KVCD: Differences between default tmpgenc vcd?
 

What exactly are the changes you made from the default tmpg vcd template? I am wondering what gives the good results

G

Re: differences between default tmpg vcd
 

The biggest change is the modified GOP structure. The standard MPEG-1 calls for a GOP sequence of I=1, P=5, B=2.
KVCD uses ( at the present time :wink: ) I=1, P=18, B=3.
This is what makes the large difference in file size.
The other things are a lowered audio bit rate, increased size of VBV buffer, droped I frame auto insertion in scene change.

kwag

What exactly is VBV buffer? What does it do?

The Video Buffer Verifier (VBV) is a model hypothetical decoder buffer that will not overflow or underflow when fed a conforming MPEG bit stream. Thus, part of the definition of a compliant stream is that it does not cause underflow or overflow of this model buffer.

Here's the whole explanation:
http://www.bretl.com/mpeghtml/VBV.HTM

kwag

Hi: In linked page it says:

Does it means that we can use until vbv=488, no matter we are doing VCD, SVCD or DVD, as it is mpeg compliant?
Could we use 488 with any SAP without problem?

Hi Prodater,

I recall doing a series of encodes at different VBV values, from 15 to 60, and the optimal (center) value was ~48. Below 35 and above 55, stuttering started to appear on standalone. This were 352x240 KVCDs.
The VBV value will have to be different, depending on the resolution used, because the higher the resolution, the more bits you have to deal with. So normally the higher the resolution, the higher the VBV value, of course, up to a point.
However :!: if you look at an MPEG-2 MP@HL at 1920 x 1080 video, you will notice that the value is indeed 488, so I assume that for now, that is the maximum value :)

-kwag