Previously we discussed trigger delaying problem http://www.staredit.net/topic/13650/#290538 and made conclusion, that the reason of this trigger delay is wait blocks.

But now I received next problem (in another map):

Old map behavior: countdown timer blocking is happen only with 5% chance (so most of games are without timer blocking), and blocking time are 1 minute long. This is good behavior for me.

I updated this map (added a lot of non-wait triggers, and added/removed some wait triggers) - and received next new map behavior: countdown timer blocking is happen with 100% chance (so all games are with timer blocking), and blocking time are 5 minutes long. This is bad behavior for me.

Goal: to make "Old map behavior" in new map (of course - ideal goal is to make this trigger blocking happen with 0% chance and 0 minues long, but this is much harder and not related to current topic).

Questions are next:

1) How to reduce trigger delaying time? By reducing "wait time" in triggers, by reducing amount of triggers with "wait" or by reducing amount of triggers without "wait" ?

2) Which ways I can make "Old map behavior" in new map ?

You can say "just use death counters instead of waits and forgot about these problems", but here are the reason to use waits instead of death counters in my map:

Crazy CPU is the map, which should show me something new every time I play it. This map contain triggers, which say "order CPU units patrol to player's base". These triggers contain wait actions.

So, as wait blocks has "random" and "unknown in some cases" behavior, I received next: sometimes cpu attack player's base often, sometimes cpu wait a lot of time and only then attack the player, also it choose attack target randomly - and this is various from game to game (and this is only one of examples of random cpu behavior).

So if I will switch to death counters - I will lose this random cpu behavior (so it will not be fun for me).

And the problem is, that new updated map just increased chance and time of this http://www.staredit.net/topic/13650/#290538 problem.

In other words - there are 2 different wait block behavior:

1) wait blocks, which delay only triggers with wait actions
2) wait blocks, which delay ALL triggers

I need 1, but I dont need 2

I attached old and new maps to to current post.

Attachments:

Post has been edited 3 time(s), last time on Aug 20 2012, 10:39 am by theleo_ua.

Do not use waits dude. They end up adding together which can make your wait time unpredicatable depending on your triggers conditions.

Bottom line use death counts, theyre extremely versatile. If you're having trouble understanding the concept of Death counts i suggest reading through this

Take the wait out of time sensitive triggers.

random is more easily achievable with death counters and a randomized switch.

Yeah it's recommended that you replace all waits with death counters. Otherwise it's hard to predict what will actually happen. Fine tuning a wait block to do exactly what you want is pretty tedious and it's also bad style plus there's probably nobody that would be willing to help you with that.

So when you replaced all waits with death counters just use switch randomization to make the computer more unpredictable. If you need more help on the topic be sure to check our stickied old wiki post and look for switches. If you're still having trouble feel free to ask here.

Also note, that triggers that do not have any waits in them will not get blocked by wait blocks. If you seem to experience that behavior there must be another reason.

Ok - i'll try this


That was really suddenly for me, but Crazy CPU map has wait action in "refresh timer" trigger. I will remove this wait and will check, how it will going.

But: Castle Fight map has no any waits in "refresh timer" triggers, but this 5% chance of all trigger blocking http://www.staredit.net/topic/13650/#290538 happens even in castle fight. I bet, that the reason of happening this in castle fight is "owner = all players". I will try to make owner of refresh timer triggers "only computer without hyper triggers", and will see how it will going.

Thanks anyway

Post has been edited 2 time(s), last time on Aug 20 2012, 12:05 pm by theleo_ua.