SolarStrike Software

If the bot is targeting a mob that is not visible:

• the bot will try for 30 sec to cast and after that time cancel that mob and try to reach the next waypoint => that looks very botlike
• if you get aggro from another mob while you are targeting a 'not visible' mob you will likely die

I would suggest:

• Instead of counting the time the bot is attacking a mob I would suggest to count the casts (without self buffs)
• if the HP of the mob dont change AND you are not in combat AND you are not the target of the mob (if you are not in group) => abort the attack after 2 or 3 casts
• if you get aggro and are in combat AND the HP of the mob don't change AND you are not the target of the mob (if you are not in group) => there is some other mob attacking you => abort the attack => clear target to be free for the attacker

If we count the cast we could also easily solve another problem:

• if the bot has choosen a target and get aggro from another mob before the bot starts to cast he will not stop and target the additional attacker. After casting that means two attackers and perhaps => the bot dies

So the bot should check directly BEFORE casting :

• is this the first cast on that mob AND are I allready in combat => abort the attack => clear target to be free for the attacker

I have allready make a modification in my bot for that points ... but that theme goes deep and needs changes on a lot of coding places. So it would be better to have a global solution within the standard bot. At least have the necessary variables (marker for skipping a target, marker for skip looting) /counters (for the casts).

Turn down your MAX_FIGHT_TIME. The name is misleading because it's purpose was changed long ago. It actually has more to do with how long before you damage the enemy. The counter resets after each successfull hit. You might be onto something with checking the combat flag, though. I'll have to think of how to best implement this.

Administrator wrote:Turn down your MAX_FIGHT_TIME. The name is misleading (...)

yeah i did some testing with this a while back and found 9 seconds to be the best value for my S/R. this is roughly a 3 skill cast timeframe and i figured this doesn't look "botish" at all.