58 PM

Darn you! How did you manage to pull that off?

<font color=red><pre>i so good i jealous of me</pre><p></font color=red>
a b à CPUs
March 11, 2002 3:26:30 PM

I got in a lot of fights in High School, and now I get paid to avoid them.

What's the frequency, Kenneth?
March 11, 2002 3:38:20 PM

Come to the dark side...the P4 1.6A is calling you...


Sorry, I'm really tired today. :frown:

<font color=orange>Quarter</font color=orange> <font color=blue>Pounder</font color=blue> <font color=orange>Inside</font color=orange>
Don't step in the sarcasm!
March 11, 2002 3:43:41 PM

Yeah, what's so bad about a P4 1.6A? Get it already!

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
March 11, 2002 7:01:02 PM

Oh no, I won't go hyping too much though. It's not like I expect it to go beyond me, but I do expect the basic of what it was designed, to have a huge IPC and to start competing at 3400. That is all I am hoping for, and if AMD delivers more, it will back whatever we said in the past and it'd show they are not in just for the fun.

--
For the first time, Hookers are hooked on Phonics!!
March 12, 2002 6:15:51 AM

Quote:
I don't mean to be rude, but you saying it and me seeing it are not the same thing. Hats off to AMD if they _do_ it though.

No offence taken, and I understand the stance, however can you point out a logical physical reason why it shouldnt follow the same curve?

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!
March 12, 2002 6:26:43 AM

the core architecture is different.

The architecture of the P4 is originally scaled to go high, but the .18 silicon capped them. Moving to .13 allowed much further scaling.

The Tbred is a shrink of what is effectively now a pretty old Athlon core (with a few tweaks). I think the Tbred itself is only scheduled for a handful of speed revisions before it hands off to clawhammer. I doubt a retail Tbred at 2.4Ghz will ever be released, whereas we know Intel will release (and are soon going to) 2.4 and 2.6 Ghz P4s.

So I think the essential difference in speed scaling and overclocking is the design of the cpu architecture, not the manufacturing process (although that obviously helps greatly within limits).

-* <font color=red> !! S O L D !! </font color=red> *-
To the gentleman in the pink Tutu
March 12, 2002 6:32:14 AM

Quote:
the core architecture is different.

The architecture of the P4 is originally scaled to go high, but the .18 silicon capped them. Moving to .13 allowed much further scaling.


And if you had read my posts on the subject, you would know why that has no bearing, the core design is already taken into account , hence the tbred will top out at 2.5-2.6ghjz while the northwood .13 has hit 3.5-3.6ghz

The archetectural designs are accounted for in the higher top speed on the .18 micron process size.

There is no physical reason why the % top speed gain for the transition to .13 will be any less for the tbred.


Look at the p3 for proof, the p3 was NOT designed to go to ultra high speeds, and at .18 it topped out at 1.13ghz, the .13 micron tually has hit 1.6-1.8ghz on overclocks, conservativly thats a ~50% gain, now apply that same 50% to the axp line and you come up with around 2.5ghz, now again I ask.


Is there any reason this would not be true?



Although I agree, amd probably wont release retail 2.4ghz tbreds, because the hammer will come before that and cut off the cores growth, that does not mean the core cant reach those speeds at .13.


"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!
March 12, 2002 6:49:09 AM

PIII is a good arguement, I'll give you that!

Well, maybe it will you know. I guess I'm just in a pessimistic mood!

-* <font color=red> !! S O L D !! </font color=red> *-
To the gentleman in the pink Tutu
March 12, 2002 8:42:01 AM

Eden,
In the post i erased, I should have qualified that they only had ONE hammer running at that speed. The rest were at the same speed as the hammer at IDF. Second, even the ones they got going "fast", are not really near "up to speed". THey have quite a ways before they reach a release speed.
Second, they only had a few running the AGP port.. the rest were still PCI. Thats why I said the Hammer wasnt doing so hot <b>right now</b>.
I am confident the coming months will see a vast improvement. I'll let you know...

