[glynnw]

In the past couple of weeks I have been physically rearranging my system from having the electronics on the side of the room  to having everything on a low stand up against the center of the front wall, for various reasons. In so doing I have plugged and unplugged everything numerous times.  On several occasions I noticed faint numbers appearing on the Tortuga's double display, even when turned off, but ignored them at the time.  Today my system was just acting weird, with sound stage problems and such, so I decided to just redo everything.  I even reset the input impedance settings on the Tortuga and then did the auto calc.  After all was adjusted I started comparing playback with different impedance settings, starting at 20 and going to 40, 60, 80 and 99.  All was well at 20K but at 99k the soundstage was skewed to the left.  I use one of the Chesky test recordings where the guy stands at different places on the stage.  At 99 the sound that should have come from the right was coming from the center and the center was coming from the left speaker.  So I have simply reset to 20k and will do some fine tuning around this area. It is really nice to have a control that lets me get this part of the sound staging right.

[tortugaranger]

At 99 the sound that should have come from the right was coming from the center and the center was coming from the left speaker.

I've found that if the audio isn't centered after running autocal at a new impedance setting (2 thru 5) for the  first time it will usually center itself if you run autocal again. The first time through autocal at each setting the attenuation table is empty (or set at extreme limits yielding zero output) so it has to start each calibration step for each LDR with a huge offset. Once it's built up the table the first time, the next time you run autocal it starts each step from the last value stored in memory so it doesn't have to hunt very far to find the new cal point. It's the software algorithm equivalent of if you've done it once before it's easier to do it again the next time.

On the other hand 99k is at the upper limit of the calibration range of the LDRs (we run them between 100 and 100k). So it may well be that it had a hard time maintaining the 99k impedance level over the full attenuation range which could have skewed the balance some. My guess is if you reran it at say 90k or 95k this issue (to the extent it was the problem) would go away.

Cheers,
Morten