Problem Description

If I test a timer (which is not set) against 0, the condition is true. If I do the test in an or-condition the result is still true, but... if I check a "platform is on floor" afterwards, it returns false, even if its true.

Edit: I attached a second file with two more cases.

Attach a Capx

https://gamumo.com/c2bug/conditionbug.capx

Edit: https://gamumo.com/c2bug/conditionbug2.capx <-- 2 more cases

Description of Capx

you can press keys a and s to check both cases. in both cases you see the monster on the floor. but when you press s, it wont jump. when you press a it jumps, although both keys have the "same" code.

Steps to Reproduce Bug

• open the capx and run the layout
• press a to see the monster jump (expected)
• press s to and see how it doesnt jump (unexpected)

Observed Result

The monster is on the floor and the condition returns false.

Expected Result

the condition should return true.

Affected Browsers

• Chrome: (YES)

Operating System and Service Pack

Windows 7

Construct 2 Version ID

233 64bit

Seems as if it doesnt work...? So, again... Ashley, could you have a look? .

hey, thanks, but I read these manuals and they don't give further informations.

In the first link, its said: "Preload sounds

Whether to download and decode sounds before the game can start. If Yes, then sounds will be downloaded while the loading bar is showing. [...]."

So option 1: There is still my question why it is loading BEFORE the loader layout you can set yourself. Feature request: Can we configure it to be loading WITH the images?

option 2: Please no. Feature request: A preload all sounds function.

option 3: that's a bad way to preload sounds .

I'm developing for computer/ browsergame.

another question: is this the right place for feature requests? I did not found another thread. cause even I dont no if its a "bug" that sounds and images are loaded separately. In this case I would post it there.

Oh how friendly... That is not very helpfull and constructive. Indeed for me it is the best way to preload them. Good that its better for you to do it in another way. I did read the manual and there is nothing said about what I mentionend - as far as I know.

Perhaps you should really read my question? I'm not saying that there shouldn't be a loading bar! I say, that Construct 2 gives you the settings for a loader layout, to load data (images) there, and now sounds are loading before the loader layout... so that there are 2 loading layouts following each other (first music, than images) before the game starts. Please show me where this is explained in the manual, if I didn't get it.

But correct me if I'm wrong. In the manual/tutorials is only explained how to use the preload option in the project settings and that you can preload sounds individually. Did I miss something? Is there another way to handle that?

Hi,

I want to preload my sounds in my game but I can't understand why they are loaded before the loader layout? Isn't it ment to be there to preload data and optimize the waiting time?

For now when I activate it, the players see the "Loader style: percentage text" for a minute and afterwards they come to my loader layout where they have to last again for a minute to see the next selfmade percentage text, which in my opinion is very confusing and "ugly".

In another topic people already asked for a feature to preload all sounds. There wasn't an answer so I'm trying to request this feature again . So, I have already over 150 sounds and I realy dont want to preload them all one by one.

Finally: Is it a bug or is it ment to be preloading sounds before the loader layout? Can we get a preload-all-sounds method for the loader layout? Please?

Thanks!

Is there a reason, why this isn't working with gamepad buttons, too? For me this counts as a "user-initiated event".

Hi again,

I expect it to be an ajax or cross domain problem (but I'm not sure), but I avoid the httpsession now and so I don't need any further help.

Thx and sorry.

Hi,

I'm not sure if I'm facing these problems due to construct 2 or my webapp, but perhaps you have an idea or hint.

I tryed setting a session attribute in a servlet/filter context in my webapp. this works fine, but when I do a second request to my webapp via my construct 2 game (via ajax post) it creates a new session. It works when I use my rest services directly in the browser (enter the url, do the (get) request myself). The session wont be created new - everything's fine. Of course I'm using the same url as in the ajax call from construct 2, same browser,...

Is there anything I need to know about Construct 2, Cookiehandling, so on... ? Never had any problems with sessionobjects...

Thanks for any hints!

It is working now and I guess it was something like a nullvalue problem. after a restard, the entry for "fade in" in the object-types-folder has been removed. it was set to 0 before. after having set 0 again, it worked fine. normaly its not possible to save it without an entry... dont know why this was removed.

Perhaps that was the problem.

Okay, now things get strange.

Made a new testproject for you, where everything works fine.

But my mainproject worked fine with the old fade behaviour. just changed the behavior to fade adv and it wont work anymore...

Going to test again, but I've got no idea...

Thanks a lot! I'm realy glad, that you share your work with us... but for me it wont work.

fade out works fine, but "wait time" and "fade in time" uses 0, no matter what I set (via initialisation or via event). I use release 184 (64bit), tested on two computers via chrome and via firefox. I'm sure you tested everything... any idea?

Edit: "fade out time" doesn't work when set in initialization; only works via event

Thats not possible, cause the function with the repeat can/will be called multiple times simultaneously. The only way I can think of, would be, to create a variable dynamicly dependig on an given id in a dictionary and refer to it later on. not very beautiful, but I guess it will fit.

but construct 2 saves the position in the event and the used objects when leaving the event cause of a wait... wouldn't it be possible to even save the repeats and their indices? sounds like a featurerequest.