Calibrating with a DVD for a DVD is pretty much straight forward and using HCFR for calibrating with an eye-one is easy - but doing it for a satellite is not so easy - how do you do it for satellite?

For example - when i use Toms suggestions for color via Dvd patterns for white and then red - no problem - but with sat or cable it seems to me the only way to do it is via "eyeball". Yes i use the saved patterns on my dvr from a channel (can't remember the name of the channel that shows them on saturdays) and they work great for contrast, brightness and sharpness - but color is a diffrerent story. And by the way - my calculated color via a dvd for dvd watching is -7. But my eyeball calculation for sat is -13.

Any ideas???

You can't. There are way too many variations in broadcast to even attempt calibrating colors. Your best bet is to eyeball it.

As D-Nice said, the channels vary widely. I set up my displays with a calibration DVD and then start with the same settings for my DirecTV programming and tweak the settings to get the most satisfying images from the "best" (my favorites) channels. It won't be right for all channels, but will be "close enough . . .".

There are calibration test patterns available from at least one of the satellite channels (I don't remember which one - depends on which satellite service you are using). You can record it, if you have a DVR - it's on early in the morning, for about 10-15 minutes. Check on the "HDTV" AVS forum, or you can go to the DBSTalk forum (you have to register - same rules as the AVS Forum).

Unfortunatelly the broadcaster patterns are made in a way you need to move the probe around the screen instead of leaving it in the same position....

For example, Sky Italy sports the following pattern:

http://img104.imagevenue.com/aAfkjfp01fo1i-10347/loc898/48452_Sky_Italy_Pattern_channel_997_122_898lo.JPG

The limitations of the above pattern are in the grayscale.
I'm in contact with them and suggesting a new more usable pattern:

http://img221.imageshack.us/img221/1944/testcopyif0.jpg

I've placed 10 grayscale steps as much as possible in a central zone of the screen to try to keep as much as low light follof on borders. Also with this pattern you need to move the probe around the screen, thus it may be unreliable, but at least we could giveit a try...