|
From: Steven M. S. <sm...@2B...> - 2007-03-20 13:25:17
|
On Tue, 20 Mar 2007, stefan wrote: > Steven M. Schultz schrieb: > > Why? Your sugessions say the same thing - just using different words ;) > > > Erm, there seems to be some confusion about what I tried to say... :-[ I said that -q should go away and an average bitrate be specified. Then you said we need a file-size mode: > The only thing I wanted to say is, that we do not need a lower(!) > boundary and that we need a "file-size-mode", too (which we do not have Now I have news for folks. Bitrate and file size are related :) SIZE = BITRATE * TIME So given a bitrate and duration you can calculate the filesize. Given the desired size and the time you can calculate the bitrate. QED > currently and which implies a real two-pass-mode). I did not want to Not really. Andrew was partially implemented a 'backup and retry' capability. Only those GOPs that exceed the target bitrate would be recoded. Could be quite a bit faster than a full multipass encoder - but alas that's also where the current bug lies ;( Cheers, Steven Schultz |