• Hello [name]! Thanks for joining the GMC. Before making any posts in the Tech Support forum, can we suggest you read the forum rules? These are simple guidelines that we ask you to follow so that you can get the best help possible for your issue.

OFFICIAL GMS2 Version 2.1.4

Status
Not open for further replies.

DukeSoft

Member
Use a backup strategy outside the IDE.
Well you should be making your own backups in any case.
Yeah, I mean its not like its GMS's fault :/

Do you keep backups of all your browser settings in case your browser randomly deletes them? Do you keep backups of your spotify playlists in case they get deleted?

Not saying you shouldn't have backups because you should, but its still pretty big f**kup if the IDE decides to randomly delete files.
 
N

Nicoloscopie

Guest
I've seen a couple of people recommending making a new folder with a different name, then copying the contents of your old project folder into that and trying to load things. Not sure if I have that right or if it will help in this situation.
I tried doesn't work, but thanks anyway, i've filled an other bug report with more info and screenshot, with hope of a response soon...

Also, if I were you I would be very angry if I lost something. I would probably ask moneyback and change engine for other one.
Thank's for the advise, i let them a couple of day's just in case but you'r right i paid 40 dollars to work on a project for two week's and lost nearly everything, i am in fact pretty angry about that so i'll ask for a refund and try something else.
 

hogwater

Member
Yeah, I mean its not like its GMS's fault :/

Do you keep backups of all your browser settings in case your browser randomly deletes them? Do you keep backups of your spotify playlists in case they get deleted?

Not saying you shouldn't have backups because you should, but its still pretty big f**kup if the IDE decides to randomly delete files.
Well, yes. But I think that learning to accept what is, rather than what should be and tailoring your approach to the situation is a very valuable lesson.

Also, if you give up at the drop of a hat because you've lost "a week's" worth of work, no game will ever be created. How many hours were in that week, btw? The IDE tracks your time spent per project.
 

DukeSoft

Member
Well, yes. But I think that learning to accept what is, rather than what should be and tailoring your approach to the situation is a very valuable lesson.

Also, if you give up at the drop of a hat because you've lost "a week's" worth of work, no game will ever be created. How many hours were in that week, btw? The IDE tracks your time spent per project.
When I buy a car I don't assume it explodes on the first 2 miles - while sometimes it does. Doesn't mean I should live like that everytime :p

On the other hand, if it explodes every 2 miles I will probably get some sort of backup car for once it hits that 2 mile again :p
 
G

Guest

Guest
Hello,

I've been teaching a class using Game Maker Studio 2 Trial Edition. This past week all of my 30 computers stopped working unless we downloaded a new update. We did this. Immediately afterwards, the Run (F5) function stopped working for about half the class.
The students can still work in the editor, but when they save and then Run, it goes through its usual line by line process and then stops. The game doesn't crash or abort. It just pretends like we never asked it to Run at all. The last lines in the window are :

C:\Windows\system32\cmd.exe DONE (0)
DoSteam
Igor complete
[Run] Run Game
Options: Z:\Mario want_1742FEE_CD36E629\MainOptions.json
C:\ProgramData/GameMakerStudio2/Cache/runtimes\runtime-2.1.4.200/windows/Runner.exe -game
Attempting to set gamepadcount to 12
DirectX11: Using hardware device

These last two lines are the strangest because we aren't using gamepads or hardware at all that I'm aware of.

Any help would be GREATLY appreciated as this class is now just sitting here waiting for the solution. I have submitted a support ticket, but its been a couple of days and I haven't heard anything back.

Dave
Roll back the runtime. See here for instructions: https://www.reddit.com/r/gamemaker/comments/89w816/if_your_collisions_are_broken_in_gms2s_newest/
 
N

Nicoloscopie

Guest
Also, if you give up at the drop of a hat because you've lost "a week's" worth of work, no game will ever be created. How many hours were in that week, btw? The IDE tracks your time spent per project.
The time spend working on my game is not the issue here, it could be one hour it could be 200 hours, the fact that i paid money for a product that doesn't work has entended make me angry, the fact that i worked for nothing make me angry. I get why you want to defend YYG but for this situation i'm not responsable for the lost of my work. I'm not gonna stop making video game just because i've lost work or i'm mad at the moment. There is a lot of other Game engine i'll have to try like unity or unreal but Gamemaker prone an easy understanding that's why i choose it first.

