digital Filter in CD Player ?

[Michael Buchstaller]

>you say 0dB, do you mean full-scale swing?  It could easily be the case that
>the filter looks for artificial (perfect) edges (like your square wave) and
>removes them.  If your levels cause a single bit to vary on every edge, that
>would look even more like an "error" to such a filter.

Yes, i have made the signal go from -16384 to +16383 and vice versa.
I wrote a simple program that generated those files.

>	I know this sounds like sacrilege, but perhaps if you add some very low
>level noise (twiddle the LSB randomly), you can fake out the filter.  Or
>perhaps round the corners ever so slightly.

Next week i get a CD player from Nakamichi, an OMS 3 E. Maybe this one
will perform better.
If not, i will try to do some ditherung of the signal as you suggested.

>	Your standard test signal CD sounds like a great idea.  Are you making
>frequency sweeps and other "patterns" as well?

Not at the moment, but i have planned to do at least some slow upward and
downward modulated sinuses to find out resonance frequencies of speakers etc.


-Michael Buchstaller