There's a lot of fuss about how stupid the Rigol engineers are due to the way they've implemented the serial decode on their scopes and how utterly useless it is. I've certainly complained about it too but the issues I had were actual bugs (ie decoding not working in segemented memory) and this has been fixed for quite some time on the DS4000 with FW 02.02 SP5 which is the scope I've got.
I'm sure they may have implemented it differently on different model so please be aware that it's a DS4000 the following tests have been conducted on. If you have experience with another model and want to comment in this thread, please don't just say "my Rigol" but make sure you include which model it is YOU are talking about.
Right, the main complaint people seem to have is that it only decodes what's on the screen and how utterly stupid and useless that is. Personally I don't know, you can only see what's on the screen anyway and why you'd want to look at hundreds or thousands of bytes on an oscilloscope screen is beyond me. But anyway, to test this I set the timebase to 100ms/div and sent a 735byte long string at 38400baud, at this timebase and baudrate the data easily fits on the screen and DS4000 does decode it. You can't see what it says but it does decode it:
To proove that it does indees decode it lets look at the event table, here scrolled right down to the bottom, 735bytes. I'll admit I haven't looked at each and every byte of the string but I haven't seen it do wrong - doesn't mean it never misses just that I didn't see it during the hour I played with this.
In the following screenshot you can see that we can zoom into the start of the string.
Keypoint here is to STOP the scope, if you don't it'll mess up!Now, what happens when we move the start of the string off the screen? Previosuly I've seen it mess up but that may very well have been due to not stopping the scope prior to "moving around" because now it continues to decode and display what's on the screen. Why do you care if the scope decodes what's not on the screen?
And here it is right up towards the end of the data, still decoding and displaying correctly. I'll admit, when panning around it takes a while (a few 100ms perhaps) for the decoder bubbles to "sync up" to actual waveform but it does seem to work just fine.
And here's a final screenshot (this is from another capture but it doesn't really matter) that shows it working in zoomed timebase mode:
So, what am I trying to say? Well, I'm saying that Rigols implementation of the decoders certainly isn't the fastest and I know there's no advanced search and navigation stuff that's available on higher end scopes but I don't understand what seems to be the main complaint about the decoders only working with what's on the screen. I Think that's a perfectly acceptable compromise.
Again, this may differ on other models so if you're commenting please make sure to say which model and which firmware version you're using.
EDIT: Replaced the screenshot with event table, initially it was the wrong one.
EDIT2: Which messed up the rest of it. So edited again.
EDIT3: Thread moved to correct category.