XP Scaling for Level >10

Unfortunately, yes.

If you want to cut down on work, you can narrow it down to critters above a certain level, e.g. CR 8 and above.

Also, some creatures already have a shared san_dying script (e.g. Hextor Invaders and Drow).

 

Re: Circle of Eight Modpack v7 / v7 NC Bug Report Thread

ToEE never ceases to surprise: apparently the dwarves in the Quarry spawn with flag #32 already set to 1. I also tested with freshly spawned Dwarves (via console) - same result.

Beyond the implication for the XP reduction (it nullifies the script because ToEE then thinks it has already reduced the CR), it raises a few questions:
1. Do other NPC protos suffer the same issue?
2. The internal variable used (obj_f_npc_pad_i_3) has so far been assumed to be unimportant. Is it? Could it influence something?
3. Likewise for other "pad" variables.

Certainly something to keep an eye for, at any rate.

For the Dwarves in particular, you can just remove the check, i.e. use

Code:

modify_CR( attachee, get_av_level() )

instead of

Code:

if should_modify_CR( attachee ):
	modify_CR( attachee, get_av_level() )

It was meant as a failsafe, but most NPCs die only once anyway

 

Re: Circle of Eight Modpack v7 / v7 NC Bug Report Thread

Well, what do you know, I learned another new thing today.

Apparently the npc_get/set script doesn't work with flag #32 - as in, it's bugged - setting any flag will mark #32 as set (has to do with binary representation of integers). So using 31 is certainly better.

The bad news is that Elmo and the Dwarves are spawned with flags 1 & 3 set, respectively. So who knows, maybe some other will have flag 31 set.

To see whether any flag is set, you should type:

Code:

[obj].obj_get_int(obj_f_npc_pad_i_3)

If it returns a number other than 0, at least one flag is set (however, to see exactly which flags are set, you have to convert it to binary representation).

A possible solution might be to switch over the flag function to internal variable pad_i_4 - I'll have to test it (I'll write a script to automatically go through every proto and see what flags/variables are being used).

 

Re: Circle of Eight Modpack v7 / v7 NC Bug Report Thread

Ok, indeed pad_4 is clear for all protos. pad_3 had values preassigned for 16 protos, in the range of bit flags 1-3.
Thus I reassigned npc_get / npc_set to pad_4. (this may influence internal testing, beware)
Tested on the Dwarf and it worked fine (but not on the first time, of course ).

Also attached:
-A long due script piece for Belsornig that will prevent his Temple from regrouping if you're working for him (=have been given the second task).
-Removed debug code from Hold Person (it was writing feedback into a text file for testing purposes, added it a long time ago and forgot to remove that bit)

 

Re: Circle of Eight Modpack v7 / v7 NC Bug Report Thread

Not sure I understand what you mean, so I'll just restate what I've changed:

- For the purpose of XP rewards, we're using flag # 31. (this is already done in the updated script)
- In general, the internal variable used is pad_4 instead of pad_3. pad_4 is clean for all protos, pad_3 sometimes has values preassigned. The change is transparent for you or other users (you use npc_get etc just the same), except if you've already started a game and set some flags before the update. Restarting the game would fix it.
- Don't use flag #32, it doesn't work. 1-31 is ok.

 

In theory, if he were to suffer from that issue (i.e. being born with some flags set to 1), then yes. The game would think that he's already said his piece. That's why I switched to pad_4, which as far as I know is all zeros for every NPC, unlike pad_3.

Specifically for Achan, I haven't noticed that he has any set flags by default, at least when I swept across all the protos.

The ones that did have some flags set were: (when using pad_3)

Burne
Rufus
Elmo
Meleny
Spugnoir
Zert
Raimol
Paida
Thrommel
Jaer
Sargen
Ariakas
Dwarf Warrior
Fireforge
King Battlehammer
Elmo (second version - proto 14723)

And the flags set were all in the range of #1-3.

 

Unless they're using flag 31, it should cause no harm.