just wanted to share some perf #s with people.
both memory @ 250mb, both with 100gb wd drives/ide, but the 1.3mhz has a 64mb vcard vs a 16mb vcard for the p3.
movie:angel eyes
dvd2avi frameserve-
p3= 47 mins athlon= 25 mins <-- almost have the time!
tmpgenc encoding
p3=11:13 athlon = 4:10 <-- huge, huge difference! this is encoding to vcd, force frame dvd2avi @ 23.976fps , with motion at high quality slow
those are pretty serious performance gains! My next test will be comparing the athlon to my dual p3/500mhz system... me thinks that the athlon will outperform the dual cpus...
+ Reply to Thread
Results 1 to 7 of 7
-
-
This isn't really very surprising.
The limiting factor to MPEG-1 encoding is almost entirely due to the CPU.
The Athlon is similar in class to the P3, thus, just on pure clockrate differences, the Athlon 1.3 should be 2.9x faster than the P3 450 MHz right?
And this is what your TMPGEnc encoding time shows...
Regards.
Michael Tam
w: Morsels of Evidence -
Wow!!!!! of course you got better preformance out of the amd. It's clock speed is more than 50%. At least compare sim. cpu's before doing test.
NOt saying amd isn't as good as intel. For the money I would rather have a amd machine. BUt I like dual cpu's to much(running 2 p3 800's). When the price of Dual amd boards go down. I will try to pick one up. -
This shows that the athlon is significantly more powerful than a pentium 3... power is NOT directly linked to clock-rate. That is, if you had a p3 that was 2.9x as fast, you would NOT get a 2.9x faster encoding.
-
That is true, Clock rate isn't everything. I agree (ps2 vs xbox cpu) All I am sayin is you are comparing a chip that came out 2 years ago to a chip that came out 6 months ago. Architecture is everything.
-
Hateslife,
That test in no way shows that the Athlon is signficantly more powerful than the P3.
Again let me say the limiting factor in MPEG-1 encoding is overwhelmingly the CPU. A P3 that has a clockrate 2.9x faster than anthoer one, can process information 2.9x faster too (all other things being equal).
As has been shown many times in the past, improvements in encoding speed for MPEG-1 is essentially proportional to the speed of the CPU.
The Athlon is quite similar to a P3 (and slightly more powerful at clockrate equivalence). As the Athlon tested is 2.9x faster than the P3, I can fully expect that the MPEG-1 encoding will be at least 2.9x faster (due to the above point -- MPEG-1 encoding is overwhelming limited by the CPU, not by any other components of your PC). This is what the test showed.
Notice the frameserving test. The Athlon is nowhere near 2.9x faster. That is because that test is not limited by the CPU to as large a degree.
Regards.
Michael Tam
w: Morsels of Evidence -
the new tempgenc is kinda slow but has loads less blocks so its worth it but I was wondering what kind of cpu would I need to convert an avi to an in real time(1:1) at top quaility settings?
Similar Threads
-
In plasma TV "Number of pixels" and "Number of dots" it's not a same term?
By flash_os in forum Newbie / General discussionsReplies: 2Last Post: 23rd Aug 2011, 12:27 -
ffmpeg Number of stream maps must match number of output streams
By tmiller_15 in forum Blu-ray RippingReplies: 1Last Post: 13th Nov 2010, 06:20 -
FFMpeg performance
By aeroliteflyer in forum DVD RippingReplies: 12Last Post: 18th Mar 2008, 19:47 -
CRYSIS performance
By retroborg in forum Off topicReplies: 4Last Post: 13th Dec 2007, 06:55 -
Still Performance
By crocker5731 in forum Camcorders (DV/HDV/AVCHD/HD)Replies: 5Last Post: 5th Dec 2007, 10:31