Thanks for your reports.
I see some people are calling this a "counter bug". It is important to note that _ALL_ measurements stop working, every single one, whether you have them selected currently when the bug hits or not. Only the Hardware Frequency Counter still works. All measurements, the "All Measure" table included, stop working, as far as I can tell. I have just been using the Statistics, Difference mode windows because they contain a sample counter, so that you can tell how long it takes for the bug to hit. It is not necessary to have the Stats windows up, the measurements will still stop working even if you only have the basic measurements, in the regular small fonts, showing. When they stop, ALL measurements now fail to work. The only way to get any valid measurements back, as far as I can tell, is to reboot the scope. And then it may only be a short time until they stop again. This is a Measure Fail bug, not just a "counter bug".
I've been running tests on both the Old Scope (with Boot version 0.0.1.2) and the New Scope (Boot 0.0.1.3), side by side, and they both have this bug. Both are running SP2 firmware. Sometimes it takes a long time for it to show up, sometimes it takes just a few seconds or a minute. It does seem that signal inputs are needed, I think.... but maybe not.
Here are some "stop counts" that I experienced over the night, using an actual bench setup that I'm using to try to monitor efficiency of an actual circuit that I'm working on. All these tests were at Average acquisition mode, 8 averages, 3M memory depth, with 3 channels + Math active, 2 us/div timebase, 5 measurements displayed:
Old Scope:
stopped at 7k14
stopped at 8k326
(still running at over 19k with no input signals)
stopped at 5k171
stopped at 279
stopped at 4k51
New Scope:
(still running at over 23k with no input signals)
stopped at 715
stopped at 10k272
stopped at 1k37
stopped at 3k910 with no input signals!
(still running at 13k240 with no input signals)
So at 3 or 4 sample counts per second it could take an hour or more for the measurements to stop, or it could take as little as one minute or even less.
I agree that this bug should probably have its own thread, with a poll to keep track of it, but I'm too busy at the moment to start one up.
I have not yet heard back from Jason at Rigol USA. It's 2 pm here in Central Texas as I write this, so it's just 12 noon in Oregon. I imagine he's pretty busy right now too.