Quoll Work Thread, 2024-11-16 #2 - 2024-11-16 19:40
The good news is that I blessed the skein! The bad news is that in doing so I made a change that requires rerunning all the tests…
Most of the changes I was seeing were expected. Mostly differences in how Ada spoke taff-speak because more things she says have been taffificated, so the RNG pulls that control that have changed. Luckily, this is very easy to just accept because the tests that change were not really testing that, it just happened as a natural result of Ada talking to Taffany (which is by design).
However, in a few cases I would notice that when the Playerr would ask Ada to list the TOPICS, sometimes Ada would go on to use the Node-Continuation nag, which is not ideal because for this node Ada tells the Playerr that they should consider listing the TOPICS. Even beyond that node-specific issue, I don’t think it makes sense to nag the Playerr when they do that because if they are listing topics, the Playerr is at least trying to converse and doesn’t need to be nagged, even if the topics listing was actually something I triggered for them implicitly (that is, within another command).
So I added this straightforward Rule:
Last Carry Out listing suggested topics (this is the don't continuation nag after listing topics rule):
now conversed-this-turn is true.
When set to true, conversed-this-turn prevents any of the Node-Continuation nagging from happening... But here’s the
catch: all of my Node-Continuation Rules do a random check to see if they should trigger to avoid nagging happening too
much:
Node-Continuation for glitch-level-1-tutorial-1-taffany-tutorial:
if a random chance of 1 in 3 succeeds:
say "Glitch and the Annoying Skunktaur Puppet seem really excited [start-taffificated-text]for us to meet
[taffany-first-name], we [end-taffificated-text]should probably ASK [start-taffificated-text]one of the
skunktaurs [end-taffificated-text]FOR the next part of the TUTORIAL."
But if I set conversed-this-turn to true when listing suggested topics happens, this random check never fires. And if
that random check never fires, that causes other RNG pulls to change. See, when I run the tests there is a set RNG seed,
and that’s fine, but it does mean that if the RNG pulls being made change, the results of tests change. This has
actually been a continually tricky problem for me, especially since I use randomization for a lot of different reasons.
(Also, I just noticed that I do not check if Glitch and the Annoying Skunktaur Puppet are actually tauric compared to anthro, which should change what Ada says here… Ah, the problems of letting the Playerr change reality…)
Either way, what that means is that there are likely many other RNG pulls that have now changed because any case where a Node-Continuation random check would have happened after listing suggested topics will now not happen, which means the next pull will change, and the one after that and so on… Which means I need to run the tests again, so that will be it for me for at least an hour or so, but possibly for tonight.
But this is exactly why I need to keep working on this on a regular basis because there’re always these little things that crop up and they aren’t hard to deal with, but if I don’t work on it, I can’t find them. To be honest, this aspect had generally kind of tamped down my excitement for working on this project a bit, but I’m too enamored of it to stop. It kind of works out tonight because I also just realized that I’m pretty hungry, so I should go get some food!
Might get back on it after I do that!