OGM : Acceptable percentage of Final size deviation

[tHe gLouCh]

I have done some tests with my OGM implementation.

I wanted to know if these results were acceptable ?

Results:
Duration: 68s - Desired size: 8192 Ko
AVI - MP3 : 8184 Ko -0.10%

OGM-OGG (IL:1x ogg): 8188 Ko -0.05%
For test : OGM-OGG (IL:1x mp3): 8146 Ko -0.56%

OGM-MP3 (IL:1x mp3): 8161 Ko -0.38%
OGM-MP3 (IL:1x ogg): 8204 Ko +0.14%

Duration: 484s - Desired size: 61440 Ko (60Mo)
AVI-MP3: 61298 Ko -0.23%

OGM-OGG (IL:1x ogg): 61575 Ko +0.21%
OGM-OGG (IL:1x mp3): not done

OGM-MP3 (IL:1x mp3): 61371 Ko -0.12%
OGM-MP3 (IL:1x ogg): 61675 Ko +0.38%

Duration: 4554s (1h15m54s) - Desired size: 512000 Ko (500 Mo)
OGM-OGG (IL:1x ogg): 512789 Ko +0.15%

[len0x]

lets move this thread up.

About percentage. Lets say acceptable diff for 700mb is one meg. which makes it 0.14% (Although I persoanlly think it should be no more than 0.10%).

I'm starting to look for a simple formula to approximate that overhead.

[bond]

Originally posted by len0x
Although I persoanlly think it should be no more than 0.10%yup me too

you can also have a look at koepis ogm bitrate calc. here (http://www.roeder.goe.net/~koepi/misc.html)

[len0x]

Originally posted by bond
you can also have a look at koepis ogm bitrate calc. here (http://www.roeder.goe.net/~koepi/misc.html)

looks really interesting. I better contact Koepi for the magic formula.

[len0x]

found this thread:

http://forum.doom9.org/showthread.php?s=&threadid=54120&highlight=ogmcalc

It turned out to be quite simple, so it mught be starting point. We'll see how presise that'll be.

[bond]

1mb per stream is really very simple :D

i hope that you and glouch can find reasonable values as gknot was always know for its great bitrate calc...

you can also have a look here (http://forum.doom9.org/showthread.php?s=&threadid=55711)

[len0x]

simple startegy for OGM didn't work for me when using DivX. I got almost 3Mb oversize with expected total size 350Mb.

On MKV side first calculator I've implemented was right on target with DivX!

[tHe gLouCh]

Originally posted by len0x
I got almost 3Mb oversize with expected total size 350Mb.

Was it on Overhead or on Video size ?

[len0x]

Originally posted by tHe gLouCh
Was it on Overhead or on Video size ?

What do you mean ?

P.S. I'm gonna change OGM calculations back to what you did with oggfactor for audio, but we still need overhead for video...

[tHe gLouCh]

Just to be sure it was not Divx fault !
Maybe it's a dumb remark but be sure to compare Video size desired/real before comparing final size

PS: i have not looked at what you have done with my calculations

[len0x]

Originally posted by tHe gLouCh
Just to be sure it was not Divx fault !
Maybe it's a dumb remark but be sure to compare Video size desired/real before comparing final size
PS: i have not looked at what you have done with my calculations

Oh, I never get size deviations with DivX and Avi container :)

I've done two identical encodes with AVI & OGM and got almost 3mb differenece. It's definitely video stream overhead (coz if I sum up OGM just video + audio - the difference is just 800K, which is AC3 stream overhead).

[nesskiel]

hello,

it's not may be the right place but i've got always a large size deviation with divx and xvid (from 5 to 60MB).that's caused i supppose but the use of trim in order to cut off the credits.i tried without the trim option and i have an almost perfect size from the first encode try.If i use the trim i must do 2 or 3 encodes to reach the exact size i want (ie 700MB).
I think it comes from the size saved from the credits audio track(s)that the encoder doesn't know before muxing the video and audio. is there a solution out there for simplifying my life? :D
regards
Ness