I think this is a remarkably good performance. Thank you, Mr. Tittle, for the heads-up. This is a case, I think, where the timings are not terribly helpful in assessing the performance. As Mr. Guerrero notes, this seems a very "fast" version based on movement timings (and the Finale is really only 21:00 when you allow for the 40 seconds of enthusiastic applause at the end - normally way too short for my taste), but because of the tempo choices Saccani makes within the movements, it doesn't feel as fast to me as some others with longer movement durations (Compare this 21 minute Finale with Solti's 21 minute Finale for an extreme example). Indeed, especially in the last movement, Saccini slows down in all the right places, producing the grandest climax I've ever heard in this movement, including Bernstein/New York (1961 and 1987) and Martinon/Chicago (1967).
There are some odd recording balances, I assume because the engineer was struggling with huge forces in a live setting (I've engineered a live recording of M3 and, though not as tricky as M2, it was no picnic). For example, at the spot in the first movement where the basses diminuendo in one of their marching episodes and the snare drum enters over top of them at a different tempo, you can hear this engineer desperately seeking a good balance between the two voices, with the drum way too loud at first, experimenting, then settling in. Also, the fifth movement seems to have been recorded at a higher level than the others, which I corrected by dropping it 6 dB in a CD-R burn.
I don't know how I'll feel next month, but right now this recording moves to very near the top of my list, along with Martinon and both Bernsteins. Oddly, the M1 discmate struck me as a non-starter, a real taffy-pull, with rhetorically meaningless tempo distortions and dynamic fluctuations, and less vividly recorded to boot. But the whole package cost only $4.00, so who can complain?
This is my first post after lurking for a long time in the old group.
James Meckley