Anyway no need to fall in the Internet rage you can usely find on forum, so thank's for your help "Hogwater", i'm not giving up on making game but if Gamemeker can't help me with my problem for something that i spend money on i'm not gonna wait and loose more money on it ^^
 

hogwater

Member
I'm not defending YYG, obviously it sucks that you lost work and the IDE shouldn't be deleting anything.

Also, of course it matters whether it was 1 hour or 200 hours lost.
 
G

Guest

Guest
I'm not defending YYG, obviously it sucks that you lost work and the IDE shouldn't be deleting anything.

Also, of course it matters whether it was 1 hour or 200 hours lost.
No it doesn't. If GMS2 is going to lose work, it's unusable, whether it's in fact lost 1 hour or 200 hours.
 

Ricardo

Member
@YellowAfterlife did something change in HTML5 scaling? My old, good and always working HTML5 scaling script is now giving me offset coordinates. It seems like the GUI size is not following the scaling. I haven't properly isolated the issue yet due to lack of time, that's why I'm asking for some advice that could ease my debugging tomorrow.
 
Last edited:

YellowAfterlife

ᴏɴʟɪɴᴇ ᴍᴜʟᴛɪᴘʟᴀʏᴇʀ
Forum Staff
Moderator
@YellowAfterlife something did change in HTML5 scaling? My old, good and always working HTML5 scaling script is now giving me offset coordinates. It seems like the GUI size is not following the scaling. I haven't properly isolated the issue yet due to lack of time, that's why I'm asking for some advice that could ease my debugging tomorrow.
Depends on whether you readjusted mouse coordinates yourself or not, I think the rough list would be around
  • A fix for a bug where scaling canvas wouldn't auto-adjust coordinates
    I think this was about having port size not match view size.
  • A fix for a bug where readjusting DPI by mismatching canvas.width/canvas.style.width wouldn't auto-adjust coordinates.
I'm not involved with fixes right now so I'd suggest to look around the bugtracker
 

clee2005

Member
@YellowAfterlife something did change in HTML5 scaling? My old, good and always working HTML5 scaling script is now giving me offset coordinates. It seems like the GUI size is not following the scaling. I haven't properly isolated the issue yet due to lack of time, that's why I'm asking for some advice that could ease my debugging tomorrow.
For me it was the window_set_position(x,y) which was positioning the window within the browser that stopped working. 2.1.3.189 was allowing me to absolutely position the window in the browser. Now in 2.1.4.200+ it is centering the window in the browser and not paying any attention to the window_set_position() method.
 

Ricardo

Member
Depends on whether you readjusted mouse coordinates yourself or not, I think the rough list would be around
  • A fix for a bug where scaling canvas wouldn't auto-adjust coordinates
    I think this was about having port size not match view size.
  • A fix for a bug where readjusting DPI by mismatching canvas.width/canvas.style.width wouldn't auto-adjust coordinates.
I'm not involved with fixes right now so I'd suggest to look around the bugtracker
Thanks! I'm not sure if I am ignoring something very obvious or if a bug was introduced in the latest runtime (since I'm been using that HTML5 scaling code for ages without issues). Anyway, I created a very small example project to show what I'm talking about.
To see the problem, all needed is resize down the browser window and notice that the sprite in the screen doesn't follow the mouse cursor coordinates correctly. I'm sure this isn't about mouse_x/mouse_y itself, since the offset problem also affect clicks events.
@YellowAfterlife I'll be awesome if you could take a look at it so I can know if there's a bug or if there is something I could do to fix it. It's totally fine if you don't have the time, though. Let me know and I can send a ticket instead.
 

Smiter

Member
Regarding crash on startup, I looked in to my yoyo account. The login sessions reported the wrong ip address, and the location in Dallas, Taxas. I, however, am in Minnesota. Not sure if its related, but it can't be good.
 

sman

Member
Using show_message() for debugging in Android now produces "this function has been deprecated." The documentation only said "this is for debug use only," and I used it that way in many places to alert me to a problem that occurred without crashing the game. Finding that those checks now result in an error and must be rewritten is not a nice surprise.
Especially if show_message was just before a crash to alert and explain the crash. Now, with show_message_async, I suppose the game crashes without showing the message :(
 
I just updated it and it doesn't work at all anymore. I'm on Windows 10. I get the usual splash screen: It's loading Languages... Loading Custom Languages... User Login... Loading Preferences... Checking RSS... Installing Runtime 2.1.4.203 and it instantly closes.
I hope this is being taken seriously.
I attached an image the instant before it shuts down.
 

