I'm curious; were adjustments made to the amp(s) used in the comparison in order to take into account the varying forward voltage drops of the different rectifiers? Typically, a 5AR4 type has a low drop, giving a higher B+ voltage to the amp; a 5R4 a greater drop, with a 5U4G type somewhere in between. Different B+ voltages make for different operating points; different operating points can make a BIG difference in sound. Or not - depending upon the amp design.
Also, indiscriminate rectifier swapping can have unpleasant results;
- a 5AR4 pulls 2A filament current, a 5U4G pulls 3A. Swapping in a 5U4G type in an amp designed for a 5AR4 might stress the power transformer, which on a consumer amp is likely not spec'd for the extra current. Many vintage amp owners blew their power trans when substituting a 5U4 for the hard to find, expensive 5AR4 (particularly before Russian/Chinese tubes arrived on the scene).
- some rectifiers are intolerant to high capacitance; anything over 4uF, and you may (or may not) make a 5R4 grumpy enough to arc. [edit] - 4uF directly after the rectifier.
[edit] - placing a 5AR4 in an amp designed with another rectifier might raise the B+ too high, potentially damaging other components (particularly electrolytic capacitors) if the design is close to the edge on component tolerances. Put a 5AR4 in a Dynaco Mk2, and boom! goes the can cap (Don't ask...)
For me, I buy expensive rectifiers when I have to; for reliability, and protection of my investment. Expensive amp, then quality NOS rectifier. And yes, they typically sound better than current production, but a well designed power supply will sound good anyway.
I think most rectifier differences folks hear are operating point changes. Excluding branding/construction differences within the same tube type, of course. My opinion only.
Now, to really blow your mind - anybody try damper diodes? My recent builds have used pairs of 6AX4's. Love 'em!