I upgraded to 1.5.1 (545c1313). I can confirm that the buzzer option is in the menu again!
Unfortunately, the AWG GUI is still buggy. For instance, if you set
waveform: sine
frequency: 24MHz
then switch to period mode, you'll see that it says "42ns", however you can't actually edit the units digit, and it actually should read 41.666666667ns. If you switch back to frequency mode, change it to 25MHz, and switch to period mode again, you'll see that it says "40ns", but again, you can't change the units digit.
Now to break things, change the 40ns period to 70ns. Then switch to frequency mode and (laboriously) zero out all the lower digits, starting with the least significant. If you try adjusting the MHz ones digit, you will find that you can no longer set the frequency up to 25MHz. It maxes out with a reading of 24,999,999.9Hz rather than 25MHz.
You can "fix" this by cycling the waveform, since it will truncate to exactly 5MHz for square and then 1MHz for the ramp/triangle, and when you cycle back to sine, you can then set to 25MHz again.
One new bug that appears in 1.5.1 is that the reference waveforms appear to be corrupted in two ways. First, they have discontinuities for some reason. Second, if you save a waveform, then save again to the same slot (or sometimes even to a second slot), you end up with TWO traces, both of which are turned on/off by the same reference control. I think there is an off-by-one sort of thing going on for reference waveform indexing, and then some other sort of issue with the memory storage. After trying to downgrade (and failing - error 4, presumably a version increment check), I then grabbed some 4k length refs, which didn't quite work - they would save one reference later. I then switched to 8k, and they appeared to start working correctly. Not sure what's going on exactly, but something to watch out for.
With respect to the CSV, it's been buggy for multiple reasons for a long time.
sequoia reported issues with CSV and unless somebody reports it directly to OWON's support team, I doubt it will get fixed just by posting the issue here.
With respect to the pulse generator, Mechatrommer, you're absolutely correct that I didn't 50ohm-terminate it - just a direct connection. Adding a 50ohm termination with a T-adapter gives the
same result as luma found.
Finally, I'm curious about your experience with the continuity testing. In my testing, the continuity buzzer is SUPER sluggish compared to my other cheapo DMM.
Here's a video comparison.