Has anyone Filmedout GH1 Footage?

[Lpowell]

Again, the P frames ONLY CONTAIN movement vectors, they do NOT contain any visual information like an I frame. You are correct that a P frame can be referenced back to the preceding frame, because it was predicted from that frame.

The changes aren't "compressed" at all. That is fact. during large movement, you have lots of changes and therefor larger bitrate spikes(as you know) and then when there is little movement, there is little to be stored in the P frame and very little bitrate change(as you also know). I think you may be confusing "compression" with "size" in this case.

I'm sorry, Svart, but you're out of your technical depth here. I've had hands-on experience optimizing MPEG and AC-3 codecs down to the assembly language level for well over ten years. It is simply incorrect to claim that P-frames "do NOT contain any visual information". Given adequate bitrate, a P-frame can encode just as much image detail as an I-frame. The reason P-frames are usually smaller than I-frames is because they encode only as much data as is needed to reconstruct the frame. That is what we mean by the term "compression".

 

[Svart]

Right and it seems that you are looking at AVC from the point of view of MPEG design. P frames in AVC generally only contain change vectors. There is no amount of bitrate you can throw at them to contain full images like those of I frames. They will always lack some amount of data.

Now I will concede that in earlier forms of intra-frame compression, like MJPEG, P frames *could* be specified as either image data or motion vector, AVC is almost completely motion vector. I think that is where the confusion is coming in.

 

[Lpowell]

Motion vectors are hardly the only type of encoding used in AVC P-frames. The range of encoding tools available to the codec is much more sophisticated than what was available to MPEG-2 codecs. The intricate complexity of AVC encoding is why its image quality is much better than MPEG-2 at higher compression ratios and why decoding requires significantly more CPU bandwidth as well.