nerding

[solipsistnation]

I have 16 waveforms in progmem and can swap them out pretty quickly. It's neat.

I need to hook on 3 more knobs. That should be easy enough-- I just need to acquire the actual knobs.

I should put this thing in a case so it's not just, like, naked electronics hanging out, especially if I add more knobs and the associated wires to hang them off.

Somebody has drum machine code somewhere on the web, including little wavetables of drum sounds. Hm.

Anyway. It's fun! HEY profesor EMBEDDED PROGRAMMING IS EASY! IS iROBOT HIRING??? I CAN TOTALLY DO THIS NOW! WHEE!!!!!

Heh.

Comments

[profesor.livejournal.com]

D00D! UR SK1LZ R00L. U CAN HAZ JORB.

[solipsistnation.livejournal.com]

Yay! I'll be over to set up my desk this afternoon!

Seriously, I already had a lot of respect for people who do that kind of programming, and this has only increased that respect, especially since I know the arduino stuff has been designed to be super easy for people to just mess around with using a fairly high-level language, with easy debugging through a serial port and easy ways to attach hardware for inputs and outputs and stuff... I can only imagine the patience it requires to do this kind of work without all the handholding and so on. Heck, just finding out that I need to worry about where stuff is in memory was a big shift from my usual perl programming, where it doesn't matter and you just declare what you want and go...

[sweh.livejournal.com]

Memory management...reminds me of 6502 assembler code. The last major program I did in that was a recursive Towers of Hanoi solver... of course I ran out of stack space (only 256 bytes; each JSR needs 2 bytes for the return address, add in local variables...) so I had to rewrite it; in the end I used the screen to store the data, which also gave me a visual display of what was going on (BBC Mode 7 for the win!) Huh, that was over 20 years ago!

[profesor.livejournal.com]

I enjoy being closer to the hardware - knowing where code & data are, twiddling actual registers, etc. It's challenging because to be good at it you have to be good at both high level design AND understand all the gory details.

I never really got perl - partially because it hid too much from me so I couldn't form a model in my head of how it was supposed to work :)

[solipsistnation.livejournal.com]

See, I like being a looong way from the hardware. I don't care what it does, I just want it to do what I want... Heck, I don't even like messing with kernels if I can help it, which I guess is kind of weird for a sysadmin.

Heck, I'm not even very good with pointers, although I'm certainly better than I was back when I rode the curve to a passing grade in CS1025 (or whatever that awful class was way back when). memcpy! memcpy_P!

Here is more or less what I've been messing with. It applies more pointer stuff in a few lines than I've had to deal with in the past ten years of perl programming...

[profesor.livejournal.com]

I like to know how things work - partly just curiosity, but also so when they break I can fix them. On the other hand, I don't like messing with things that aren't broken - I don't like messing with kernels either if I don't have to.

For a language designed to simplify some aspects of embedded programming, that PROGMEM interface seems kind of clunky to me. I can see why they did it that way, but it seems strange from a programmer's point of view to declare a variable but not be able to access it directly without calling some function.

[djcozmik.livejournal.com]

JOE WONG - Edible Lingerie Model (I think only stophittinyrslf will get that, but it was worth it)

[zeeke42.livejournal.com]

I know Joe Wong, but I don't get this. I'm pretty sure that's a good thing.