More questions on Bots

[Angelox]

The bots are at a point now where you only need download from the SVN and play.
Windows executables are already pre-made and posted. And usually, who ever runs a Linux server will know how to arrange the makefiles.
If someone will tell me this works, maybe I can come up with a script you can run before you compile so you can select 'bots' or 'no bots'. Or maybe someone else wants to do this, maybe place it in the SVN. I was thinking of a Linux and Windows version.

[hayward6]

Originally Posted by Angelox

The bots are at a point now where you only need download from the SVN and play.
Windows executables are already pre-made and posted. And usually, who ever runs a Linux server will know how to arrange the makefiles.
If someone will tell me this works, maybe I can come up with a script you can run before you compile so you can select 'bots' or 'no bots'. Or maybe someone else wants to do this, maybe place it in the SVN. I was thinking of a Linux and Windows version.

I think bots have been generaly excepted and would benefit most servers. Wouldn't it be easier to just impliment it across the board, and then go in after and raise the status needed to run the bot commands in the database? That way everyone will have working bots if they want it, with little work, and if they don't they can change the command access.

[AndMetal]

I think the logic is that, because not everyone is going to run bots (although they are quite awesome), it's better to use a preprocessor definition during compile than a rule, since it will decrease a lot of the required resources (memory/processor cycles) if you're not using it.

[Angelox]

Those makefile/ files I posted work for me, I just want to make sure it works for others too.
I think the bots are more intended for a 'play at home' version, alone or maybe a friend or two (even if I'm a 'bot addict' too. I'm trying to get a few more needed commands at least started (can always be changed for the better), such as pacify and evac. Things that when alone, I always think "Damn I wish I had that right now ..."

[hayward6]

Originally Posted by AndMetal

I think the logic is that, because not everyone is going to run bots (although they are quite awesome), it's better to use a preprocessor definition during compile than a rule, since it will decrease a lot of the required resources (memory/processor cycles) if you're not using it.

A good point. I feel this is one of the best things to happen to the project in a while! It's exactly what I wanted when I first learned about the emulator, but have no skills to work on it

[Andrew80k]

Originally Posted by AndMetal

I think the logic is that, because not everyone is going to run bots (although they are quite awesome), it's better to use a preprocessor definition during compile than a rule, since it will decrease a lot of the required resources (memory/processor cycles) if you're not using it.

Is this actually the case? I know the executable for the zone would be bigger, but I wouldn't think it would use much more resources unless you actually spawned BOTS. If you don't I would think that resource usage would be very similar to not having them compiled in. But that's a rough guess based on looking through the code a couple of times.

[Angelox]

Here's another problem I saw with the Bots;
When you have a good Bot group, mobs get real easy to kill - The game really gets unbalanced; lower end red-cons become easy to kill.
So, lets say you upped the exp to 2-3x , if you kill a Red, exp is like 10x! Even with normal exp, the reds are like 4x , which is ok if they are in a no-bot game (bigger challenge/ risk for a greater reward).
When I lowered the red con Exp to something more sane, then blues and lite blues gave little or no exp. If I had new players on the server that played alone (no bots) or just were starting out, I would get complaints about no exp.
So here's what I did, that may be an idea; Its a macro called 'CON_XP_SCALING', I never could get it to compile on windows with the macro, so I quoted it out and ran like that.
All the settings are changed in CON_XP_SCALING, and the original EQ settings in variables and rule_values are default.
The effect I got was fast exp on all cons, but nothing crazy for red/yellows. It may not seem to make any sense, but it works with whatever other scaling there is in the source and counters.

//#ifdef CON_XP_SCALING
		if (conlevel != 0xFF) {
			switch (conlevel)
			{
			case CON_GREEN:
				//Message(15,"This creature is trivial to you and offers no experience.");
				return;
			case CON_LIGHTBLUE:
					add_exp = add_exp * 175/100;
				break;
			case CON_BLUE:
				//if (lvldiff >= 12)
				//	add_exp = add_exp * 6/10;
				//else if (lvldiff > 5)
					add_exp = add_exp * 250/100;
				//else if (lvldiff > 3)
				//	add_exp = add_exp * 9/10;
				break;
			case CON_WHITE:
					add_exp = add_exp * 175/100;
				break;
			case CON_YELLOW:
					add_exp = add_exp * 200/100;
				break;
			case CON_RED:
					add_exp = add_exp * 225/100;
				break;
			}
			/*
			if (otherlevel >= 65)
			{
				int add = add_exp*((otherlevel-49)*20/100);
				add_exp += add_exp*((otherlevel-64))*2;
				add_exp += add;
			}
			else if (otherlevel >= 50)
			{
				add_exp += add_exp*((otherlevel-49)*20/100);
			}*/
		}
//#endif

Originally Posted by Andrew80k

Is this actually the case? I know the executable for the zone would be bigger, but I wouldn't think it would use much more resources unless you actually spawned BOTS. If you don't I would think that resource usage would be very similar to not having them compiled in. But that's a rough guess based on looking through the code a couple of times.

[Congdar]

Bots, like people.... when you have a full group, mobs are easier to kill. The group exp bonus applies and you get good exp. I don't think the bots make any difference in exp gains or mob difficulty than what you would get with a full group of actual players. The only differnce is, the bots go all out 100% where a player might conserve mana or the like.

If you can show in the code somewhere that exp gain is somehow increased due to bots that would be different than if you had actual players, or that somehow mobs become easier because of bots instead of players, then that would be the place to put in a fix.

[Angelox]

Well, for one , Bots don't make mistakes, PCs do -
When I first tried the bots, I got about 30 levels in a few hours. Might be just the way EQemu always has been, most people do not want low exp rates, so exp in general goes up much faster than it would as if I played on live.
When I played on live with full, experienced groups, killing a red was always a big challenge and many times proved disastrous. Not so with the bots.
All around, EQ Live is much harder than EqEmu. I don't like the way Exp drags on on live either. But I do like the game to last a little longer than a day or two; but that's just me.
I do have the fix in what I post at my site for many weeks now, and about a dozen people use it (it's that last 'fix' that still makes my version of the source custom).