pixel wrote:Oh dear. Let me rephrase: you get the highest accuracy there's to have. The irregularities of datassettes didn't exactly escape my attention over the last couple of months. But I didn't really care about measuring them.
They are significant.
On another note: A few months ago, I entertained myself to build a test bench for datasettes which allows me to directly attach the digital input/output lines to a function generator/oscilloscope. I used that to create a reference tape with a 2.5kHz signal. I also generated another tape with a range of different frequencies to study the
transfer function of the built-in filter-integrator of the datasette.
See:
http://sleepingelephant.com/ipw-web/bul ... =datasette
I also used it to convert the playbacks from a datasette to a signal that can be captured by a PC's sound card and generate .TAP files. The reverse also works.
A modified version of this test bench could effectively be used to generate perfect reference tapes assuming you know exactly the type of signal you want to put on there and I would be happy to use it to help you with your project.
The funny thing I discovered about the datasette is that because of the playback filter, if you try to play music with it you'll get junk but the opposite is completely untrue; you could convert your music with a high speed 1 bit DAC and you'll get top quality recordings albeit in mono.
IMHO, one of the worst problems of the datasette is head magnetisation and it is getting more and more difficult to obtain good demagnetisation tools. Magnetised heads weaken the recorded signal with each playback of a given tape.
Be normal.