First, OP (on this disc) and ED (on the second disc) are short versions that are from different recordings as far as I can tell. They have slightly over 3dB better dynamic range than the versions in the OP/ED CD single.
Despite that, it actually gets a lower grade because I have higher expectations for instrumental tracks, which this CD does not meet.
Many microphones, placed in proximity to instruments, are used to record the music to multi-track master, then artificially mixed to create the final stereo version.
First problem I have with the mix is that the placements are unrealistic. I'd go as far as to say that it's arbitrary (as in random and meaningless).
Second problem is that amplitudes and dynamic ranges of different tracks don't have proper relationships with each other. To do this properly, one would have to record another version with just two microphones, with the orchestra arranged in a real hall. Then, using that as a reference, try to reproduce that relationship using the multi-track version. Why not just use that stereo version? I personally wouldn't have minded that so much, not that such version even exists, but having the microphones far away drastically reduces signal to noise ratio, so proximity recording is actually a good thing. It's just that generating the proper final mix requires a lot of work, which they obviously didn't put in.
