Concerning VDMod - what's up with those 5kbps?

[Gew]

Hiya!

Here comes another I-country question..

When about to encode (and print in .AVI container) through VirtualDubMod, using Xvid/vfw (koopi default latest build), I enter the specific length for audio/video (in the internal calc feature), to get appropriate bitrate for desired output size. I get the calculated bitrate for desired container, so that I can easily copy/paste it into the field. However, you don't even have to do that, 'cuz the desired bitrate-value is automatically inserted in the bitrate-field when "OK-exiting" the calc window.

BUT(!!)..
An additional 5kbps seems to be added to the bitrate that the calc had first given me, always.. sort of.. Why is that?

Is it 'cuz of the specific way VDMod produces its AVI container (a mixture of legacy/OpenDML(?) or otherwise, somehow, whatsoever), and thus "correcting" the xvid-calc-given bitrate, for output..? Proposing 5kbps of bitrate extra.. should I take the offer? ..or overriding by re-entering the digits that xvid-calc gave me in the actual calc window.

[Gew]

Just had my late lunch, and kept thinking about this. Then it (reality?) suddenly hit me. After all, this occurred when setting file size in final pass of 2nd-pass-encoding. Did xvid add these 5kbps on top of the bitrate-calc-given bitrate after considering the stats-file from the 1st pass?

<XviD engine> Hmm, a little low-need here and there, I guess overall I could offer him 5kbps more..

Something like that, am I on the right track? Or is it still a mystery?

[Gew]

No, it isn't that.

I keep on fudding around with single-pass mode and applying different scenarios in the calc window. It keeps on adding a few kbps of bitrate when I press ok, as for 'target bitrate'. Average bitrate is not equal to target bitrate, for some matter. I mean, it couldn't possible pre-calculate the real(!?) outcome of "average" anyways, so.. How come it keep changing the value?

[Gew]

Finally, I've managed to wrap my issue/question into a final one-sentence line (I hope).

Here goes..


De facto:
When using xvid's bitrate calc, the "average bitrate" number given in the calc (after feeding it i.e. audio/subs/proposed container) is always a few kbps lower than the "target bitrate" it feeds back to rules for encoding.

Question:
Why is this? Is it as simple as xvid "rounding off" upwards? Or is there some more advanced underlying basis?

Finally:
To achieve (closest to) the desired file size, what value should I use? Should I leave it at the proposed a-few-bits-higher-than-calculated, or copy/paste the "average bitrate" determined in the calc window?


Damnit, couldn't bring it down to a one-liner anyways.. ;p