Multiple Counter objects with Analog Triggers issues.

[AlexanderTheOK]

Thanks for the input. I was actually using all four analog inputs before. The code above is only here to demonstrate the issue, and to show that it isn't caused by anything like multithreading of my own. Again, with all four inputs, the first is still the only one to count, no matter which one it is, 0,1,2, or 3. Each works fine on their own, but no more than one functional counter can be created it seems.

Does anyone know a different way to do rollovers? I've been experiencing lag spikes in actual code up to 200ms so simply running code that manually checks for rising and falling edges doesnt quite work. I't misses ticks sometimes.

I'll try to use interrupts today to see if I can avoid the above problem (the lag). Will report if it works or not.

[jhersh]

https://usfirst.collab.net/sf/go/artf4010

[jhersh]

Originally Posted by AlexanderTheOK

I'll try to use interrupts today to see if I can avoid the above problem (the lag). Will report if it works or not.

Based on the bug report, the issue is that the Analog Trigger -> Digital Source is what's broken, so Interrupts won't work either. Perhaps just correct the HAL source (i.e. don't change the channel if analog trigger bit is set) and rebuild. Not sure when another release will come out of WPI that will have a fix.

[AlexanderTheOK]

Hmm. I've got the source here. Any resource on how to build the source and set it up for Eclipse if and when I find the pesky line that's messing things up?

[BradAMiller]

I'll try to get this fixed tonight or tomorrow morning and will post a version of the library for you to test to see if it fixed your issue. It looks like Joe correctly identified the issue, I just need to verify an FPGA usage thing. When it's ready, I'll post on this thread.

Now that all the source code is public, we'll put up a page of directions on building the library and the plugins. I'll try to get to that soon.