Benchmarks are like sex, everybody loves doing it, everybody thinks they are good at it.
March 12, 2002 10:44:21 AM

The one thing I hate is people saying "the P4 is designed to ramp in clock speed", when there is no proof other than the 20 stage pipeline. Ray has yet to prove me other than that, WHAT causes it to go further. Since this is twice the lengh, you take a 0.13m P3 MAX speed, you double it, you get the NW's reference max speed estimate. Around 3.4GHZ. That's what I think.

--
For the first time, Hookers are hooked on Phonics!!<P ID="edit"><FONT SIZE=-1><EM>Edited by Eden on 03/12/02 07:44 AM.</EM></FONT></P>
March 12, 2002 10:46:32 AM

Oh ok that I didn't know. I guess each company can rant and get problems eventually, it's normal that it gets that on a new architecture. I would wait of course. Thanks for letting us on track though man! Don't want full exposure, just some insider tid-bits to keep us until release.

--
For the first time, Hookers are hooked on Phonics!!
March 12, 2002 11:23:11 AM

Fine, I'm just going by empriacle evidence that we already see aircooled (selected) P4s running at 3Ghz. Even the Williamettes overclocked pretty ok I think.

So, on the one hand I see actual processors overclocking, and on the other I see a report that OEMs are just getting samples of Tbred in limited quantities. Call me old fashioned but I'm a "1 in the hand over 2 in the bush" kind of guy.

You know what, you're right. The engineers at Intel, obviously having no clue, took a strategic look at the future of Intel processors and said, "heck, we don't really know what we are doing, let's cobble something together from a bunch of leftover cup-cakes and 3rd grade sciance project and see what it does."

I think it is fair to say that designing a cpu that would scale further/ramp up/have longer legs/get a good production run/need redisigning less/...../yada yada... was probably a fairly important goal for them, and from what I have seen, they are probably going to do it.

-* <font color=red> !! S O L D !! </font color=red> *-
To the gentleman in the pink Tutu
March 12, 2002 12:02:10 PM

Eden, the p4 IS designed to scale higher in clockspeed, and it does.

However, as I have said before the % gain from a process shift is basically the same regardless of core type, the reason being the benifit derived from the cores design is ALREADY accounted for when you figure the %.

When the axp tops out at 1.8ghz and the p4 topped out around 2.2ghz on the .18 micron process the fact that the p4 was ALREADY 400mhz faster will be bore out in the updated new top numbers.

The nw p4 currently looks to hit around 3-3.4ghz in the sweet spot, this is roughly a 50-60% gain over the .18 p4.
(and may even go higher, but the northwood didnt just do a process shift, they changed the core as well, so as always there will be variations)

When the p3 went to .13 micron also gained ABOUT 50-60% and the axp when it goes to .13 should, according to logic, also gain 50-60%.

The same tale is born out on all transistions within the same cores, when the p3 went to .18 from .25 it gained roughly the same % as the athlon going from .25>.18.(of course small things like on die cache and copper ic technology add variations, as seen with the p4, but the same general rule applies).

So the core design has effects on the top speed of a processor when compared to other core designs, but as I have hopefully shown, a process shift has a nearly UNIVERSAL effect on a cores clockspeed.


Raystonn pointed out in another post that design may also put limits on a cores speed.(IE if you shrunk a 486 to .13 it would not run at 23123132ghz) which is true. However design limitations are reworked when a company lays out the chip on the new process so the major timing issues(trace lengths etc etc are removed), so this limit, which exists, has yet to show itself.(except arguably on the p3 1.13 which would not go much higher regardless of cooling/voltage IIRC).

Leading me to ask everyone once more.

Can you, or anyone, think of a reason WHY the axp will not gain 50-60% top clockspeed from the .13 process shrink.

(I seriously want to know, no one has yet(asides from rays design limitation counter, which btw were still waiting for proof on ray) to come up with a physical reason that fits the situation as to why it would not.)

"The Cash Left In My Pocket,The BEST Benchmark"
No Overclock+stock hsf=GOOD!
March 12, 2002 12:35:48 PM

Don't be rude about my Northwood. You're making feel alienated and unloved. So I demand a two page apology addressed to Charlie's Northwood which may just go a small way to repairing the pyschological trauma you caused it.

Never spit into the wind. Never eat yellow snow.
March 12, 2002 1:26:11 PM

Quote:
Can you, or anyone, think of a reason WHY the axp will not gain 50-60% top clockspeed from the .13 process shrink.