Attachments

Thanks! I'm not sure if I am ignoring something very obvious or if a bug was introduced in the latest runtime (since I'm been using that HTML5 scaling code for ages without issues). Anyway, I created a very small example project to show what I'm talking about.
To see the problem, all needed is resize down the browser window and notice that the sprite in the screen doesn't follow the mouse cursor coordinates correctly. I'm sure this isn't about mouse_x/mouse_y itself, since the offset problem also affect clicks events.
@YellowAfterlife I'll be awesome if you could take a look at it so I can know if there's a bug or if there is something I could do to fix it. It's totally fine if you don't have the time, though. Let me know and I can send a ticket instead.
Something definitely changed in the update, HTML5 co-ordinates were not adjusting accurately when the canvas was scaled, and on a high DPI device (iPhone 7 in this case), the co-ordinates were incorrect but definitely scaled by the DPI. On top of that the resolution of graphics were still terrible. I'm sticking with 2.1.3.189 and my custom approach for now until there's a fix.

This is including the runtime 2.1.4.200, 2.1.4.202 and 2.1.4.203.

I see that a previous bug regarding this issue was re-opened today: https://bugs.yoyogames.com/view.php?id=28596
 

Ricardo

Member
Something definitely changed in the update, HTML5 co-ordinates were not adjusting accurately when the canvas was scaled, and on a high DPI device (iPhone 7 in this case), the co-ordinates were incorrect but definitely scaled by the DPI. On top of that the resolution of graphics were still terrible. I'm sticking with 2.1.3.189 and my custom approach for now until there's a fix.

This is including the runtime 2.1.4.200, 2.1.4.202 and 2.1.4.203.

I see that a previous bug regarding this issue was re-opened today: https://bugs.yoyogames.com/view.php?id=28596
Nice! I was about to send a ticket about the issue, now I think I'll just wait. I'm feeling very frustrated because the latest update (2.1.4.200 to 2.1.4.203) has a lot of interesting HTML5 fixes but I can't re-build my project with the scaling broken. Scale is an essential feature in HTML5 projects... I really hope they release a hotfix for this so we don't have to wait 3 or more months for a new version. We'll see.
 
Nice! I was about to send a ticket about the issue, now I think I'll just wait. I'm feeling very frustrated because the latest update (2.1.4.200 to 2.1.4.203) has a lot of interesting HTML5 fixes but I can't re-build my project with the scaling broken. Scale is an essential feature in HTML5 projects... I really hope they release a hotfix for this so we don't have to wait 3 or more months for a new version. We'll see.
This has been the case for so long, I think we could all benefit from the HTML5 fixes in 2.1.4 but if scaling doesn't work, it honestly makes the whole module pointless.

Looking at the release notes of the runtime, there is a pattern of at least 2 updates a month in some cases, but I agree with you and I also hope we don't land up waiting months for a fix.
 

Smiter

Member
I just updated it and it doesn't work at all anymore. I'm on Windows 10. I get the usual splash screen: It's loading Languages... Loading Custom Languages... User Login... Loading Preferences... Checking RSS... Installing Runtime 2.1.4.203 and it instantly closes.
I hope this is being taken seriously.
I attached an image the instant before it shuts down.
Same here. I already posted and filed a bug report, hopefully they respond soon. I'm in the middle of a big project, and I don't really want to switch engine.
 

Smiter

Member
So I finally figured it out. GMS was failing to access the runtime folder in order to install it. So simply run the "gms.exe" as administrator resolved that.
Hopefully this can resolve other's problem as well. Now my project is only behind one week:)
 

Japster

Member
In this update, I notice a distortion of the sound. Especially audible at high frequencies and on the Android device. In the sounds (especially at high frequencies) there was hoarseness.
This problem was described in the latest update of 1.4 >>>
Now I hear it here.
@Andrey - seconded - I have a jukebox in my game, works flawlessly in 2.1.3.x, but in 2.1.4.x, one of my tracks cuts out after a short while, and restarts - literally after about 30-40 seconds, when it's a 3-4 min track. The other tracks seem unaffected.... Re-imported, rebuilt, etc, no joy. Even reverted my runtime, and it still does it. Not sure what's going on....

EDIT:- Actually, I found a way to correct it...

