There seems to be a problem in my computer. l11e3H$XrTY# should produce 0.25 seconds of white noise. On my computer it doesn't work quite right. But it happens in Matlab too (not just in MATL), that's why I'm pretty sure it's my computer's fault. Can you check?
With MATL on Matlab, instead of 0.25 of noise it sounds like stuttering, with the noise interrupted by a brief silence.
With MATL on Octave, the sound lasts much less than 0.25, almost imperceptible, as if there was an initial unwanted silence that blanks most of the sound.
The saved file (replace T by F) seems to be correct.
I was thinking, if you implement sound in the online compiler, what if the program produces several sounds? Same with images actually, how is that done currently?
Funnily, we had a very good one about chord recognition. I sent the link to a friend who's a musician, and she liked it a lot. Maybe I will get her into code golf :-P
@LuisMendo It's coming along. I haven't had a ton of time to work on it lately. But right now I can fetch all questions and do some basic stuff based purely on the data from SE (# of answers per user, average score, answers over time, etc.). I've now started integrating the actual parsing of the answers
I assume that it's kind of slow because it has to write the PNG data to a file on the hard drive first before then opening the file again and sending the contents across the wire