Hello Rik,

> A new version of SlowJT9, with correct reports and the cycle bug
> fixed, will be uploaded later today.

v0.9.02 installed and monitoring 630m JT9-2 here.

> Regarding the drift: keep in mind that in the JT9-2 -> JT9
> conversion process frequencies are doubled, so any drift also
> will inevitably be doubled in that process.

Yes of course. The drift I measured was the actual RF drift of the
signal, measured visually with a high resolution waterfall external
to SlowJT9. The actual drift passed to the decoder would be double
the amount I reported.

> About the crash: what was the content of the decoded.txt file?
> Maybe that will give me a clue what happened.

Contents of the file:

0359  18  -15   0.3   2188.   0   CQ WB4JWM EM83         JT9

> I have SlowJT9 (in
> JT9-2 mode) running since 9 Nov 20 UTC and it was still running
> this morning (12 Nov 6 UTC). It is running side by side with
> WSJT-X and Google Chrome, no other apps. I must admit that I
> haven't paid much attention to intercepting errors so far,
> another item to add to my to-do list.

I was not complaining! It ran all night the previous night on JT9-1
and for at least 8 hours last night on JT9-2 prior to the crash. I
don't think it is important to put too much work into tracking down
the reason for a few crashes right now. Better to continue testing
and see if it will be worth continuing the project.

Thank you again for your work! I look forward to more testing and I
 hope SlowJT9 can help with trans-Atlantic QSOs on both 630m and
2200m. As I have said several times before, we are in need of some
help on 2200m. JT9-1 requires quite strong signals and there are few
or no alternatives aside from DFCW/QRSS.

73,
Paul N1BUG