VC-1 StartAnalyze BeginWriting 2-pass Chicken Or Egg Conundrum Nathan 12/19/2008 5:38:01 PM I was wondering how to implement 2-pass encoding for video AND audio using WMF11 and the VC-1 sdk. From what I can tell it can't be done in one fell swoop, as it's a sort of "which came first, the chicken or the egg". For VC-1, you have GetCodecPrivateData which can only be retrieved after StartEncode. Yet, for the ASF profile, you can't set append the CodecPrivateDataSequenceHeader to the WM_MEDIA_TYPE because that has to be done before the WmWriter BeginWriting and you can't BeginPreProcessing until after that. It's looking like the source file will have to be read 4 times (pre and post process for audio and video), then mux. I'm hoping someone can find the goose that laid the golden egg that I have been overlooking. Nathan

RE: VC-1 StartAnalyze BeginWriting 2-pass Chicken Or Egg Conundrum Nathan 12/19/2008 5:41:00 PM And really, I shouldn't have said the source would need to be read 4 times, when I could create 2 different writers and then mux when it is done, but that is still an extra step and extra storage. Nathan