I've seen the term GOP used here but am not familiar with it. The only GOP I've ever heard of is for Grand Old Party (the Republicans).

Could someone explain what GOP means?

Thanks.

Yes, we could, but why would we, its only the second entry. http://www.google.nl/search?hl=nl&source=hp&fkt=2422&fsdt=3594&q=gop&meta=&aq=f&oq=&aqi=g6g-s1g3

I.e. http://en.wikipedia.org/wiki/Group_of_pictures.

Hmm, A GOP can contain the following picture types:

I-picture or I-frame (intra coded picture) - reference picture, which represents a fixed image and which is independent of other picture types. Each GOP begins with this type of picture.

'Each' is incorrect, as there are 'infinite GOP' structures that do not rely on I frames to establish the base image, but reconstruct one from a series of P- and or B- Frames. Quite an exceptional case, but still, it does exist.

From a practical point view, what does GOP have to do with either video recording or video rendering? Is it something that can be adjusted or modified? How does it affect the video?

There is no practical aspect to it from a user point of view; you have no control.

These are technical characteristics about how video is encoded, specifically how elements of the compression operate. Video is ultimately a sequence of photos and a GOP - group of pictures - represents a segment of related ones. The first element is a reference picture (i-frame) and then subsequent frames that specify changes relative to that picture. Thus you can get considerable inter-frame compression when the amount of change between adjacent video frames is relatively small because it can represent this in high fidelity with a small amount of data. You seen this efficiency break down when there is considerable change between frames (e.g., a fast pan, flickering lights, running water, etc.) and thus it requires a lot of data to encode these changes, which is then limited by the bitrate.

With that said, if you understand how the camcorder works and what its fundamental limitations are, it can help you take better video and avoid its weaknesses (e.g., avoid fast pans and zooms).

AVCHD and MPEG2, are common video formats in M2TS files from DVB receivers and DV Camcorders.

Each key frame with its following non-key frames are called a “group of pictures,” or GOP. Non-key frames can NOT be decoded without its key frame, while key frame can be decoded independently.

Some time we want to cut video into pieces. So, if you cut on a non-key frame, you will got a corrupt clip, with some leading frames not being decoded correctly, often mosaic, or glitch. You must copy all the frames in a GOP to make them correct. This, is not frame accurate.

Another way, although not very happy but applicable, re-encode some frames in a GOP to make a new GOP, this could help, and frame accurate, but costs much time. AVCHD and MPEG2 re-encoding usually works slow. Many NLEs re-encode the whole timeline, while not necessary, since many time, only several cut points are there.

It could be a 'smart rendering' or 'smart cut', only re-encode the GOP on the cut point, copy other parts directly without re-encoding, this gives highest speed while reserving highest video quality. (We know, each time the video re-encoded, its video quality will be degraded once.)

Some softwares can do this, like CyberLink PowerDirector, although $100 is a little expensive to me. Others are Womble Video Wizard, only support MPEG2; Smart Cutter for DV and DVB, supports AVCHD and MPEG2, only a video cutter, lacks video effects.

But if you often apply many video effects onto the whole time line, then the whole re-encoding is necessary. Then forget the GOP, who is really a naughty boy, unwilling to be easily handled.

Is it something that can be adjusted or modified? How does it affect the video?

Can't change Gop at the cam level, but you can change it somewhat while re-encoding. The mpeg encoders that come with the more expensive editors like Premiere pro or Sony Vegas Pro will allow you to choose the number of I frames, or B frames, or even overall gop size (12, 15, 18....). But for the most part you need to understand what gop is all about and what these changes specifically do.... otherwise you end up with an unwatchable video.

From a practical point view, what does GOP have to do with either video recording or video rendering? Is it something that can be adjusted or modified? How does it affect the video?

Yes, if you do your own encoding, you can choose the GOP structure. Quality of re-encodes and the optimal choice of GOP structure under a given compression method used to be discussed on fora like doom9.

And, when you put out your own edited product, you could choose your favourite GOP structure.