| 12-29-2002, 06:48 PM | #1 |
Guest | I tested this map that i made, in all single/LAN/Bnet games. So far no sign of the infamous "wait" bug appearing when different number of users is in the game. Based on the numerous tests, i don't think the wait trigger is the issue when a map plays perfectly in singleplayer but sometimes causes annoyance in multiplay. HOWEVER, this test assumes that the internal diseases cloud time does not vary with # of player present and sticks pretty much to real time. Correct the triggers if u don't think the test is valid. Thx to those on bnet that tested out this silly map wit me heh :D Anyway, Try it out urself and i hope this settles whether there is a wait bug in multiplayer. |
| 12-29-2002, 11:55 PM | #2 |
One wait per trigger is allowed. The wait bug seems to exist if you do several waits in one trigger. I suggest that you test it that way. Anyways: It's good that someone puts some work into the elaboration of bugs and when they appear. :D Keep up providing us with any information you get, please. |
| 12-30-2002, 01:44 AM | #3 |
Guest | Anyway, after some preliminary new testing as suggested by SuperIKI, it seems the wait bug DOES exist. Tho i haven't pinned down the exact formula of how the wait cmd varies with # of player present (i suspect it probly will turn out to be the "X secs / # of players" theory, but I can confirm that the wait bug appears only when the # of active players is more than one users (active players do not include comps) and that it matters on the # users present in at the time of the waits are being executed (ie. if game started wit 2 users and wait cmds is executed. Then if a user leaves the game, the bug disappears) This would so far explain why in singleplayer, everything with waits are fine and why it gets wacky once in multiplayer. DL the new map /w the fixed waits. Still more testing/data needs to be done be4 exact root cause can be reveal...but it looks like the wait bug behaviours are like most of wc3 mappers predicted Stay tune... P.S. Thx SuperIKI for the suggestion, otherwise I would have lead y'all to the wrong conclusion. |
| 12-30-2002, 01:46 AM | #4 |
Guest | oops forgot to add the file....:D |
| 12-30-2002, 02:55 AM | #5 |
You can use a simple trigger to fix to problem. I did a test on bnet with a trigger that displays the amount of time 'waited' between things. I found that a 5 second wait time with 1 player = 5 seconds (duh) 2 players = 2.7 seconds 3 players = 1.9 seconds 4 players = 1.65 seconds And judging by this calculation I can give you the wait time for all players up to 12. x = (T/P) + 0.1P Basically it's time divided by players plus about 10% of the amount of players. I don't know how it works this way, but I guess it just does. But with this theory the rest ( up to 8 ) would be 5 players = 1.5 seconds 6 players = 1.43 seconds 7 players = 1.41 seconds 8 players = 1.45 seconds etc. A trigger for this would be fairly straight forward with some simple algebra, but this is what I have come up with. |