'Output YUV data as...' question
 

I'am confused about the the settings for TMPEG's Quantize matrix
'Output YUV data as Basic YCbCR not CCIR601'. Having this option
switched off (as recommanded in some of kwags posts) produces
sigificant less filesize and i can not detect any visible difference
in encoded mpgs when this option is check or unchecked. So why is
this option checked by default?

Can somebody please explain this option and its effects to me....

Re: 'Output YUV data as...' question
 

There are two ways of representing video data, generally called PC mode and TV mode (the latter is technically known as CCIR-601). PC mode means the pixel values are between 0 and 255 inclusive, with 0 being pure black and 255 being pure white. TV mode means they're between 16 and 235, with 16 being pure black and 235 being pure white. In TV mode, values outside this range are called "superblack" and "superwhite", and can cause problems with some TVs.

When TMPGEnc reads the input file, the data is generally first converted to RGB. If your input is in YUY2 format, for example, this conversion results in the existing (correct for TVs) scale of 16-235 being "expanded" to 0-255.

If you check this option, TMPGEnc will not scale the values back to 16-235 before encoding. With the option unchecked, TMPGEnc will "re-correct" the range back to 16-235.

So if you use YUY2 or YV12 source material (say, a DVD) and Avisynth with the ReadAVS plugin, you should always leave this option unchecked.

I hope that was clear enough -- it's a fairly complex issue which requires some poking around in the source code to figure out (at least, it did for me).