P4 3.2GHz EE Beats Down A64

[knewt]

The first real benchmark I've seen are out. AMD's flagship has definitely upped the level of competition. I have to admit it perfoms rather well in a few benchmarks. But... that being said, the new P4 chip is a monster. It dominated the A64 in most of the tests. Sorry fanboys, better luck next time. There will always be the next generation of superchip from AMD in say 6 to 8 years.

Anyway, here's the <A HREF="http://translate.google.com/translate?sourceid=navclient-menuext&hl=en&u=http://www.x86-secret.com/popups/articleswindow.php?id=91" target="_new">link</A>. I haven't really picked it apart, just scanned through. But I thought it to be interesting.

Quote from the article: "Athlon 64 Fx-51 is a monster of power which one can compare without problem with Pentium 4. Equalizing or exceeding it in many benchmarks, it remains however still behind in the large applications like 3d Studio max or Lightware."

Another quote from the article: "INTEL answered the evil by the evil and Pentium 4 ' EE' is a success in term of performances. Catching up with Athlon 64 FX in all the tests (except for UT2003 in BotMatch) and exceeding it, sometimes largely, in others, Pentium 4 ' EE' are as powerful as expensive and inalienable"

<font color=white>---</font color=white>
Wanted: Large breasted live-in housekeeper. Must be a good cook, organized, and willing to pick up after me.

 

[TknD]

Strange to be making those conclusions after one review...

So here's more info for you:
<A HREF="http://www.theinquirer.net/?article=11674" target="_new">http://www.theinquirer.net/?article=11674</A>

The 64bit minigzip is interesting:

<A HREF="http://216.239.37.104/translate_c?hl=en&u=http://www.x86-secret.com/articles/cpu/k8-2/a64-7.htm" target="_new">http://216.239.37.104/translate_c?hl=en&u=http://www.x86-secret.com/articles/cpu/k8-2/a64-7.htm</A><P ID="edit"><FONT SIZE=-1><EM>Edited by TknD on 09/22/03 12:52 PM.</EM></FONT></P>

 

[speeduk]

How much does/will that EE chip from Intel cost?

<A HREF="http://service.futuremark.com/compare?2k1=7000747" target="_new"> 3D-2001 </A>
<A HREF="http://service.futuremark.com/compare?2k3=1284380" target="_new"> 3D-03 </A>
<font color=red> 120% overclocker </font color=red> (cheapskate)

 

[pIII_Man]

lightwave is so intel optimised it would be extremely hard for amd to pull off a win there, its like a PC beating a mac in Photo shop.


Proud owner of DOS 3.3

 

[TknD]

How much does/will that EE chip from Intel cost?

Probably Extreme Pricing?

I expect it to be either the same price or probably more than the A64 FX. (last dicussed it would be in the 700s range at least)

 

[knewt]

They've already said the chip will be expensive. So will the A64FX.

Flagships are not about value purchases. They are strictly about bragging rights.

<font color=white>---</font color=white>
Wanted: Large breasted live-in housekeeper. Must be a good cook, organized, and willing to pick up after me.

 

[pitsi]

Nahh .. I wouldn't draw any final conclusions yet! We still have Tuesday and Wednesday for that! OMG, it's going to be a lot of reading 😛

 

[SJJM]

yea, this true but look at ut 2003. it is the same, amd beat intel pretty bad. don't look at those test that are optimised for the cpu, look at the other test. the memory test and cpu test should that they are closely matched. they go back and forth, it is just that intel wins a few more then amd.

<font color=blue>"You know, that my backstab attack does double the damage. I can make an off button for him." </font color=blue> 😎

 

[Mephistopheles]

yea, this true but look at ut 2003. it is the same, amd beat intel pretty bad.

Yes, but quite honestly, I don't play a lot of UT and I don't care a lot about it. Never really liked it that much... Besides, it is "the" extreme case of AMD winning, and it does not represent a realistic picture.

don't look at those test that are optimised for the cpu, look at the other test.

