Ah gotcha. Yeah it can be a pain to change everything to fixed and there are moments when I suddenly realize something in the filters gone awfully wrong. Turning out that I forgot to shift enough bits for some coefficient!
As always cost vs time, or rather how much time vs the cost advantage :)
M4/M4F have DSP instructions added to the chip. This facilitates a handful of things that make some of the current applications that I am trying to do. As for the FPU, yeah, I can take the time to convert everything to fixed point math, but I rather spend the few evenings I get to play doing other things rather than beating my head against a wall trying to convert all my math to fixed point. The kicker is that when you have done that, and then you need to change something.
If you are looking for bleading edge performance, floating point is probably not the right direction, but considering the FPU and the rest of the DSP that allow for float math to not be computationally expensive, I just use it. In the end, it usually ends up being a mix with most things being fixed point and a lesser quantity in floating point.
You know, I always wondered why people prefer a floating point core to a fixed point one. An older engineer I usually talk to once said, 'the only reason you need a floating point is if you're doing military applications/have cash to spare'. And in general for most of my applications I found that (with proper shifting) fixed point gives plenty of precision for most digital filters(admittedly not more that 10-taps or so).
Any chance you could elaborate on why the Floating vs Fixed? Or is it just a M4 MIPS thing?
I really want to play with some of their parts. I have a dev kit, just I have not yet found the right project yet. They really seem to have some interesting features. I really wish that they had one with a M4F core. I think that this would make for a very nice chip.
I'll see what I can come up with. I probably won't have any pictures of set-up and take down. I usually find myself thinking that "it would have been nice to get a picture of that" just AFTER we finish some tricky bit...
@Elizabeth: You know this could be just the thing to use for you BADASS display.
I know, I know ... the great thing about how I'm implementing the BADASS display is that I can swap controllers in and out -- for the moment I'm going to continue with my current plan (Arduino Mega + chipKIT MAX32), but after everything is up-and-running I may well experiment with a PSoc 4 and a Teensy 3.1 and a ... the world is my lobster.
Replay available now: A handful of emerging network technologies are competing to be the preferred wide-area connection for the Internet of Things. All claim lower costs and power use than cellular but none have wide deployment yet. Listen in as proponents of leading contenders make their case to be the metro or national IoT network of the future. Rick Merritt, EE Times Silicon Valley Bureau Chief, moderators this discussion. Join in and ask his guests questions.