I can't. It is a prefectly reasonable deduction.

I think the real questions though are more along the lines of:
1) Will anyone <i>other than overclockers</i> ever see that clockspeed once AMD releases their ClawHammer in competition to their own Athlons?

2) So far it appears that each die shrink of the P4 has (and will) not only increase the overall clock speed, but has also give Intel the opportunity to put components that were stripped out of the original P4 design back into the marketted P4s. This improves IPC as well as clock speed, making each new P4 core's performance boost more than just a simple ramping matter.

3) AMD, while having the opportunity to add more to each Athlon core improvement, has done very little to improve the IPC of the core, even if the IPC is already excellent. (Though I have to admit, supporting both 3DNow! and SSE was a nice move.) That in mind, and considering how much AMD is putting into the Hammers, will the Athlon cores of the future still have a better IPC than the P4s to make their lower clock speeds remain competetive to the higher clocks of the P4?

4) A question to which no one can actually answer yet, but sort of puts number 3 into place: Will the ClawHammer actually have an IPC improvement over all Athlons and be available at the same clock speeds?

<pre>If you let others think for you, you're the only one to blame when things go wrong.</pre><p>
March 12, 2002 2:03:55 PM

"
2) So far it appears that each die shrink of the P4 has (and will) not only increase the overall clock speed, but has also give Intel the opportunity to put components that were stripped out of the original P4 design back into the marketted P4s. This improves IPC as well as clock speed, making each new P4 core's performance boost more than just a simple ramping matter."

Actually, AMD has done the same with Athlon. The only Athlon they may *not* do this with is the Thoroughbred. Each previous Athlon upgrade has seen noticeable improvements in IPC. Palomino brought the most, perhaps, because of the SSE support as well as the major improvement to the prefetch, but the other Athlons generally also saw noticeable IPC enhancements.

"3) AMD, while having the opportunity to add more to each Athlon core improvement, has done very little to improve the IPC of the core, even if the IPC is already excellent. (Though I have to admit, supporting both 3DNow! and SSE was a nice move.) That in mind, and considering how much AMD is putting into the Hammers, will the Athlon cores of the future still have a better IPC than the P4s to make their lower clock speeds remain competetive to the higher clocks of the P4?"

Well, I somewhat answer this in my answer to 2) Namely, you are wrong when you say that AMD has done very little to improve IPC. I'd say the 25% or so improvement in IPC since the first Athlon is a nice improvement.

As for Hammer; we'll see what we see I guess :>



When all else fails, throw your computer out the window!!!
March 12, 2002 2:20:24 PM

Quote:
Actually, AMD has done the same with Athlon. The only Athlon they may *not* do this with is the Thoroughbred. Each previous Athlon upgrade has seen noticeable improvements in IPC. Palomino brought the most, perhaps, because of the SSE support as well as the major improvement to the prefetch, but the other Athlons generally also saw noticeable IPC enhancements.

While I will agree that each Athlon core change has seen <i>an</i> improvement to IPC, I did specifically state that AMD "has done very little" to improve the IPC of their cores. Doing very little can still achive noticable improvements. So I think you have misinterpreted what I said.

Which still leaves my points in tact. Intel is specifically trying very hard to improve their IPC. AMD is also improving their IPC, but my observation is that they are not putting in as much effort towards this, so much as they are simply trying to improve electrical and thermal performance to allow them to ramp to higher clock speeds. The IPC gains in the AMD chips (other than the SSE support, the move to a 133MHz FSB, and the use of an on-die cache) have been mostly a side-effect of their primary goal, not an actual goal in and of itself.

<pre>If you let others think for you, you're the
only one to blame when things go wrong.</pre><p>
March 12, 2002 2:36:49 PM

And What I am saying is you are simply wrong in saying that. Adding SSE support is a significant change. completely redoing the prefetch cache mechanism is a major change. Reorganizing much of the core layout is a major change. If anything, Palomino showed significantly more changes/improvements than the Northwood has over Willamette and the relative difference in IPC improvement seems to validate that. Palomino reflects nearly double the IPC improvement of NW.

Moving cache from off chip in early Athlons to all on chip is a much larger change than simply douybling the existing onboard cache as NW did.

