...at first I was thinking it was important. but Now as I was testing the best way to monitor a serial data stream is with a long buffer (using best Mem depth), then analyze by using the zoom feature , where the decoding does work. and to record long frames, then zoom and scan the Data.
I'm afraid you've missed the whole point of Segmented acquistions (RecordMode) in the first place. In almost all communications protocols, there are not only gaps between requests & responses, but much longer gaps of 'dead time' between those pairs. By recording just one Frame worth of data at each trigger, you can easily extend the time you can monitor by 10x, 100x, or even 1000x. Using the shortest frames, that means up to 200k (!) msgs could be captured. And Rigol
brags about that 200k frame capability!
Let's take 1Mbit CAN for example. I'd need to oversample that by at least 10x, so 10MSa/s. Even with the superdeep 140 Mpts of the DS4000, that's only 14 seconds. Which is <10k msgs at 10% bus loading (10x/20x worse). With lighter traffic, it's even worse yet. But with segmented captures, 100k msgs could be captured with ExtendedIDs (and 200k for StandardIDs). And if I were using a more selective trigger (say specific address, not grabbing
every message), I could filter out a lot of "uninteresting" msgs, and extend that time coverage even further. But not with the method you're espousing.
I can tell you if I dropped the kind of money a DS4000 costs, I would d@mn sure expect this very useful function to work. Great in theory (and advertising marketing), but
doesn't work in practice is a huge
.