Oh really? And why the hell not? I use a lot of 3dsmax, and it is highly P4-optimised. Should I just ignore that? Of course not. What users want is PERFORMANCE. It doesn't matter if it comes from hardware or software! It doesn't matter if it comes from high IPC or high clock! The more performance, the better. That's it.

<font color=red><b>M</b></font color=red>ephistopheles

 

[Mephistopheles]

yea, this true but look at ut 2003. it is the same, amd beat intel pretty bad.

Yes, but quite honestly, I don't play a lot of UT and I don't care a lot about it. Never really liked it that much... Besides, it is "the" extreme case of AMD winning, and it does not represent a realistic picture.

don't look at those test that are optimised for the cpu, look at the other test.

Oh really? And why the hell not? I use a lot of 3dsmax, and it is highly P4-optimised. Should I just ignore that? Of course not. What users want is PERFORMANCE. It doesn't matter if it comes from hardware or software! It doesn't matter if it comes from high IPC or high clock! The more performance, the better. That's it.

<font color=red><b>M</b></font color=red>ephistopheles

 

[SJJM]

Mephistopheles,
that was not directed to everyone, I was only telling him that he should look at the test that shows the all round score. I understand your point, and if I worked with certian programs I would look carefully at that.

<font color=blue>"You know, that my backstab attack does double the damage. I can make an off button for him." </font color=blue> 😎

 

[jclw]

<tangent>
The P4 EE core they have is unlocked. I wonder if they all will be (since they are aimed at the enthusiast market) or if is just as ES chip.
</tangent>

- JW

*Dual PIII-800 @900 i440BX and Tualeron 1.2 @1.74 i815*

 

[Mephistopheles]

Erm, actually, sorry if I was a bit too agressive there... But I was only trying to make a point... (and I did... but not very delicately...)

<font color=red><b>M</b></font color=red>ephistopheles

 

[c0d1f1ed]

The real big strength of Athlon 64 is its extended register set (16 instead of 8 general purpose, MMX and SSE registers). When optimized for it (requires recompilation), applications that require many register could benefit from it enormously. It's hard to put a figure on it, but I'd say 30% will be no exception. Of course, hand-optimized code for either 8 or 16 registers can still be very comparable in performance. It's especially an advantage for compilers that don't optimally manage register usage. For very good compilers like Intel's own, there is hardly any need for more than 8 registers. There's a setting that returns how many times it runs out of register, and this is mostly very little. Of course when it occurs in intensively used code, it does influence performance noticably...

So, what I'm really curious about is how things will evolve in a few months. Especially, will the Athlon 64 optimizations be enough to keep up with Prescott, which won't require recompilation to take advantage of its new architecture?

 

[eden]

Intel says in its Docs that Prescott somehow uses the PNIs with no optimizations or writing needed. Just how truthful is that and how far can it go on its own?

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol:

 

[eden]

The multimedia results have been straightforward with the many past reviews featuring an A64 3200+, which leads us to assume and extrapolate good logical estimates. In that respect, I don't think I'll need tommorow's results to tell me the A64 still can't compete with the P4 in Multimedia.

What I would like, is tons of gaming benchmarks. However, Intel really pulled a fast one with the EE, and it definitely destroyed the BIG gap the A64 could have left. It's smaller now. In the end, we'll see tommorow, and if the gap is not big overall, you will need to look at the combo pricing for mainboard, CPU and RAM for the P4 and A64 FX to determine who really wins

Only thing that did strike me as worrying, was the GZIP 64-bit mode version. That was insanely faster. It worries me because no one tried to dissect the results and explain what happened. If that actually was what 64-bit can bring overall to any x86-64 optimization for the A64, then expect a huge turn in opinion from me, and many others. I'd be definitely surprised. Alas, I don't know if I can bet on that happening.

--
<A HREF="http://www.lochel.com/THGC/album.html" target="_new"><font color=blue><b>This just in, over 56 no-lifers have their pics up on THGC's Photo Album! </b></font color=blue></A> :lol: