Update 1.24.42 is available

ForumNinja

Mac Creator 1.24.42 Fixes:

• Games that have tags with no name will now open correctly

Engine 1.24 Fixes:

• Fixed another issue with center of mass
• Saving and loading boolean attributes works again
• The random function works correctly again

PhilipCC

Slick fixes, thanks.

drizzling

Thank for the speed in fixing!!!

epicman

Great

V 1.24 is very sturdy and robust now, been playing with many of the features and it seems to be working.

Now we are ready for V 1.25

myke66

I'm not sure if this is a bug or not, but when selecting the round function, it brings up round(x, n) where as roundTo brings up roundTo(x). Shouldn't it be roundTo(x, n) and round(x) not the opposite?

This is on the latest Windows creator

drizzling

I experience a new problem with this version in displaying text/number from table cell value.

1) When press an actor that has a rule on table cell value being added/subtract, the number displaying the table cell value **sometimes stop altogether as if hang there... (not displaying the actual number of the table cell value). But the rule still works.. except that the text isn't able to display the correct table cell value (a integer in my case). My number display from the **Game attribute integer (Not Table cell value) has no issue and is displaying correctly and able to catch up with the rule in the same scene.

2) Only When change scene or reset the game which has an actor displaying the text showing the same table cell value as refer by the previous scene, the text then now display the correct table cell value. (This shows that nothing is wrong with the rule of changing the table cell value since it is still being calculated in the background though the text doesn't display the changed table cell value in the previous scene.

3) This issue happens randomly and unpredictable ... as when it hangs, it just doesn't display correctly anymore.. Only when reset the game/ change scene, the display of the table cell value then becomes corrected.

Hope Im being clear enough in explaining this issue... I suspect its a hang issue when being referred to the Table cell value... Please look into this. Thanks

drizzling

Oh ya.. the above issue is more prominent in adhoc test. In creator it is less likely to happen. It only happen once in creator.. but multiple times in my adhoc test.

wilsongalucho

Yes I also noticed this happening with my game.

The_Gamesalad_Guru

Send that info to GS

Armelline

Remember to report bugs using http://bugs.gamesalad.com and to include steps to reproduce the issue.

wilsongalucho

1.24.5861.17890 current windows version .. Just update it .. I have an attribute constrained to this expression and at some point it stops constraining the attribute

Constrain to floor(game.Clock.Second+(60 * game.Clock.Minute)+(60 * 60 game.Clock.Hour)+(24 * 60 * 60 game.Clock.Day)+(31 * 24 * 60 * 60* game.Clock.Month)+(12 * 31 * 24 * 60 * 60(game.Clock.Year-114))/10000000)

it also gets stuck if it's set to Change attribute every 1 sec

I've tested the the day, hour, minute, sec game.clock attribute and it keeps counting, but the attribute doesn't update, which causes everything in my game to get stuck.. I've submitted a bug report, just wondering if anyone else has had t this issue, it only happens on Android devices..

Hopscotch

@wilsongalucho , I think android returns the month as 0 for Jan, to 11 for Dec. Also, the year is less 1900. Check the values individually to confirm. This will make your calculation always be 0. Set a game.OSFlag and treat the constrain differently on iOS and Android to accommodate this.

wilsongalucho

Hi @Hopscotch thanks for the info, but it shows 19/1/2016 ... and the time stamp does count sometimes.. but sometimes it stops.. I think its bit specific to my game.. but it definitely has got something to do with the new update as it was working fine before.

I've tried to change it to Feb anyway. Before when I changed the time on the device, the timestamp would move forward.. now it doesn't doesn't, it seems like the rule stops responding.

I remember the timestamp was the very first issue I came across when I started developing my game, there were like 8 people here on the forums trying to solve the issue, and finally it was solved, but this is something to do with the rule being triggered nothing to do with the expression itself.

Somehow changing scenes triggers the rule again.. but at some point will stop..

Hopscotch

@wilsongalucho ok, now it sound like another bug I reported
http://bugs.gamesalad.com/show_bug.cgi?id=1509
Is it possible for you to delay the rule for a split second and see if it works?

wilsongalucho

Hello @Hopscotch I think I manage to fix it. I think I had a conflicting rule saving the timestamp, so far its working. Thank you for your tips!

Hopscotch

Good stuff @wilsongalucho

However, @GeorgeGS , I just added a new project to Bug http://bugs.gamesalad.com/show_bug.cgi?id=1509

This must lead to all kinds of issues.

Here a sample of the problem:

self.State = 0

Rule: if key=pressed AND self.State = 0 then 
         self.State = 1
EndRule

Rule: if self.State =/= 0 then
         Timer: after 0.5 seconds Run to completion
                  self.State = 0
         EndTimer
EndRule

If the timer ends and resets the attribute in the same draw frame as the key is pressed, then the attribute will get stuck in state 1. The second Rule will not react to the second state change.

This happens either by pressing the key in quick succession or just by holding it down.

ycan

Android performance is bad. Can be it fixed in 1.25 ?

wilsongalucho

Ok, now I'm sure something is definitely going on with Android and Windows creator and might be the error that @Hopscotch mentioned ...

This rule stops triggering

If Hero Stamina > 0
Training Time <= 60

Set Hero Stamina to (Hero Stamina-1000)

and my Timestamp stops counting on this sceene .. after a few secs ...

And the same rule on the previews scene keeps counting

wilsongalucho

@ycan what kind of performance issues are you referring to?

viking

I notice a lot more beachball slowdowns when I am using the Creator with this version compared to the last few versions - even when I quit and restart every 10 minutes.

I understand that new features added will slow down performance before they are solidified in the code base and optimized for memory leaks, but it would be nice to have this latest releases cleaned up before the release of 1.25, which will probably have a lot more issues initially.

MentalDonkeyGames

@viking said:
I notice a lot more beachball slowdowns when I am using the Creator with this version compared to the last few versions - even when I quit and restart every 10 minutes.

I understand that new features added will slow down performance before they are solidified in the code base and optimized for memory leaks, but it would be nice to have this latest releases cleaned up before the release of 1.25, which will probably have a lot more issues initially.

This happens a LOT when opening / exiting actors that have lots of rules. It sucks...

BlackCloakGS

can you file a bug with us and attach a sample project. So we can profile creator and fix the issue.

Armelline

@BlackCloakGS said:
can you file a bug with us and attach a sample project. So we can profile creator and fix the issue.

It happens with every project. Literally any project I open. It's not as bad as early-Yosemite days, but it's fairly bad and has been bad for a few versions now. I haven't noticed it being particularly worse in this version. Pretty much since the new numbering scheme there have been problems.

viking

@BlackCloakGS filed and attached a test project, but as @Armeline writes, it is not project specific - just an overall increase in memory leaks it seems.

Another thing that I have noticed is that the implementation of the new version of Box2D physics engine made my golf ball physics unpredictable. With 1.23.1 my ball would always take a hop forward when it landed and the spin back a little bit. And it would do so 100% predictably on every shot. Since the new physics version was added 90-98% of the balls land, jump forward and roll way past the landing spot and then 2-10% of the time the ball hop and spins back like it used to. I could live with the change in physics if it was consistent. I would just tweak the friction to make the ball spin back when it lands, but I can't do that now because the few times the ball physics react the way it used to the ball will spin back wildly and it completely breaks the illusion of golf.

I am guessing it has something to do with the custom collision and the change in how the engine deals with "fast moving actors". I sure hope this will be fixed though. We need to know that the physics engine in the game will give us predictable results or else players will get so frustrated that they will stop playing our games.

Agent Argyle

@Hopscotch wrote:

@wilsongalucho ok, now it sound like another bug I reported
http://bugs.gamesalad.com/show_bug.cgi?id=1509
Is it possible for you to delay the rule for a split second and see if it works?

@BlackCloakGS I believe my game is experiencing the problem Hopscotch reported.

viking

@BlackCloakGS the more complex actors also load a LOT slower than previously. As I am scrolling down the logic in the actor I sometimes have to wait for 5-10 seconds before the loading of the second part of the logic in the actor is complete. This used to be instantaneous.

68kStudios

@drizzling said:
I experience a new problem with this version in displaying text/number from table cell value.

1) When press an actor that has a rule on table cell value being added/subtract, the number displaying the table cell value **sometimes stop altogether as if hang there... (not displaying the actual number of the table cell value). But the rule still works.. except that the text isn't able to display the correct table cell value (a integer in my case). My number display from the **Game attribute integer (Not Table cell value) has no issue and is displaying correctly and able to catch up with the rule in the same scene.

2) Only When change scene or reset the game which has an actor displaying the text showing the same table cell value as refer by the previous scene, the text then now display the correct table cell value. (This shows that nothing is wrong with the rule of changing the table cell value since it is still being calculated in the background though the text doesn't display the changed table cell value in the previous scene.

3) This issue happens randomly and unpredictable ... as when it hangs, it just doesn't display correctly anymore.. Only when reset the game/ change scene, the display of the table cell value then becomes corrected.

Hope Im being clear enough in explaining this issue... I suspect its a hang issue when being referred to the Table cell value... Please look into this. Thanks

I have a lot of issues with table management too. When I write on a table or use a table value, sometimes it works, sometimes it doesn't...

AlchimiaStudios

I think I might be noticing the rule triggering issue with some recent logic too now. Seems to get stuck, but if i make a way to force the attribute to re-trigger it's state it starts working again.

One thing is for sure, my FPS performance has been cut by about 50% from 1.23.1 to the latest "stable" in some situations. Sent GS some vids and projects, surprised no one else has reported drops in the forums.

wilsongalucho

I came across a bug with my game ... basically a simple rule such as ... when press change attribute to attribute - 100 ... This rule would get stuck after a few presses... I managed to fix it by creating a new attribute ...

Clock

In mac, its very slow when i am working on a big project. Back and ford takes more than a sec. Even click on an actor, it takes more than a sec. please fix this. Previous version is much faster.

drizzling

@68kStudios i have submitted this bug and they found it an issue in 1.24 series and this bug is absent in 1.23 .

They mentioned that the bug seemed to resolve in 1.25 and so hope they will release 1.25 soon! Hang in there!

Maybe the table cell display issue could be just a lag.. as in it couldn't catch up with the performance in the ad hoc devices (Both in android and iOS) .. So if performance of 1.25 is better/faster .. this will resolve by itself