While working on the new version of RUU, I've discovered two (relatively minor) firmware bugs involving SCPI commands (and added to the list on page 1). One of them is old (I noticed it already back in v.01.00.00.03) - one of them is new in this version. When reading display memory from the DSO, you should always get back 1400 bytes (unless you STOP the DSO and zoom out).
1) OLD BUG: When in HighRes mode and reading display memory from the DSO, if the time base is <= 20us, the DSO returns 1398 bytes instead of 1400.
2) NEW BUG: When the DSO is STOPped and reading display memory from the DSO while zooming out (i.e. make the time base bigger so the waveform is shrinking on the screen), the DSO returns (correctly) a decreasing number of display memory bytes until < 6, and then it incorrectly starts sending 1400 bytes every time (which is a nonsense waveform) instead of 2 (a single line).
You can see bug #2 in action (non-destructively) if you run RUU, then RUN and STOP the DSO with a small sample length. Then start zooming out. When the waveform reaches a single vertical line on the DSO screen, RUU will start displaying a wild, nonsense waveform - which is some part of overwritten display memory which the DSO is sending (incorrectly) to RUU. Kind of funny...