[Trigger] Triggering Myth - .25sec wait

[apocalypse_dude]

I hope this is the right forum, didnt know where to post else

Ok I repeatedly read that the minimum wait for

• Wait x seconds

is 0.25 seconds


Today i watched the Mythbusters busting some myths and so i feel inspirated to prove all those people wrong

Ok here you go !

• buster1
  • Events
    • Player - Player 1 (Red) types a chat message containing start as exact match
  • Conditions
  • Actions
    • Set bool = True
    • Trigger - Run runner <gen> (checking conditions)
    • Wait 1.00 seconds
    • Set bool = False

• buster2
  • Events
  • Conditions
  • Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions
    • Wait 0.01 seconds
    • Multiple FunctionsIf (All Conditions are True) then do (Then Actions) else do (Else Actions)
      • 'IF'-Conditions
        • bool Gleich (==) True
      • 'THEN'-Actions
        • Game - Display to Player 1 (Red), at offset (0.00, 0.00) the text: blabla
      • 'ELSE'-Actions

First trigger starts the second one, waits 1 second and then sets bool to false which will prevent trigger 2 from displaying messages
If you run this it will show 7 to 9 messages
Which basically means that
Wait time is bigger then 0.01 (or if then else takes a while to be executed)
Wait time is a lot smaller then 0.25!
Wait time is NOT! constant

Ok just to prove my theory a bit more
modify the wait time in trigger 1
Set wait to 0 => no messages
Set wait to 0.1 => one message

So you see there might be a minimum wait time but it is definetely a lot shorter then .25 and i think it might depend on your computers speed



Ok now i feel like Hyneman until someone proves me wrong (or tells me that this is already been posted )

 

[PurplePoot]

A far more accurate test, done with GUI since this is the GUI forum.

It displays between .12 and about .25, but it is far more often at .25 (2/3 of the time). Waits are also significantly slower on multiplayer. Also, the generally fastest accepted wait time is <.1, but that's for 0 or -1 second waits.

I'm not sure you fully understood the topic (Or that most do).

And this just proves further, not less, that waits suck.

• Timer
  • Events
    • Time - Elapsed game time is 5.00 seconds
  • Conditions
  • Actions
    • For each (Integer A) from 1 to 6, do (Actions)
      • Loop - Actions
        • Countdown Timer - Start Tim as a One-shot timer that will expire in 99999.00 seconds
        • Wait 0.01 seconds
        • Game - Display to (All players) the text: (String((Elapsed time for Tim)))

 

[emperor_d3st]

It actually has been proven long ago. Wait time is affected by processing time, as far as I know. I remember reading something like using timers instead of waits results in more accurate "wait times", but I have never tried it and I don't really care until I want to use these tiny wait times.

 

[PurplePoot]

Timers are accurate to .0035 or so seconds, but the fact that you need to use a callback function turns most people (Especially GUIers who thus can't use them for MUI) are turned away.

 

[apocalypse_dude]

Ok, its just that i read about this limit like 2 times a day minimum, hopefully some people read this thread and change their opinions

And yeah your method is way more accurate but it basically proves the same thing, waittime is not .25 secs
so... your Hyneman, and I'm Savage, can live with that


(and of course wait sucks, didnt want to question that)

 

[-Berz-]

Since we are talking about timers a i have a question:
What is better to use wait or use wait game time?
i know its a stupid question...

 

[emperor_d3st]

It's not a stupid question at all. I would also like to know because the stuff I googled up about this is very contradictive.

Most probably Polled wait (aka wait game time) waits through lag but the simple wait doesn't.
Polled wait uses a timer, but as far as I know it's as accurate as a simple wait, in addition it leaks 4 bytes every time it's used..

I have also found a function named polledwait2 by Vexorian:
http://wc3jass.com/viewtopic.php?t=206
I haven't tested it yet, dunno if it's any good, there are several reports that it caused the users' map to crash.

In my opinion the Wait is a more lightweight wait function probably intended for single player / campaign use, while Polled Wait is intended for multiplayer, but I have no idea why are they so inaccurate. Well, if you want to go for precision you have to use timers or periodic events.

 

[Alkonis]

Polled wait is nothing but a timer, really, but it leaks and it's slow. I think if you're really worried about the normal wait screwing up, you're much better off with a timer or PolledWait2, which makes Polled Wait faster and and removes the leak.

 

[PurplePoot]

PolledWait is slightly more accurate than TirggerSleepAction for some intervals, but still mostly sucks.

And Apocalype_dude, the thing you're missing is: waits are still about .25 66% of the time, and that is on single player where waits are rather fast.

 

[spiwn]

I found PolledWait to be more inaccurate(a lot) than TriggerSleepAction but I haven't put that to extensive tests.
I think timers have even less minimum interval, but it causes lag due to the number of executed functions per second.

 

[PurgeandFire]

I found PolledWait to be more inaccurate(a lot) than TriggerSleepAction but I haven't put that to extensive tests.
I think timers have even less minimum interval, but it causes lag due to the number of executed functions per second.

Timers are more accurate, and have a really low interval, yes, it has been verified.

But generally, if you loop and wait, or run a periodic timer and perform the same amount of executions, it will lag.

It isn't the timer's fault, it will generally lag if you do too many executions with a really low interval.