Tom's hardware-MPEG-4 Encoding With DivX 4.11, P4 on top?

Toggle sidebar Toggle sidebar
D

damocles

Diamond Member
Oct 9, 1999
5,105
5
81
Uwe Scheffel (whoever that is) certainly seems to think so

link

The latest free DivX codec, 4.11, is a revolutionary product. For the first time, the Pentium 4/2000 will be able to break the real-time barrier when simultaneously converting video and audio signals, even in worst-case scenarios. While it's true that Intel has a head start on AMD, as we clearly see in the benchmark discipline, it's evident that AMD can still hold its own.

(link fixed)
 
R

Rahminator

Senior member
Oct 11, 2001
726
0
0
Dead link so I can't compare any scores, but my XP 1600 @ 1.56 does DivX 4.11 at about 30fps average (just video stream) at relatively high bitrate (1650) and resolution. Eat that P4!

[Edit] I was done encoding Final Fantasy 2-pass encode in just 3 hours!
 
D

Diable

Senior member
Sep 28, 2001
753
0
0
The reason the P4 wins in THG's benchmarks is the Divx4.11 codec is SSE2 optimized. As a P4 owner I still like Divx3.11 better.
 
S

Soccerman

Elite Member
Oct 9, 1999
6,378
0
0
well what do u expect? it's been optimized by Intel themselves!

sheesh, u don't think the P4 would benefit more than an Athlon when running software optimized by intel, do u? :p

They only mention SSE (not SSE2) optimization in the review, but I personally have no doubt that they changed a few things for the P4..
 
D

damocles

Diamond Member
Oct 9, 1999
5,105
5
81


<< The reason the P4 wins in THG's benchmarks is the Divx4.11 codec is SSE2 optimized. . >>



I think it is just SSE optimised

(oops beaten to it)

True, it is optimised but one of the promises of the P4 would be that we would see more optimised code appearing, which is still lacking more than a year after release.
 
S

Survey23

Member
Oct 8, 2001
50
0
0
INtel had always lead the race until the 1900+ xp passed it...so big deal...

The real laugh is the fact that sse2 optimization ( And yes it is obviously sse2 optimized considering the original was already sse optimized and since amd has same sse codes yet didn't realize same percent boost) doesn't seem to reverse the fact of the low fpu making mhz to mhz comparison fall way short. IN most instances the p4 beats the -333mhz slower athlon by 2-5 percent. I knew it was all crap about how sse2 optimization was this end all saver to the p4 lackluster performance...Still laughing...

Big laugh is the fact that most seriuos users don't use xmpeg since it is extremely buggy and slow compared to some other programs. I mean most programs....

Most movies I do I am forcing film at 23.97fps so I guess all of this cpu's here have passed that mark. Hell my 1.4tbird (266fsb) did a 2 hour movie in 2hours and 4 minutes w/o sound on second pass (the real encoding leg of the program). Sound was done with acidlame...I did it with gknot and nanub
 
D

Diable

Senior member
Sep 28, 2001
753
0
0


<< well what do u expect? it's been optimized by Intel themselves!

sheesh, u don't think the P4 would benefit more than an Athlon when running software optimized by intel, do u? :p

They only mention SSE (not SSE2) optimization in the review, but I personally have no doubt that they changed a few things for the P4.. >>



Version 4.11 of the divx codec is optimized for P4's and the optimization weren't done by Intel it was done by Divx Network.
 
R

Rahminator

Senior member
Oct 11, 2001
726
0
0
Doom9 said that Athlon optimizations are in the pipe so expect the next version to outpace P4 :).
 
P

PhiI2e

Banned
Jul 6, 2001
664
0
0
THG is sooo intel biased. Intel had havepaid them for that video they made. With the kind of influence THG has, I would seriously doubt they're not paid to be biased to one party or another.
 
R

RedShirt

Golden Member
Aug 9, 2000
1,793
0
0


<< THG is sooo intel biased. Intel had havepaid them for that video they made. With the kind of influence THG has, I would seriously doubt they're not paid to be biased to one party or another. >>



Right.. Tom has bashed Intel countless times.
 
