Timeline for Would a PIC suit me better than an AVR?
Current License: CC BY-SA 3.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 23, 2011 at 19:27 | comment | added | dren.dk | Well, in my experience there are two kinds of code, the complicated algorithms that need a proper unit testing framework and proper debuggers, those get tested on the host, not the target, the other kind of code is the simple hardware manipulation code which can only run on the MCU and is frequently timing sensitive, so you can't break in with a debugger and get a useful result. | |
| Sep 23, 2011 at 17:57 | comment | added | Olin Lathrop | @Kevin: The lecture you linked to sounds like he's got a axe to grind. He dismisses debugging as difficult and time consuming to set up but provides no support for that argument. He even advises using printf and blinking LEDs first because they are supposedly easier. I don't know what debuggers he has used, but that's completely backwards from my experience. Plugging a RealIce into the USB and the other end to the ICSP port on your board, is pretty quick and simple, certainly faster than soldering LEDs on and modifying the code to drive them appropriately. | |
| Sep 23, 2011 at 17:18 | comment | added | Kevin Vermeer | Here's an outline for a lecture I attended which supports your view: andrewsterian.com/424/Lecture23.pdf - He has a very similar priority/debugging set, a notable addition is having someone else look at the code. | |
| Sep 23, 2011 at 12:06 | history | answered | dren.dk | CC BY-SA 3.0 |