P4 vs. Athlon DDR -- FlasK Performance Revisited

[frankel]

Please, please guys, always bear in mind and keep reminding yourself of the fact that the Pentium4 we use to compare with Athlon DDR is 300MHz faster in clock frequency. This is a 1.5GHz QDR P4, and that is a 1.2GHz DDR Athlon.

Maybe the Athlon 1.2GHz DDR is so good that we've got used to see it as a "fair" counterpart of the 1.5GHz QDR P4.

Read the following data table, the fact is plian and simple (please consider the 300MHz GAP):

=============================
Flask /DivX MPEG4 Compression (Athlon is actually faster!!!)
=============================
High Quality iDCT x87 (fps)

(www.tomshardware.com)
Original Re-Comp SSE2 Opti
P4 1.5G 3.83 14.03 18.96
Athlon1.2G DDR 6.43 11.14
P3 1G 4.39 8.03

(http://www.amdzone.com/flask.cfm)
Original Re-Comp 3DNow! Opti
Athlon1.2G DDR 6.85 16.23 17.75

(*We may take the difference between 6.43 and 6.85 as showing the differnces between testing hardware and software environment.)

================================
A Simple Ananlysis of Tom's Data (Athlon is not the loser!!!)
================================
Re-Comp Performance Gap Clock Gap
P4 1.5G 14.03 2.89(25.9%) 300mhz(25%)
Athlon1.2G DDR 11.14 3.11(38.7%) 200mhz(20%)
P3 1G 8.03


:frown:

 

[frankel]

Sorry, I just re-formated the text for an easy read of the data table:

Please, please guys, always bear in mind and keep reminding yourself of the fact that the Pentium4 we use to compare with Athlon DDR is 300MHz faster in clock frequency. This is a 1.5GHz QDR P4, and that is a 1.2GHz DDR Athlon.

Maybe the Athlon 1.2GHz DDR is so good that we've got used to see it as a "fair" counterpart of the 1.5GHz QDR P4.

Read the following data table, the fact is plian and simple (please consider the 300MHz GAP):


Flask /DivX MPEG4 Compression
=============================
High Quality iDCT x87 (fps)

P4 Optimized Re-Compilation(tomshardware.com)
_______________Original___Re-Comp___SSE2 Opti
P4 1.5G__________3.83______14.03______18.96
Athlon1.2G DDR___6.43______11.14
P3 1G____________4.39_______8.03

(http://www.amdzone.com/flask.cfm)
_______________Original___Re-Comp___3DNow! Opti
Athlon1.2G DDR___6.85______16.23______17.75

(*We may take the difference between 6.43 and 6.85 as showing the differnces between testing hardware and software environment.)

A Simple Ananlysis of Tom's Data
================================
________________Re-Comp___Performance Gap___Clock Gap
P4 1.5G__________14.03______2.89(25.9%)_____300mhz(25%)
Athlon1.2G DDR___11.14______3.11(38.7%)_____200mhz(20%)
P3 1G_____________8.03______0_________________0

 

[yathinkicare]

now factor in the price gap


:Q

amd -48% of intel!

 

[yathinkicare]

now factor in the price gap


:Q

amd -38% of intel!

 

[Sunner]

Pretty much equal, kinda dissapointing IMO, the P4 should easially pull ahead assuming the 3DNow! and SSE2 optimisations are fairly equal.

 

[Quickfingerz]

I Want This Version of Flask that's optimized for the Athlon. I'm using the PX3 version that gives me around 14 fps.

 

[LocutusX]

If you're getting 14fps in Flask, you can easily get real-time in mpeg2avi. Do a google search for it and you'll find it.

 

[noxipoo]

so i still have an old computer, whats flask?

 

[IBMer]

What he is forgetting to tell you is that these are even real numbers.

If you have read the site you would know that the numbers are theoretical and not real benchmarks.

 

[Dean]

The way i see it, the athlon is GOOD HARDWARE that will enable software to run well. The P4 is hardware that requires GOOD SOFTWARE to run well.