BatBrain -- a central nervous system for consult scripts

Again, I suspect you have some other source of +spell that you're not accounting for. With capped Intrinsic Spiciness, Stream of Sauce has a very small window in which it actually varies.
Quite possible. I'm going by what mafia reports, but since I'm using the maximizer, I'm wearing all sorts of strange stuff. I should probably try it more naked.

Silent Slam checked out. Prediction and actual damages matched both with and without clubs.
 
One more note: It's a hell of a lot easier to spade spell damage formulas when your +spell% is an even multiple of 100, since then you don't have to worry about which direction the rounding goes.

Also, I retested most of the sauce spells.
Salsaball: 2-3+0*myst, cap 8.
Stream of Sauce: 9-11+0.2*myst, cap 24.
Saucestorm: 2 hits of 20-24+0.2*myst, cap 50.
Käsesoßesturm: 33-38+0.4*myst, cap 100. (Yeah, this one was weird.)
Surge of Icing: 14-18+0.2*myst. unchanged. Unaffected by Intrinsic Spiciness.
Wave of Sauce: 35-40+0.3*myst, cap 100.
Saucecicle: 35-40+0.3*myst, cap 150.

Oh cool. I missed this post and the next, when I took too long to write a reply.
 
Last edited:
Oh, I get something that was really confused me. KoL caps are not the same place batbrain.txt caps are used. Because batbrain.txt just has an average, people have moved the caps deeper into the expression. That's not good. Maybe batfactors needs a range/spread function so which for now just evaluates to the average, so formulas can be entered more correctly.

With the wiki listing caps in myst, batfactors listing caps in multiplier*myst, and KoL actually capping BaseRange+multiplier*myst, this is more confusing than it needs to be.

EDIT: Oh, right. {min,ave,max} already exists.

EDIT: So these new sauce spells are capped like Jarlsberg spells. I really hope the cap doesn't vary with MP cost.
 
Last edited:
So does
Code:
Stream of Sauce: 9-11+0.2*myst, cap 24.
Mean this:
Code:
spelldmgpercent*min({9,10,11}+0.2*buffedmys+spelldmg+elembonus,24)
 
Correct (re: formula). Cap is independent of mana cost (checked with wave of sauce and saucecicle, added Black Body spray).
 
Last edited:
Correct (re: formula). Cap is independent of mana cost (checked with wave of sauce and saucecicle, added Black Body spray).
Thanks. Sorry it took me so long to understand that. No wonder I was capping that spell. I'll add all the sauce spells you spaded out. That's a huge help.
 
Thanks so much hee^3. I've got all of those sauce spells into batfactors.txt and uploaded. Saucestorm was a pain with the split damage, but I got it working. I tested them all capped and uncapped and got everything to match KoL, except Käsesoßesturm when it isn't capped. The disagreement is small, only a few points of damage with actual damage being a percent or two higher than WHAM predicted. The wiki formula is even worse, so I put hee^3's formula into batfators. It's good enough and won't get anyone killed.

Group (aoe) effects are working fine too. I'm not sure why I saw WHAM report too low values against School of Many earlier, but I wasn't taking notes on that multi at the time. I tested against Irritating Series of Random Encounters and BatBrain.ash multiplied the damage correctly.

I'm happy to have made some progress, but finally a bit burned out on testing spells.

EDIT: School of Many isn't listed in batfactors.txt, that's why it didn't work.

EDIT2: School of Many now added to batfactors.txt and uploaded.
 
Last edited:
Hm, yeah, Cheesestorm looked really weird, so maybe I'll go back and recheck that.

edit: Huh, that's weird, preliminary testing seems to reveal 35-40 base damage, with, uh, 30% scaling? How did I think it was 33-38 +0.4*myst?
 
Last edited:
Hm, yeah, Cheesestorm looked really weird, so maybe I'll go back and recheck that.

edit: Huh, that's weird, preliminary testing seems to reveal 35-40 base damage, with, uh, 30% scaling? How did I think it was 33-38 +0.4*myst?

So it's like stench wave, except without the group effect or delevel?
 
So glad to see a bunch of helpful posts in this thread that actually save me work. Thanks guys!

r29 gives us Crowther's previously posted fix for Intrinsic Spiciness (used to be always applied by KoLmafia, so BatBrain had to remove it from non-sauce, but that's evidently changed). I believe this is all that's needed to complete Crowther and heeheehee's splendid work thus far on updating batfactors with new formulas.

