So - after the pre-HoTS patch my arcade map I am developing has been riddled with new minor bugs. However, one show stopper for me is the triggers that change the state of water.
In my map I use lava inside a multi-tiered coliseum. The levels of the lava raise dependent upon progression and bosses. Since the patch, this transition has been EXTREMELY laggy. We're talking 3 FPS type lag, and though I have not yet logged system performance during transition periods to pinpoint whether this is a performance issue or game issue I know that on my powerhouse-tower that has amazing hardware and performance potential; I experience the same lag as well as reduction of performance as compared to the transition lag on my laptop which has at least 50% less performance power (rough estimate-comparison). Has anyone experienced the same or similar issue since patch? Any resolutions?
As well - the triggers seem to be the same. I was worried maybe the editor was reacting to now-legacy scripts, and perhaps I would need to rewrite the trigger in newer trigger actions format. However, the water state change action seems to be exactly the same as before. I am not sure what is causing the issue.
Here is the trigger I use to RAISE and LOWER the water states. Keep in mind, both transition suffer the same effect until the state transition duration expires. Have a look:
WLavaLvlUp
Events
Local Variables
Conditions
Actions
Sound - Play TTosh01Adjutant00013 for (All players) (at 100.0% volume, skip the first 0.0 seconds)
Unit - Order Door[7] to ( Close Gate) (Replace Existing Orders)
Unit - Order Door[8] to ( Close Gate) (Replace Existing Orders)
Unit - Order Door[9] to ( Close Gate) (Replace Existing Orders)
Unit - Order Door[10] to ( Close Gate) (Replace Existing Orders)
General - Wait 1.0 Game Time seconds
Sound - Play Emitter_Lava for (All players) (at 35.0% volume, skip the first 0.0 seconds)
Environment - Change water to Char, Alternate Setting over 8.0 seconds, using a Exponential blend curve
Environment - Set lighting to Lava, blending over 8.0 seconds
General - Wait 4.0 Real Time seconds
Variable - Set WLavaDeath? = True
General - Wait 4.0 Real Time seconds
Trigger - Turn WLavaLvlDn On
Trigger - Turn WLavaLvlUp Off
Also, there are 2 types of lava used side/by/side so that one half of the coliseum can be off limits and force players to play on the other half where the lava state is default. So if you could imagine; there are 2 triggers of the above that designate either Char or Redstone lava states to be adjusted to the defined height. I would also link the trigger for my the lower states but the symptom is problematic with both in equal fashion, so if the correction is discovered with the above trigger, then the same solution could be applied to the lowering triggers.
How do you call the lava state change triggers? I noticed you have no event, which is fine, but you also turn the triggers on and off... which has no effect if you are calling the triggers via an action.
I have a setup trigger when arcode mode is selected. An action to run the trigger and wait until finished is used. Your point is valid, the on/off actions are relics feom function testing that i have since deleted.
Are you running any other threads while the lava is rising/falling? Do you have events firing rapidly?
Lag like this is typically called by too many events being fired at once. Another source would be too many sound files being played in rapid succession.
For the most part, its pretty much the only thing happening at the moment the state change is executed. For now I changed the transition to 1 sec duration. It has a huge 2s hiccup and then completes.
Before the 2.0.4 and 2.0.5 patch this was not a problem; transitions did not create any performance drain. The difference between patches is extreme.
If anyone has the time to experiment; it'd be interesting to see if the same problems occurred for a quick map you makeup to test this, but I wouldn't expect anyone to spend that much time trying to figure this out for me heh heh. If only water state changes were more popular in games maybe more people would be looking into this haha.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
Hello All,
So - after the pre-HoTS patch my arcade map I am developing has been riddled with new minor bugs. However, one show stopper for me is the triggers that change the state of water.
In my map I use lava inside a multi-tiered coliseum. The levels of the lava raise dependent upon progression and bosses. Since the patch, this transition has been EXTREMELY laggy. We're talking 3 FPS type lag, and though I have not yet logged system performance during transition periods to pinpoint whether this is a performance issue or game issue I know that on my powerhouse-tower that has amazing hardware and performance potential; I experience the same lag as well as reduction of performance as compared to the transition lag on my laptop which has at least 50% less performance power (rough estimate-comparison). Has anyone experienced the same or similar issue since patch? Any resolutions?
As well - the triggers seem to be the same. I was worried maybe the editor was reacting to now-legacy scripts, and perhaps I would need to rewrite the trigger in newer trigger actions format. However, the water state change action seems to be exactly the same as before. I am not sure what is causing the issue.
Here is the trigger I use to RAISE and LOWER the water states. Keep in mind, both transition suffer the same effect until the state transition duration expires. Have a look:
WLavaLvlUp
Events
Local Variables
Conditions
Actions
Sound - Play TTosh01Adjutant00013 for (All players) (at 100.0% volume, skip the first 0.0 seconds)
Unit - Order Door[7] to ( Close Gate) (Replace Existing Orders)
Unit - Order Door[8] to ( Close Gate) (Replace Existing Orders)
Unit - Order Door[9] to ( Close Gate) (Replace Existing Orders)
Unit - Order Door[10] to ( Close Gate) (Replace Existing Orders)
General - Wait 1.0 Game Time seconds
Sound - Play Emitter_Lava for (All players) (at 35.0% volume, skip the first 0.0 seconds)
Environment - Change water to Char, Alternate Setting over 8.0 seconds, using a Exponential blend curve
Environment - Set lighting to Lava, blending over 8.0 seconds
General - Wait 4.0 Real Time seconds
Variable - Set WLavaDeath? = True
General - Wait 4.0 Real Time seconds
Trigger - Turn WLavaLvlDn On
Trigger - Turn WLavaLvlUp Off
Also, there are 2 types of lava used side/by/side so that one half of the coliseum can be off limits and force players to play on the other half where the lava state is default. So if you could imagine; there are 2 triggers of the above that designate either Char or Redstone lava states to be adjusted to the defined height. I would also link the trigger for my the lower states but the symptom is problematic with both in equal fashion, so if the correction is discovered with the above trigger, then the same solution could be applied to the lowering triggers.
Thanks in advance for your time/responses.
How do you call the lava state change triggers? I noticed you have no event, which is fine, but you also turn the triggers on and off... which has no effect if you are calling the triggers via an action.
I have a setup trigger when arcode mode is selected. An action to run the trigger and wait until finished is used. Your point is valid, the on/off actions are relics feom function testing that i have since deleted.
@Gaelidan: Go
Are you running any other threads while the lava is rising/falling? Do you have events firing rapidly?
Lag like this is typically called by too many events being fired at once. Another source would be too many sound files being played in rapid succession.
@BasharTeg: Go
For the most part, its pretty much the only thing happening at the moment the state change is executed. For now I changed the transition to 1 sec duration. It has a huge 2s hiccup and then completes.
Before the 2.0.4 and 2.0.5 patch this was not a problem; transitions did not create any performance drain. The difference between patches is extreme.
If anyone has the time to experiment; it'd be interesting to see if the same problems occurred for a quick map you makeup to test this, but I wouldn't expect anyone to spend that much time trying to figure this out for me heh heh. If only water state changes were more popular in games maybe more people would be looking into this haha.