Some interesting reading/tools from MSU:

Brief explanation about some of the most frequently used objective metrics (PSNR, SSIM, VQM, ...), with some formulas and image examples:

http://www.compression.ru/video/qual...e/info_en.html

A tool to compute the above metrics:

http://www.compression.ru/video/qual...t_tool_en.html

Some standard methods for SUBJECTIVE comparison of Video:

http://www.compression.ru/video/qual...s_info_en.html

Basically two videos are presented to the expert that gives a grade to each one, the various methods are variations on this theme.

A tool to perform the above tests (SUBJECTIVE), and to compute the average results (because to draw valid conclusions you'd better run the tests several times, with different experts, and average the results some way):

http://www.compression.ru/video/qual...y_tool_en.html

The subjective tests are the most interesting part here, because they can help us to decide what "looks better" based on pure subjective judging, but hopefully with no bias and in a standardized, statistically valid way.

And there is nothing else to say. That's the kind of test I used to participate 15 years ago when the specifications of MPEG2 were not finilized yet. Everything else is just bla bla bla (and "masturbation de neurones" in french in the text, the aimed personne will understand ).