Might take awhile to find the material, but believe me, when DVD was emerging, this was talked about quite a bit - especially in all the AV magazines (Widescreen magazine, etc.) It's very true.
"some people say DTS is superior because it's 768kbps compared to AC3 at 448kbps, when in fact it's the actual sound mix that's better."
To my knowledge, there was no real "remixing". Dolby and DTS both received the same master soundtrack (The individual streams), and compressed it to their own formats. (This is not to say that there weren't instances of Dolby and DTS using different mixes, or having a second mix made specifically for one or the other, but this was not the standard method.)
"So theoretically, the Star wars episode I dvd at 448 should be better than the lasers 384, but it's not the case."
Because, again, the difference between the SW LD and DVD isn't just the bitrate. The LD is a different mix as well.
And your SPD LD comparison makes perfect sense.
[EDIT]
Doing a quick search, and I ran across this:
The process of "channel coupling" you are referring to is actually rather more complex than simply "mixing channels". Additionally, the 10kHz number refers to AC-3 5.1 encoded at 384 kb/s. At 448, the channel coupling, when it occurs, is above 15kHz. It is also worth noting that at bit rates above 448, AC-3 can be fully discrete up to 20 kHz, which is one of the factors I credit D-Theater soundtracks with for superior soundstaging. So, while DTS chose to roll off frequencies above 15 kHz with their coder, Dolby instead felt these upper frequency fundamentals were important enough to keep and instead use channel coupling to improve efficiency when necessary. - LINK
There's a lot of good info in that thread, and it gives some idea of why channel separation of the same soundtrack would give the effect of a different mix. Also, I had completely forgotten about D-theater (digital videotape) which had full bitrate Dolby Digital soundtracks. I rarely read a bad review of those versions.
[EDIT]
After locating all the materials of the film, Van brought in Gary Rydstrom, the movie’s original sound designer and the re-recording mixer who had won two of his four Oscars for T2, to begin re-mixing the entire movie and adjust it to near-field listening environments. “We did the mix last November,” Van recalls. He and Lightstorm representatives went to George Lucas’ Skywalker Ranch to remix the movie’s soundtrack for use on the DVD. It took them about a week to do the work needed for both versions of the film and several other elements for the disc. “Luckily in 1993 when we did the Special Edition Laserdisc, we tried to do all the audio materials in six track, which helped us immensely now.” LINK
There's a lot of good info in that thread, and it gives some idea of why channel separation of the same soundtrack would give the effect of a different mix. Also, I had completely forgotten about D-theater (digital videotape) which had full bitrate Dolby Digital soundtracks. I rarely read a bad review of those versions.
[EDIT]
After locating all the materials of the film, Van brought in Gary Rydstrom, the movie’s original sound designer and the re-recording mixer who had won two of his four Oscars for T2, to begin re-mixing the entire movie and adjust it to near-field listening environments. “We did the mix last November,” Van recalls. He and Lightstorm representatives went to George Lucas’ Skywalker Ranch to remix the movie’s soundtrack for use on the DVD. It took them about a week to do the work needed for both versions of the film and several other elements for the disc. “Luckily in 1993 when we did the Special Edition Laserdisc, we tried to do all the audio materials in six track, which helped us immensely now.” LINK
"near-field listening environments" says a lot. This means the theatrical soundtrack was remixed for a smaller listening environment, like a family room. This was not done for early LD releases.