so far I’ve been focused on making sure the profiles match the plotted spec. even though my measurements were all within my self-imposed spec tolerance, the behavior of the profiles interacting, specifically the wow/flutter values when toggled, was wrong.
this means when the user goes from TASCAM 424 clean to worn, they get less wow/flutter, not more. my tests were passing because I was only testing a small portion of the big picture. it’s like if a building inspector signed off on the blueprints [1], then after construction was well underway and several subsystems installed, came in to inspect the work, then signed off on the plumbing without even turning on the water. just because the spec is correct, doesn’t mean it works right in practice.
testing behavior is where it’s at
sigh. as I just wrote that line, of course it is. over the years of software development, I’ve shifted from being wholly focused on unit testing (testing in isolation), to integration testing (test the subsystems interacting). I know there’s value in both. unit testing is for validating the shape of each part stays the same (spec testing, in this case). integration testing is for validating the parts interact correctly with each other.
it’s not any different just because it’s different
i guess my mind categorized this software project as different because it’s not web software. in the end, software is software is software; parts working together to produce output when users do things through the interface.
or maybe
maybe it’s that this is all very new to me. when I’m building a web app, I already have all of the parts mapped out in my mind before I even start writing any code. I’ve been writing web apps for a long time, so there isn’t a whole lot involved that I don’t already expect. I’m not adept at building digital pedals. and honestly, this pedal is way different than the first one I built, and is way different than the next one (also in progress, but not even close to be ready to talk about yet). several things I’ve learned with this pedal are being used in the next one, but it’s different.
when you find out you have a long way to go
yes, the title of this post is a gilmore girls reference. my wife and I have watched through the whole series, start to finish, four times. she’s watched through at least double that amount. the second series was shit, but that’s a whole nother blog post.
the love tapes [2] pedal is taking some serious shape, but as I’ve found, I have a long way to go to get the behavior right. luckily for me, I have some great tools and new skills to help it get there.
[1] I’m pretty sure they don’t actually sign off on blueprints, but I’ve never built a house, so I don’t actually know. it’s a metaphor, people, lighten up!
[2] okay, love tapes is sticking as the name. final answer (maybe).
[*] happy independence day! but seriously, the USA still sucks ass for immigrants, non-whites, and almost all women (at least the ones who don’t swallow the patriarchy as they continue to rape the life out of everything). I’d much rather be celebrating the hard work everyone is doing to continue to fight the white nationalist and fascist uprising that’s trying so desperately (and failing) to take hold. is that related at all to effect pedals? nope. but if you don’t like it, you can fuck right off; this page and the things I make aren’t for you. if it weren’t for the checks and balances in the courts, we’d be fucking cooked right now. congress has been all but worthless because the GOP are sycophantic babies afraid to stand up to Trump for fear of losing their power. you have to be willing to lose to actually do anything worthwhile. sitting at home and praying the world gets better actually allows it to become worse.