But i'm also considering to drop the DSO's LA in favor to an external device, so i put on order a DSlogic plus to evaluate this solution, if it works as expected, the DSO choice will be easier.
I was thinking along similar lines. It seems that an external LA might be a better option than the ones built into these scopes. I was considering the Hantek 4032 as it has 32 channels. I also looked at the DsLogic, but don't like the way that they have essentially taken the Sigrok project work, gave it their branding and passed it off as their own without giving the Sigrok project credit. Still, technically I hear it is meant to be a good option and is affordable.
If I went down the route of an external LA, then I might want to find non-MSO oscilloscope that can cover a similar bandwidth range to the MSO5000 or SDS2104X Plus for a similar or lower price range. However, I don't think there is such a thing on the market. I am therefore wondering whether it might make sense to simply save the money I am considering spending on an LA probe and purchase a separate LA instead. Sigrok covers so many more protocols than either of the MSOs. I already own one of those cheap $25 ones, but need something a bit faster. The Saleae might have been an option if the 16-bit version didn't cost more than an MSO!
This is eternal debate.
Some facts.
What most people nowadays call LA is not really a proper LA. These are some simplified devices that can record state from digital inputs, coupled with some software that can analyse such streams.
Real old school LA had staggering complexity and almost infinite capabilities compared to modern "toy" devices.
Problem is that once speeds on certain interfaces went too high, and most of slow data transfer(and some of fast ones) was migrated to on board serial protocol links of all sorts, these old school LA stopped being made. There was not enough demand in industry to keep making them. We came to the point that new DDR4 and DDR5 interfaces cannot even be instrumented directly to connect to some measurement device without destroying the very signal margin you try to measure. FPGA devices are being equipped with internal eye diagraming instrumentation to be able to verify SI..
Etc etc..
I don't deal with legacy 8 computers or retro computing electronics. So even if I bought ZeroPlus LA (that has both trig in and out and can do synchronous sampling like a proper LA) I used it maybe 5 times total in 5+years.
It support literally more than 100 protocols, and is well made device.
But for modern electronics it is either what you can do with simple MSO scope or you need 5GHz+ BW and probes that cost as much as my apartment.
As I said, many a time this was discussed and always ended the same: everybody should own a 20something $ Ebay LA that can be connected to sigrok. Why: it cost nothing, and while limited it can do a lot of stuff. Sort of a it's better than nothing argument, but in this case way better. It's clunky and stuff but you can do surprisingly lot with it. 20$ Ebay clone with Sigrok will have same rudimentary triggering as Saelae and actually more protocol decodes....
For people with more money there is DSlogic and those with even more Saelae.
DSLogic actually have much more capable hardware and triggering than Saelae, which I consider an Apple of these LA: deliberately simplified device but what it has works out of the box.
Saelae is functionally equivalent to 20$ Ebay clone, so you are paying for software, in this case just works ease of use.
DSlogic OTOH is much more powerful.
These 3 LA devices all will work very well for decoding and looking into long sequences of serial messages.
They are very good to a software person writing software stack on a solid proven hardware.
They will not tell you if message was sent wrongly because of software or if it got corrupted by signal integrity issues on the bus. That is very important on multipoint buses and noisy buses.
For that you need scope that has to have sufficient bandwidth to look at the communications bus SI in analog domain.
And this is how it was made long time ago: with analog scope you were looking at SI, running all kinds of synthetic data patterns (PRBS or specially designed pathological patterns to provoke certain error states) and once engineer doing that was happy, hardware was passed to software guys to do their thing.
With time MSO scopes appeared and new stuff was possible: time correlation of analog signals and digital data and messages. Also with decoding from analog channels, you could do both SI and message debug at the same time. One place where that made big change was CAN bus, for instance... Nodes stepping on each other is much easier to spot with MSO, while on CAN message level you just saw an error ...
That is a landscape in short. Everybody decide what is important for them.
Home
Help
Search
Login
Register
