My apologies for the slow reply. Also I will be away for the next 5 days,
but I will try to respond to any messages when I get back.
On Tue, 22 Aug 2000, David W. Tamkin wrote:
> OK, I'm confused. I thought originally that you said that the Sony machines
> were combining the right channel of sample N-1 with the left channel of sample
> N. So lopping off one sample of leading silence from the right channel would
> move sample N(R) to position N-1(R), where the MD recorder would pair it with
> sample N(L).
My wording was not exactly optimum. A diagram should help.
Original signal, flat with a noise spike (think of the signal as 'moving'
in the direction of the arrow with time)
| |
L --------------------------|
| --->
| |
R --------------------------|
^
Input samples taken here
The signal which the MD combines to make the mono output is
| |
L --------------------------|
| --->
| |
R --------------------------|
^
Input samples taken here
So, removing the first sample of the right channel before feeding into the
MD will restore the original when it is monauralized.
> But when I tried it with a mono track that peaked at 100%, lopping off a
> sample from the start of the right channel cut the peak to 98.5%, while doing
> it from the left channel instead cut it to 99.1%. The only thing that got
> proper results was the R3's method of dropping the right channel.
What did a straight mono recording with no phase shifting give (or is
that what you are measuring things relative to)?
> Unfortunately I'm having no luck with the files you uploaded; they don't seem
> to come out right for me.
What happens to the files? For the mono one, does the level meter stay on
full throughout or does the reading start to decrease steadily after about
the first 5/6 seconds of the recording of the file? Also, when you say
that the mono track peaked at 100%, do you mean the actual original, or
when it has been recorded in stereo to the MD (comparing with the level
after stereo recording to the MD would eliminate the effects of the
ATRAC, which could possibly account for or at least affect the results).
I have been thinking about this - what is strange is that, assuming
the MD did not have the problem I have described, the same
(frequency-dependant) volume loss pattern would occur when the signal was
phase shifted in *either* direction, which is not happening. If the MD
did have the problem, there would be a 'worse' direction and a 'perfect'
direction when phase shifting and a straight not-shifted recording would
be somewhere in the middle.
Jonathan
-----------------------------------------------------------------
To stop getting this list send a message containing just the word
"unsubscribe" to [EMAIL PROTECTED]