S

ST4RCUTTER

Platinum Member
Feb 13, 2001
2,841
0
0
I've never seen Tom's Hardware outright "bash" a product, but they do occasionally point out the poorly designed aspects of one.


That said, THG is notorious for screwing up facts to the point where their articles are hardly credible.
 
P

Pabster

Lifer
Apr 15, 2001
16,986
1
0
Diable wrote:

"Version 4.11 of the divx codec is optimized for P4's and the optimization weren't done by Intel it was done by Divx Network."

Riiiiight ... And the patch AMD supplied to reviewers to allow WME to use SSE routines was coded by Intel. Who are you trying to fool?

Intel wrote the code. Just as they wrote the code for the P4-optimized Flask routines. And there's nothing wrong with that. I mean, who better to code a piece of software for maximum performance than the chips' creator?
 
D

DSTA

Senior member
Sep 26, 2001
431
0
0
That said, THG is notorious for screwing up facts to the point where their articles are hardly credible.

Hm. I've found the latest MPEG encoding piece to be a bit difficult to read, language wise. Might have to do with the copious amounts of Vinho Tinto I've had tonight. I'll check tomorrow if the German version makes more sense to me :).

Back on topic: IMVHO it would have been more interesting to see how the new 4.11 DivX Codec would react to Athlon / P4 platform in a more isolated test - i.e. doing a transcode from MPEG1 to DivX MPEG4 since that gets rid of the MPEG2 decoding issues. IIRC the guy behind videotools.net once wrote an article called "benchmarking the FLASKmess". Sums it up pretty well IMO, since I believe it's really tricky to draw any meaningfull conclusions from FLASK based benches.
 
A

AGodspeed

Diamond Member
Jul 26, 2001
3,353
0
0
Gee, I don't think the P4 was that much faster. The largest margin the P4 won by in any of those tests was 8%, but for god sakes, the rest of the tests the AXP and P4 switched leads.
 
N

Nefrodite

Banned
Feb 15, 2001
7,931
0
0
sorta ot, but need an answer:)

Most movies I do I am forcing film at 23.97fps so I guess all of this cpu's here have passed that mark. Hell my 1.4tbird (266fsb) did a 2 hour movie in 2hours and 4 minutes w/o sound on second pass (the real encoding leg of the program). Sound was done with acidlame...I did it with gknot and nanub

you can force a different fps without losing audio sync? how? or have the newer encoder/flask progies fixed this.
 
D

Diable

Senior member
Sep 28, 2001
753
0
0


<< Diable wrote:

"Version 4.11 of the divx codec is optimized for P4's and the optimization weren't done by Intel it was done by Divx Network."

Riiiiight ... And the patch AMD supplied to reviewers to allow WME to use SSE routines was coded by Intel. Who are you trying to fool?

Intel wrote the code. Just as they wrote the code for the P4-optimized Flask routines. And there's nothing wrong with that. I mean, who better to code a piece of software for maximum performance than the chips' creator? >>



Pabster, may I ask how do you know this? The Divx Network guys who post in there own forum have never said Intel did the optimizations to version 4.11. So I guess your basing your asunption on the Intel of Germany where did you get this bit of info?
 
S

Sugadaddy

Banned
May 12, 2000
6,495
0
0


<< why every time i use FlaskMpeg and it crashes?? >>



Because flaskMPG is crap. Use VirtualDub/Nandub or Fairuse.




<< also why can't i copy VOB file to my HD? >>



Because it's encrypted. Use Smartripper or DVD Decrypter to extract the VOBs to your hard drive.
 
J

j@cko

Diamond Member
Feb 13, 2000
3,814
0
0


<< THG is sooo intel biased. Intel had havepaid them for that video they made. With the kind of influence THG has, I would seriously doubt they're not paid to be biased to one party or another. >>




PhiI2e, I think you just haven't been around much eh... THG is QUITE biased to AMD rather than Intel.... So, your statement there is totally the oppposite.
 
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom