BatBrain -- a central nervous system for consult scripts

I think this hasn't been answered yet?

zarqon said:
This is my one remaining question with Soul Blaze: As it's not labeled a spell, does it succeed or fail versus spell-resistant
Click to expand...

Soul blaze against fideaux with snail part:
Code: 
Your spell bounces off of the horrible beast's snail shell and hits you!

Soul blaze against beer golem:
Code: 
The golem sees you preparing to cast a spell, removes one of the bottles from its body, shakes it up, and hoses you down with beer before you can finish casting. Blech.
 
My spading results so far. Please correct me if I'm wrong. The five elemental and one electrical OCRS auras hit you whenever the monster attacks, even if it misses. This includes misses due to modifiers. Stunning, staggering and blocking all stop the auras. Cowardly and frozen are probably 90% automiss like lazy and narcissistic, although I wonder if it's more like 95% for all of those. Twirling seems right at 50%. I know that's not much, but more than I knew before.
 
Last edited:
It appears the wiki is wrong (again) about the damage formula for ranged attacks (see Veracity's post and hee^3's post). I was also able to confirm this although my numbers are slightly off. BatBrain is using the same wrong formula from the wiki. I switched BatBrain to use this:
Code: 
res[$element[none]] = max(0,max(0,floor((my_stat(to_string(current_hit_stat())) - monster_stat("def"))*ltsadj*radj)) +
And now the predicted attack values are much closer to the actual values. They are still off by a little bit, but so are my hand computer ones, because I suck at this (help!). This change I made might break weaponless combat. As I moved where that modifier is applied as well. My own spading says that weaponless attacks and ranged attacks are not the same formula using .25 vs .75, but actually different and apply those factors at different stages, but I don't trust my spading since my numbers are all slightly off (there is something I'm not accounting for).
 
Veracity noted that the damage formula apparently was missing damage resistance initially - is BatBrain calculating Spelunky DR?
 
I can test weaponless.
Bigface (has one B, and I'm in BHY), 9999 HP, 13333 attack, 13333 defense. I have 14457 muscle and am weaponless. 0 ML, 15 bonus weapon damage. Offhand equipped to prevent Way of the Surprising Fist from kicking in. Punching does 15 physical damage.

Tattered, buffed muscle. 57848 muscle. Another bigface. Punch did 1144 physical damage. That fits with 57848/4 - 13333 + 15.

Tattered again, buffed muscle again, this time to 78508. Mercenary of fortune has 11111 defense. Punch did 8531 damage. This fits with 78508 / 4 - 11111 + 15.

I agree with Crowther's findings.
 
Thanks hee^3! Odd that they don't work the same way, but I should be able to code that into BatBrain anyway. Except it's been a very busy weekend until just now.
 
I finally got around to making that change. I split apart the weaponless and ranged modifiers and moved the ranged one out after the monster defense is removed. I left the LTS modifier where it was. I don't have a whole lot of time for the game at the moment, but I wanted to get this done before I forgot it. I did a quick test and the predicted damage for range attacks are more accurate now, but they are still off for some reason.
 

Attachments

Having a new error as of today, using r16119 and batbrain r81, WHAM r43:

When I try to use WHAM to fight a CCS, I get "Script execution aborted (java.lang.NullPointerException): (BatBrain.ash, line 1893)".

That line is: " map_to_file(happenings,"BatMan_happenings_"+replace_string(my_name()," ","_")+".txt"); // do this here (once) rather than in set_happened()"

No clue what changed, but everything still works smoothly with r16115, batbrain r81, WHAM r43.
 
If you're willing to investigate please roll back to r16117 and see if you encounter the problem. If not please try r16118 and try again. If you have the problem with r16118 please report it in the main bug reporting forum. r16118 made a change to how some data types are handled.
 
Valliant said:
When I try to use WHAM to fight a CCS, I get "Script execution aborted (java.lang.NullPointerException): (BatBrain.ash, line 1893)".
Click to expand...
By the way - that would have caused a stack trace to be printed to a DEBUG log. Can you post that, please?
Thanks.
 
Veracity, not exactly WHAM related, but is there a way to set your combat action (change it from "attack with weapon" to "custom combat script") via a CLI command or ASH command?
 
Zen00 said:
Veracity, not exactly WHAM related, but is there a way to set your combat action (change it from "attack with weapon" to "custom combat script") via a CLI command or ASH command?
Click to expand...

Something like:
set battleAction=custom combat script
or
set_property("battleAction", "custom combat script");
 
Fascinating. Well, I suppose a null check would prevent that from happening - but there should never be null values in ASH maps, and it was only because I broke the compareTo method of Value that it got one. That is fixed, now, and this should never happen again.

Thanks!
 
BatmanRe cannot be switched OFF and doesn't calculate actions (calculates forever).
At CLI there is red string: Bad monster value: "clingy pirate" (SmartStasis.ash, line 271)

Also I have issues with two zones where script always autofight after I used any move at first round - it's Yossarian quest and Collosseum at the Sea.
 
You're reporting a mix of issues, none of them in the correct thread. Both BatmanRE and SmartStasis have separate support threads.

BatmanRE can be "switched OFF". Type the following in the gCLI:

Code: 
zlib BatMan_RE_enabled = false

and the problem with the Clingy Pirate in both BatmanRE and SmartStasis was fixed a few hours after it was reported. If you update your copy of the BatManRE script the issue should be fixed for that script. Of course there is now a similar issue with Mimics. :rolleyes:

The fix for SmartStasis may be found here
 
Last edited:
You must log in or register to reply here.
Back
Top