EQoffline, bots and more..

[Angelox]

Originally Posted by Congdar

I've added some, but I'm trying to add ones that can't be easily acquired other ways. Levitate and see invis can be found on items, or if you have a wizard bot you can get levitate. I'm confused with what you have in the () you want these commands to make is so you don't lev or wolf or seeinv?
see post #750

What I meant was, Instead of the Bot constantly casting group wolf or lev (and lev-laced spells) on you, you could command when needed (make it choose able). But since I'm just a 'spoke on your wheel', I felt i'd consult you before making any changes.

Also for the Yaulp and 'Yaulp like' short -term spells (Fleeting fury), there could probably be some code to where it could be used when engaged only, so as not to be yaulping all the time.

[Congdar]

Originally Posted by Angelox

But since I'm just a 'spoke on your wheel', I felt i'd consult you before making any changes.

I've been the lone programmer on the bots for over a year now, reviving it from non-working to what it is now. I made many decisions on spell lists and ai that worked for me and my playing style. Feel free to make whatever changes you feel are needed. If they work, great... if not try something different. If it's too wacked, i'll say something. I welcome any help on the bot code. Nice work, btw, on the botname000 fix, that one had me baffled. Here's my list of things that need to be worked on.

Command structure reworked. There are too many commands that take multiple(too many) words. Group and Raid commands can be combined and the check for raiding or not done in the code.
Each class with it's own AI code. currently its Cleric(which I did) or all casters or all melee.
Ranger bots need to be able to use a bow... and toggle back and forth to melee.
Spell lists need a second look. I made some interesting choices on what spells to include and what spells to ignore and how to categorize the spells. The Excel file I made would be a good reference and need to be updated.
There's lots of utilities that could be added so that choosing spell lists is easier:
#bot ranger track --- list of mobs on track
#bot ranger track [NPC name] --- tracks the closest NPC with that name.
#bot wizard port [location] --- Port to Wizard Spires with wizard in group
#bot druid port [location] --- Port to Druid spires with druid in group
#bot wizard evac
#bot druid evac
#bot [melee] discipline [name] --- where [name] is the name of the discipline, and melee is either a specific class (Shadowknight) or just "melee" for all melee classes.
#bot [class] activate [aa] --- where [class] is a class and [aa] is the name (or number) of an activated AA. Examples:
#bot cleric activate Turn Undead
#bot mage activate Pet Swarm (don't know the real name)
#bot pacify - aggro reduction
#bot resurrect - get exp back
#bot summoncorpse - if you really want it back
#bot identify - get item lore

[Angelox]

Originally Posted by Congdar

I've been the lone programmer on the bots for over a year now, reviving it from non-working to what it is now. I made many decisions on spell lists and ai that worked for me and my playing style. Feel free to make whatever changes you feel are needed. If they work, great... if not try something different. If it's too wacked, i'll say something. I welcome any help on the bot code. Nice work, btw, on the botname000 fix, that one had me baffled. Here's my list of things that need to be worked on.

I'm trying, but in the C programming dept, I'm a far cry from what you all are.
I can see all the code and have some idea of how it works, the rest is puzzle work, back tracking what everyone has done, and looking for clues. Very slow process, took me a just about a whole week of trial and error with the botname000 fix.

I've made my ax_classic database compatible with the SVN code, so I don't have to worry about updating special codes for ax_classic anymore.
PEQ is light years ahead of my DB when it comes to content, I just have a lot of custom stuff in mine, and someone might enjoy a change once in a while.

One problem I had with the bot code was, when you had a bot group, you leveled much faster because you could take out reds much easier, so if I turned down the exp ration to where it was right for bot groups, then people who played with out bots complained; they were not getting hardly exp from blues and lite blues. I found a solution to this, might be of interest, I'll look it up and post it.

[Rocker8956]

I just starting using Bots on my minilogin server. They are fun but I noticed something strange.

My bots hit for about 250 dmg at lvl 4 with no weapon.

I tried setting their mindmg to 1 and maxdmg to 8 in the database since they were zero but that did not help. So I then gave my lvl 7 beserker Bot a rusty warhammer and he hit for 25 dmg almost every time.

This is using svn rev 69 (fairly certain that is the version I am running atm)

Is anyone else seeing this or did I screw up my database?

Thank you

[Rocker8956]

Small update...

When I set my bots' level to 64 it hits in a more appropriate range.
If I set it back to level 1 it hits for 250+ on mobs in droga

[Congdar]

I copied over the bot src from the public release code, but the svn code has some things different that don't work the same way the public release code does. The hate list is different and i'm guessing some of the combat stuff is different too. I'll be looking into the differences and making updates.

[Congdar]

when you say 'set my bots' level...' how are you doing the 'setting' ? I can take a guess that bots do not work with GM 'level' commands and that your bot was really your level and was squashing level 4 things at a high level.

[Rocker8956]

I would set my level to 65. Zone into droga then spawn my bot.
He was hitting normal for a naked level 65 (will need to look at numbers again to get exact but I think it was between 15 and 20 for most hits)
I then zoned out of Droga, set my level to 1, zoned into droga and spawned my bot. He was level 1 but smacking stuff for 250+ while naked.

It could be my database or code. Can someone else give this a try and see if they get the same numbers?