definitely, that's also where all of the fun design discussion emerges. like, how did *you* experience all of these aspects of the game, what did you love, what didn't work for you, what if we changed X, etc

i've mainly just been irked by discussions treating game design as a linear bad↔good scale

that is to say! i'm always down to hear a nuanced discussion with fresh perspectives

i would be interested to hear your thoughts from a gamedev perspective. but i've also grown a bit weary of all the "i love silksong, BUT..." pieces

i'm very much in the target audience for SS. it's been tiring to hear people constantly point out things that i loved about the game as "design flaws"

Meanwhile, Beast's attack is great against flyers, its pogo is surprisingly forgiving, and you can kinda stay in endless high-range fury mode. Architect is enough red tools to just delete crowds. Witch is blue tools, good range, a strong AoE attack on heal, and i-frames on pogo and dash attack.

Wanderer is great for single bosses, but terrible for gauntlets. You don't have time to land multiple blows, low range makes it so you can only hit one enemy at a time (also makes flyers hard to reach), it lacks red tools to deal with both crowds and flyers, pogo-spamming gets punished heavily, etc.

I'm in my second playthrough, trying to use the tools/silk skills/crests that I didn't use the first time. Turns out some things (like gauntlets) are just way easier if you customize for the specific challenge at hand. But I think the memory lockets kind of disincentivize experimentation early on.

I'm gonna take a guess, I might be wrong. Have you relied mainly on one crest in your playthrough, the same one you were using in the prev videos you posted? Because that crest turns gauntlets into pure torture

Animation that you can hear

More generally, commits are cheap to make, and interactive rebase lets you edit, combine, and shuffle around commits as you please. The downside is that it's easier to end up making conflicting commits though.

Not super smooth, but: Make a commit for each feature in each file, with message format `(wip) ${feature_name}`. Next, interactive rebase of your n² commits, sort them by commit message, and mark everything except the first commit of each feature for fixup. Squashes down to a nice feature chain.

the entire camera market might be flipped on its head if any manufacturer just slapped 3 extra programmable dials onto a midrange model

i went back to ISO layout with a modified US keymap a while back. what made it work for me was moving shift to the ></ key, then mapping underscore to the now vacant shift key. a few additional remaps make common symbols easier to reach as well, like swapping [] with (). out there, but comfortable!