I selected another method of sound (it was set to Compressed - Streamed) (correct for audio tracks apparently) - Changed it to "Compressed - NOT Streamed", seemed to work fine after that. On a hunch, as it had forced a conversion, I tried the original setting again. It's now working fine, and playing through to the end again! - Weird, but all's well now...

Might be helpful for any other users with audio track issues after moving to this update....
 

Japster

Member
Just to add....

I can confirm that my game is now crashing like hell - every single time I run it from the IDE, even not doing anything other than sitting at the title screen, or playing a few turns on the game itself. Sometimes it crashes the moment it's built and is launched...

Yes, I've reverted to 2.1.3.x runtimes, and no joy either.... I'll produce an EXE tomorrow, exit the IDE, and run it, to see if that is a contributing factor, but atm I'm knackered and going to bed....

This is REALLY frustrating, guys....

Sod it... I've just created an executable, closed GMS2 completely, and the same thing's happening (albeit took a lot longer to crash - I was getting hopeful at one point, but noooo)... Am I right in thinking that by using the 2.1.3 runtime, I should not be affected AT ALL by having to upgrade the IDE and Runtime to 2.1.4? I'm stumped atm.....
 
Last edited:
B

Brian C

Guest
I keep getting the following error, see pic... how can i get past this error with installing the runtime?


GameMaker 2 Error Pic.png
 
So I finally figured it out. GMS was failing to access the runtime folder in order to install it. So simply run the "gms.exe" as administrator resolved that.
Hopefully this can resolve other's problem as well. Now my project is only behind one week:)
Tank you so much! This solved it!
 

LEGOlars

Member
The new IDE is incredibly slow when having a very large project opened :( I have a 5 second delay after every mouse click or any keyboard shortcut or any action at all. Is there any way I can reset to the old IDE until that's fixed?

all the info I have:
I tested it on multiple computers. In the task manager it doesn't seem to be using much CPU or memory. Also uninstalled and installed multiple times. I used to have a similar problem in GMS1.4 when I opened the project, but it disappeared after launching the game once. When porting my game to GMS2 I never saw the issue again up until this update.
 
Last edited:

Japster

Member
@rmanthorp - Well, after a shedload of testing, pulling my hair out etc, I can say with some confidence, that I'm sure that this release seems to be screwed in some way....

I had luckily compiled and built a Windows executable of my 0.99 project snapshot back in March, so proved a perfect comparison - and I can run this without crashes, for 20 minutes at least during my test, back and forth between the menu and game, playing a game to completion, choosing human or CPU players, etc, etc...

I take that EXACT SAME project snapshot that this build was created from - it pulls it into the IDE, SAVES(?) it immediately on loading (possibly with changes to file structure etc, which might be causing an issue, I dunno), and it's totally screwed - I can't get in or out of the game screen more than 2 or 3 times, even after exiting seconds afterwards (as in my testing with the earlier-created executable...), and it crashes out CONSTANTLY, with no GMS error message, just a standard Windows error. I compile and build a fresh EXE of this exact same snapshot, and it crashes just as badly as in the IDE. I NEVER got this with the previous, superb release of 2.1.3... Either in IDE, or building it.

Oh, and it seems to do it randomly - even when I simply test it, sometimes the moment it runs even? - If I remove my "GameController" (basically the game and all it initiates) object, it doesn't crash at all (basically because I'm doing nothing, I guess (no grids, no tiles, etc, but it does show that SOMETHING in my game is causing it to crash NOW, when it didn't with the previously-built executable built with the exact same snapshot, and is probably therefore worth investigating by YoYo?). However, this object isn't even loaded or live until the game screen is chosen, and the game sometimes crashes instantly, ie. on the title screen, which has no instance of the Game Controller object even created yet?

Sometimes, when choosing Debug mode, it will even seem to run almost flawlessly, as long as I have a breakpoint in there somewhere, pushing it along every time it stops on it, although it will still eventually crash, so perhaps it's related to alarms, timing etc, I'm not sure at this point... Everything I *thought* it might be, it wasn't....

I am really, REALLY hacked off here with this mandatory update - I just want my earlier build back, that simply WORKED. Does anyone know how to get this running again? Development has stopped completely, and I've lost 3 days just trying to find out if it's down to me and something I was maybe getting away with before, to now feeling certain it's a newly-introduced instability - I was *almost* resorting to hacked earlier releases I'm seeing online, which is a ridiculous state of affairs when I've legitimately paid for GMS2?! - I was happily developing in what I thought was a superb, stable engine, and now this...

