Darren New wrote: > James G. Sack (jim) wrote: >> Can you not create test data streams. Might that not be something you >> end up doing anyway -- maybe recording some live data and reusing it? > > Sure, but it doesn't really test your code in a way that's particularly > helpful. It helps in the same way that testing graphics code says "yes, > the screen redraw function doesn't crash" without telling you whether > you're still drawing the right things. >
Well, assuming it's a deterministic system to begin with, and assuming that it's possible to calibrate/validate it with human assistance (for input X, correct output is Y), then it seems that should at least tell you whether the transform function ever changes. I appreciate that measuring and tolerances can be tricky, but in principle, for a given input, you want a specific output, no? I have no appreciable realtime expertise, I'm just asking whether this is not one way to test. Regards, ..jim -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
