On Oct 20, 7:21=A0pm, John Stone wrote:
It seems strange that there would be such a difference. Modern high quali=
ty
preamps should be transparent. Are you doing any level matching when you =
are
doing your comparisons?
I did some measurements with Room EQ Wizard. Levels are calibrated
with pink noise and log sweep is used for getting frequency response
(phase is also recorded, but I'm not sure what to do with that). I
think the error here in setting the calibration is at least 0.2 dB,
though. I plotted the relative level in dB between each device used
vs. going direct from the DAC to the amp.
http://www.mediafire.com/?spuun4alsdv5e
The devices a
Bryston 3iX tube pre-amp with 1000 Ohm output impedance
Burson Buffer, which is an active buffering device
Gryphon XT solid-state pre-amp with 50 Ohm output impedance
(I think I may have messed up the calibration when measuring the
Gryphon).
I don't know how to do octave smoothing with gnuplot, so you have to
eyeball the trends in the graphs. Maybe someone can suggest something
free that can do a better job plotting audio data and that can take
columnar data from a textfile. However, it seems pretty clear that
the BAT 3iX has a rising midrange and a falling treble.
I also tried measure the Vrms output from the amp with a voltmeter,
but mine does not seem to be accurate over 1kHz or so.
Dave Cook