P,B, and I Frames
 

I was looking at my custom rendering options in Vegas and saw that you could choose different amounts of these P-frames, B-Frames, and I-Frames. What are these? I have no idea and how do they affect the final rendered video if I change the values? Thanks

Mitch

I-B-P frames are called GOP's (Group Of Pictures).

Basicly, if I can remember correctly, the GOP is how the rendering process will render each individual frames.

I-frames I believe stands for "Independent" frames which is essentially a frame indepedent of previous and forthcoming frames. a Keyframe. It's encoded with its own set of rules.

B-frames get their information from the I frames and are encoded according to the keyframes.

P-frames I think are encoded according to the B frames information.

DO NOT QUOTE ME ON THESE! Wait for someone to confirm this information, but one thing I do know for a FACT is that IBP frames are called GOP (Group Of Pictures) and that it is the way the video is encoded/rendered.

I might have mixed up the B and P frames and their functions, but I know for a fact the I frame is essentially a keyframe.

Messing with the GOP will affect more your file size than anything, dependingg on how you set your GOP it can make the quality uglier or prettier (but bigger).

Mitch, Roger's information is mostly right, but you have lots more reading to do about compression before you go into custom compression settings. Feel free to experiment, but not on a paying job. For now, use a custom preset until you know more about specific technical needs of your projects. Incorrect compression settings can cause problems by going outside the parameters of the playback software/hardware on the other end. Even professional software can cause playback problems due to unforseen issues on the other end. Don't worry, everything should be fine on the default settings and you will probably be learning all about compression as you work on more projects. It's a big topic, so take your time.

I still have lots to learn about compression, but I can make a quick summary.

I = Intra frames that are keyframes. More of these make a more accurate image, particularly during fast-motion sequences.

P = Predictive frames that are partial frames based on what frame came earlier. They record only what has changed in time since the last frame.

B = Bi-directional frames that use information from not only before, but after their point in time.

Using only partial information, the areas that have changed between the other frames, the P and B frames can be very highly compressed. The P and B frames are where compression schemes like MPEG, Quicktime, and Divx can save so much data. The I-frames are also compressed, but they have enough information on their own to make a full frame. HDV has about two I frames per second and does a decent job with fast motion while having a great deal of compression. Less I frames means more compression, but if there is a problem with the tape/transmission there will be visible artifacts until the next I frame (half a second in the case of HDV).

I googled this topic and found this post on another forum that probably does a better job explaining all of this if you are still curious.

http://forum.digital-digest.com/showthread.php?t=9785