4 ms·
The big down side is of course the limited sample memory. Many (especially the very popular, because cheap, fx2la) pc-based ones simply stream the data, which e
by danhor 5y ago
The big down side is of course the limited sample memory. Many (especially the very popular, because cheap, fx2la) pc-based ones simply stream the data, which enables the use of all PC RAM and essentially infinite capture times.
- MayeulC 5y agoLast time I tried a cheap ($8) saleae fx2la clone using pulseview/sigrok (open source combo for visualization, with open source fx2lafw), I didn't find a continuous option, but that might have been a GUI issue, thanks for the tip!
- danhor 5y agoYou can just set the sample count really high (up to 1 trillion). The samples are not compressed in RAM, so the maximal capture time is determined by the size. For 8 GiG and 24 MHz this means you can capture for ~6 minutes. I tend to simply set the sample count higher than needed and stop the capture prematurely.
- buescher 5y agoIf you have a decent oscilloscope with logic/protocol analyzer features, you can trigger the analog scope side of the tool on data. This is a very powerful debugging tool for finding the physical cause of odd data. Yes, with a pc-based tool you can capture everything to disk and go back and search, but that usually works out to be kind of clumsy.
- kevin_thibedeau 5y agoTektronix introduced some nice LA units 20+ years ago that could be configured with analog mux outputs to pass probe channels over to a scope. The instruments could be cross triggered if you wanted to but it was fun to cycle through a whole address and data bus in real time and look for SI problems.