[Research] YIQ color Space vs. YUV color Space for NTSC

I was doing some research into the YIQ color space for composite/S-Video reasons, and I stumbled on this:

The original color space used for encoding NTSC video in 1953, which was very similar to YUV. Only a small number of TV sets that decoded YIQ were ever manufactured, and it became obsolete. All NTSC encoding today is YUV. See YUV.

So from what this is saying is CRT’s didn’t use a YIQ color space it used a YUV color space.

Wouldn’t this mean the composite and S-Video shaders should be using RGB/YUV/RGB instead of RGB/YIQ/RGB. (To be more accurate.)

As with all of this stuff, the answer to that is “it’s complicated”. :stuck_out_tongue:

The NTSC standard is YIQ, though YUV is used for some NTSC variants, apparently. YIQ decoding was indeed not done much because the decoders are expensive, but that just means that they used a cheap filter to separate the I and Q components rather than actually decoding them (at least, as far as I can tell from reading).

With all that said, YUV and YIQ are extremely similar (the first row of their conversion matrices are identical). Switching maister’s NTSC shader to use YUV matrices makes a very slightly different image (yiq first, followed by yuv:

1 Like

I actually like the YUV image better…

It’s doing something with the red that’s noticable on mobile.