Just my 2c - for serial comms analysis and decoding I much prefer a dedicated LA to the decoding features included in entry level scopes. Unless you only need to inspect a few data packets that easily fit on the screen, a dedicated LA offers more flexibility and can generally be easier to capture and locate the data you are looking for. A cheap $20 LA that works with the open source Sigrok software is sufficient for most serial comms protocols. If you do frequent decoding analysis, you might consider stepping up a notch to something like the $149 DSlogic Plus LA that has a much greater feature set.
Horses for courses. I use both LAs and MSOs for serial decode, but the workflow and features are different.
o Triggers
Scopes with serial decode tend to have more complex protocol specific trigger capabilities available that cheap USB LAs lack. This has a fundamental affect on how one uses a scope compared to a USB LA.
o Memory depth & streaming
USB LAs often support unlimited streaming, whereas scopes rely on and are limited by memory depth for longer captures. Using a scope's triggers and segmented memory can usually mitigate the lack of unlimited streaming.
o Sample rate
Scopes tend to have higher sample rates over a larger number of concurrent channels compared to USB LAs. This typically impacts higher speed SPI decodes.
o Mixed signal time correlation
MSOs have integrated time correlation so you can directly analyse the interaction between digital and analogue signals.
o Workflow
For longer runs, on a scope, because of the triggers and lack of streaming, one tends to set up relevant triggers interactively and iteratively to capture specific events. On an LA that supports streaming, one tends to set up a long capture and then analyse the results retroactively.
For longer serial data streams, particularly longer slave device configurations, the results of a USB LA can often be more easily viewed and manipulated on the host PC than navigating and interpreting a longer capture on a scope.
I tried decoding on my old SDS1104X-E scope but the constraint of being able to only decode what was captured on the screen was frustrating for me. I suspect the decoding features on the newer models are very similar.
On the SDS-1104X-E, yes, you need to have the entire capture, and have sufficient sample rate for the capture, but it will decode long captures, and you can save the entire decode to a CSV file. The implementation on this scope is certainly one of the better ones IME. The LA digital channel implementation on this scope is, on the other hand, terrible!
The DHO800/900, the decode is flakey AF. It misses bytes in I2C, and it subsamples, although it's not quite as bad as the DS1000Z. Getting a decode export on anything more than a short stream won't work. The SPI decode on the LA channels of the DHO900 series is currenty completely broken. I2C decoding on the analogue or digital channels is a constant fight, and is very unreliable.
There was certainly a significant drawback with the DS1000Z series where it undersamples the capture, so only a few bytes at a time are decoded. Typically you can work around it, but it's slow and messy.
Home
Help
Search
Login
Register
