I currently have a 15 year old Picoscope with a bit lower specs than their current bottom of the barrel model (2204A), in general I am very happy with it but in some ways it is a bit limited, more so with the "new" Pico 7 software as there is a bug limiting the memory usage while streaming and being an obsolete model it doesn't seem to be a priority to rectify.
What I like about my current Picoscope are things like:
Portability, I can throw it and a laptop in my backpack and use anywhere, runs off the laptop battery. Not fragile, no screen to scratch or exposed knobs to break off.
Ease of sharing recorded data. Waveforms can be saved as images and shared anywhere online, or as data files where anyone still can zoom, apply math channels or do whatever, just like I could do right after capturing it. Great for help with analysis.
Good user interface, lots of functions in the software.
Reliability. As said, it's 15 years old and still going strong. No guarantee a new one will last as long of course, but it comes with a 5 year warranty and I haven't seen many complaints about Picos breaking randomly.
What I'm missing:
Memory. I have 8kS for high speed signals (rarely used) or 2*2MS per waveform in streaming mode (continuous transfer over the USB, not saving one at a time waveforms in the 8kS scope buffer) for lower speed signals (max 1MS/S, half that is the practical limit to avoid cutouts in the waveform), up until the total 100MS memory the Pico 6 software allows is full, so close to 25/50 waveforms @2MS per channel.
Faster streaming, or equivalent results using other methods - streaming would be unique to USB scopes, I suppose, and not a limiting factor for bench scopes and such.
More than 2 channels would be nice.
DC offset. I'm currently stuck with voltage ranges centered around zero volts, i.e. -20 to +20V for example. A lot of wasted resolution when I'm just looking at a signal varying between +5 to +15V. Alternatively, better vertical resolution would work, so I could zoom in without losing too much detail.
Upgrading to a Pico 2206B or 2406B would solve that. They have:
32MS memory buffer in the scope.
I think 9.6MS/S max in streaming mode, 100MS max in one waveform (shared) using Pico 6, the newer software Pico 7 is supposed to allow even more I think. I think the auto save feature in the new software would practically allow you to keep it going until the computer hard drive is full, should you have the need to.
2 or 4 channels, depending on which model I choose. There's quite a price difference, and a Rigol DHO804 would cost somewhere between those options.
DC offset.
Reliability? As a lot of the hardware is in the laptop and the laptop is easily replaced, a USB scope can avoid worn switches, aging power supply capacitors, scratched screens and so on. Less parts to break.
Continuous software development. The 15 year old scope works excellent with the last version of the Pico 6 software from a year ago or so, the current Pico 7 software has a bug for these obsolete models but it
mostly works. Doesn't guarantee the same applying to the future, but seems promising.
The possible downsides of 2206B/2406B:
It's a relatively high price for the specifications, compared to for example the DHO804. Bought locally (not a fan of international shopping for expensive items with warranties) the 2 channel would end up about $100 cheaper, and the 4 channel $150-200 more expensive. Rough numbers, I haven't checked all possible sources.
The Picos are only 8 bit vertical resolution. Good enough for just about everything I do, sure, especially using the DC offset, but when you compare it to 12 bit you kind of realize there is room for improvement.
The Pico 2000B models have been around for a number of years now. On one hand that means tried and true, on the other hand it means old, and that usually says you'll pay a lot for limited function compared to newer tech.
The Rigol DHO 804... Well, you guys probably know that far better than I do. When it comes to bench scopes I haven't used anything newer than 1990s analog scopes. Those were in no way suitable to throw in the backpack.
The possibility to have a portable USB C power supply seems nice, that could suit me well.
How practical is it to share collected data?
What kind of life span should we expect out of it? I'd like something that'll last quite a while, when I spend several hundred dollars.
Does the Rigol has the ability to use combinations of functions for math channels? For example, in the pico I have used frequency math channel to show an engine RPM, and then applied derivative to the rpm to get a curve of how rapidly the rpm changes (with low pass filtering to avoid the curve changing so much it 's hard to interpret).
Are there other good options? I don't need much bandwidth, 2 channels may be okay, 4 is better. Math channels similar to the pico would be very helpful sometimes.