In addition, I added a few tweaks to Infectious Bite (you can use it more than once, but only the first use grants MP) and Plague Claws (only the first use grants MP, and it doesn't deal damage unless you've used Bite first). It also considers the stacking damage up to the second use (since happenings don't currently track the number of times you've used a skill, that's as far as we can go at the moment).
 
So glad to see a bunch of helpful posts in this thread that actually save me work. Thanks guys!

r29 gives us Crowther's previously posted fix for Intrinsic Spiciness (used to be always applied by KoLmafia, so BatBrain had to remove it from non-sauce, but that's evidently changed). I believe this is all that's needed to complete Crowther and heeheehee's splendid work thus far on updating batfactors with new formulas.

In addition, I added a few tweaks to Infectious Bite (you can use it more than once, but only the first use grants MP) and Plague Claws (only the first use grants MP, and it doesn't deal damage unless you've used Bite first). It also considers the stacking damage up to the second use (since happenings don't currently track the number of times you've used a skill, that's as far as we can go at the moment).

Oh. I may have to look at batbrain wham interaction for the zombie slills then as wham has its own handlibg for at least infectious bite.
 
Hm, yeah, Cheesestorm looked really weird, so maybe I'll go back and recheck that.

edit: Huh, that's weird, preliminary testing seems to reveal 35-40 base damage, with, uh, 30% scaling? How did I think it was 33-38 +0.4*myst?
I finally got around to entering that. So, we have cold, stench and hot spells all with the same formula now. WHAM was really loving on the new Saucestorm today.
 
BatBrain doesn't know that Cadenza can only be used once a combat. Should BatBrain be updated with the information that candenza is used only once a combat like chefstaff jiggling, or should batfactors be updated with this information like with disco dance of doom?

In the meantime, lacking an answer I added the keyword "once" to all the cadenzas. (I also added the candenza for Shakespeare's Sister's Accordion because it deserves to be used all the time.)
 
It's just one of those ridiculously nice quality of life bonuses the Smith's Tome likes to give. The most egregious is the Hand that Rocks the Ladle's Utensil Twist. (Seriously, Entangling Noodles and two Utensil Twists will kill almost anything that isn't killed by a single Utensil Twist -- Smith Tome almost makes every other Pastamancer spell unnecessary.)

Does BatBrain know how to handle Utensil Twist yet? And the super-dooper bonus from the Hand that Rocks the Ladle?
 
BatBrain is awesome!
Having written my own script to directly interact with BatBrain for the first time, I was surprised how easy it was. It really is an awesome work.

Getting all the new skills into BatBrain is still a work in progress. I managed to write a rudimentary spell damage testing script, which I'll probably post in the scripting help forum at some point, because it's ugly and needs help.

And then there's Friday. No, not VD, but the next special challenge path. Which will likely bring lots of new skills to spade, so I'm clearing my KoL plate as best I can.

Here's a patch to add support for Butts of Steel. As I write, I'm still debating sending it out. It's a bit of a half measure. Additional damage is missing for turtle helmets and shields as well as TT blessings. Some of which aren't fully spaded. Still, this patch is a step in the right direction and will make the skill much more useful for WHAM if you have Butts of Steel.

View attachment butts.patch.txt
 
I've not done any spading re: butts of steel. Also, good to see someone else is picking up my slack (re: spell damage spading etc)!

I also still need to get around to writing a library that uses batbrain data to generate probability distributions as results of actions (most importantly: probability of "winwinwin" = X%, probability of "lose" = Y%)
 
I've not done any spading re: butts of steel. Also, good to see someone else is picking up my slack (re: spell damage spading etc)!

I also still need to get around to writing a library that uses batbrain data to generate probability distributions as results of actions (most importantly: probability of "winwinwin" = X%, probability of "lose" = Y%)
We've see how long I can keep it up for.

I'd really like something that noticed when the damage done was outside the expected range. Ideally you could have a probability assigned to each possible damage value. With a miss just being another value (often zero). That would work for critical hits as well. Then you'd know a formula was wrong if you did a damage with zero probability and could do repeated testing to see if your distribution matched up.
 
Back
Top