I'd say AMD has worked just as hard to improve IPC of Athlon as Intel has of the P4 and the older P3.

Mark-

When all else fails, throw your computer out the window!!!
March 12, 2002 2:37:58 PM

hmm, what was their goal in releasing hardware data-prefetch?

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
March 12, 2002 2:52:12 PM

<b>As I said</b>
Quote:
they are not putting in as much effort towards this

As much does not mean, nor has it ever meant: none.

<pre>If you let others think for you, you're the
only one to blame when things go wrong.</pre><p>
March 12, 2002 3:14:23 PM

Quote:
And What I am saying is you are simply wrong in saying that. Adding SSE support is a significant change. completely redoing the prefetch cache mechanism is a major change. Reorganizing much of the core layout is a major change. If anything, Palomino showed significantly more changes/improvements than the Northwood has over Willamette and the relative difference in IPC improvement seems to validate that. Palomino reflects nearly double the IPC improvement of NW.

1) That's <i>all</i> in the single Palomino core change though. And the reorganization of the core change was for electrical/thermal reasons, <i>not</i> for IPC reasons. IPC improved from it, but that was a side-effect. Just as SSE support was really done not to improve the Athlon's overall IPC, but actually just to give Athlons a boost in applications which were optimized for Intel chips and not for AMD chips. It provides no IPC boost to applications which already supported 3DNow!.

Yes, the Palomino had many good changes made, and more physical core changes than we saw from the Willamette to the Northwood. However the Northwood lifecycle is not yet complete, <i>and</i> I never gave such a narrow scope. My scope was from the entire lifespan <i>and</i> future plans of the P3 and P4 compared to that of the Athlon.

Quote:
Moving cache from off chip in early Athlons to all on chip is a much larger change than simply douybling the existing onboard cache as NW did.

Your argument here is completely nonsensical and irrelevant as Intel also moved the cache from off chip to on chip in their CPUs as well. The performance gain for this improvement was done by <i>both</i> companies, so the efforts cancel each other out when weighing the efforts put into IPC improvements. Just as moving the bus from 100MHz to 133MH was done by both companies, so again these efforts weigh each other out for comparative purposes.

Quote:
I'd say AMD has worked just as hard to improve IPC of Athlon as Intel has of the P4 and the older P3.

Which you have every right to say. Just as I say that AMD hasn't done much of anything to improve IPC <i>specifically</i> that Intel hasn't also done. The real debate is if the opposite is true or not. Has Intel done or is preparing to do anything more to improve IPC in their P4s that AMD has not done or is preparing to do in their Athlons? I think the answer here is pretty clear. Intel has SSE2, Hyper Transport, and a cache size increase. There is also the rumor of putting the originaly-designed P4's FPU back into Prescott. (Which is admittedly just a rumor, but AMD doesn't even have so much as a rumor to match that.)

Yes, ClawHammer may very well see some of these things itself. However, that was never part of my debate, so it doesn't count. The debate is between the Athlon and the P4 and the opinion of if the Athlon will have any staying power against the P4 in the future based on what Intel and AMD have done in the present and past.

<pre>If you let others think for you, you're the
only one to blame when things go wrong.</pre><p>
March 12, 2002 3:21:53 PM

Quote:
The debate is between the Athlon and the P4 and the opinion of if the Athlon will have any staying power against the P4 in the future based on what Intel and AMD have done in the present and past.

And keep in mind that this is purely <i>opinion</i> as none of us have a time machine to go into the future with. :)  It is mean to be fun. To make us think. A valid response could be as simple as, "My opinion is that AMD will make whatever core changes are needed to the Athlon to keep it competetive with the P4 if for some reason the Hammers fail." While being a generally uncreative response, it is a perfectly viable and acceptable counterpoint.

Have fun with this people. :)  I'm not trying to debate if AMD or Intel is better. Who knows, maybe in the future VIA will surprise everyone with a new CPU. Maybe part of the team who worked on the Alpha will quit and start their own desktop CPU business that will shock everyone. The future is unknown, but that just makes it all the more exciting to speculate on it now. :) 

<pre>If you let others think for you, you're the
only one to blame when things go wrong.</pre><p>
!