Seriously guys, there is something WRONG with this release that's been introduced along with it, I'm certain of it. I tested everything I could think of, disabled/removed huge amounts of events and code trying to find the problem, and YES, I also tried different runtimes (I could only revert to my favourite runtime, which should have worked by all means, but couldn't try earlier ones, as my parent/child object relationships didn't seem to be recognised by the 2.1.2.x runtime even - I just got build/compile errors), so yeah, I also checked for memory leaks etc, but nothing is showing - 20mb usage, doesn't deviate up or down by more that 100kb, and no huge drops in frame rates etc - I double-checked my ds_grid destroys and resizes as pertinent, the game now crashes even at stupid/random points, ie. even when a move has been played, the score is being added, and is zooming back into the screen. Sometimes crashing going INTO the game, sometimes when coming OUT, sometimes when simply moving a piece around the board (or not)....

EDIT - I just removed even more code, replaced objects with dummy ones (no events etc), tried a few times to run it, got similar results to above, then got this exception code etc one time, literally the instant it executed, if it helps:-

 
Last edited:
@rmanthorp - Well, after a shedload of testing, pulling my hair out etc, I can say with some confidence, that I'm sure that this release seems to be screwed in some way....

I had luckily compiled and built a Windows executable of my 0.99 project snapshot back in March, so proved a perfect comparison - and I can run this without crashes, for 20 minutes at least during my test, back and forth between the menu and game, playing a game to completion, choosing human or CPU players, etc, etc...

I take that EXACT SAME project snapshot that this build was created from - it pulls it into the IDE, SAVES(?) it immediately on loading (possibly with changes to file structure etc, which might be causing an issue, I dunno), and it's totally screwed - I can't get in or out of the game screen more than 2 or 3 times, even after exiting seconds afterwards (as in my testing with the earlier-created executable...), and it crashes out CONSTANTLY, with no GMS error message, just a standard Windows error. I compile and build a fresh EXE of this exact same snapshot, and it crashes just as badly as in the IDE. I NEVER got this with the previous, superb release of 2.1.3... Either in IDE, or building it.

Oh, and it seems to do it randomly - even when I simply test it, sometimes the moment it runs even? - If I remove my "GameController" (basically the game and all it initiates) object, it doesn't crash at all (basically because I'm doing nothing, I guess (no grids, no tiles, etc, but it does show that SOMETHING in my game is causing it to crash NOW, when it didn't with the previously-built executable built with the exact same snapshot, and is probably therefore worth investigating by YoYo?). However, this object isn't even loaded or live until the game screen is chosen, and the game sometimes crashes instantly, ie. on the title screen, which has no instance of the Game Controller object even created yet?

Sometimes, when choosing Debug mode, it will even seem to run almost flawlessly, as long as I have a breakpoint in there somewhere, pushing it along every time it stops on it, although it will still eventually crash, so perhaps it's related to alarms, timing etc, I'm not sure at this point... Everything I *thought* it might be, it wasn't....

I am really, REALLY hacked off here with this mandatory update - I just want my earlier build back, that simply WORKED. Does anyone know how to get this running again? Development has stopped completely, and I've lost 3 days just trying to find out if it's down to me and something I was maybe getting away with before, to now feeling certain it's a newly-introduced instability - I was *almost* resorting to hacked earlier releases I'm seeing online, which is a ridiculous state of affairs when I've legitimately paid for GMS2?! - I was happily developing in what I thought was a superb, stable engine, and now this...

Seriously guys, there is something WRONG with this release that's been introduced along with it, I'm certain of it. I tested everything I could think of, disabled/removed huge amounts of events and code trying to find the problem, and YES, I also tried different runtimes (I could only revert to my favourite runtime, which should have worked by all means, but couldn't try earlier ones, as my parent/child object relationships didn't seem to be recognised by the 2.1.2.x runtime even - I just got build/compile errors), so yeah, I also checked for memory leaks etc, but nothing is showing - 20mb usage, doesn't deviate up or down by more that 100kb, and no huge drops in frame rates etc - I double-checked my ds_grid destroys and resizes as pertinent, the game now crashes even at stupid/random points, ie. even when a move has been played, the score is being added, and is zooming back into the screen. Sometimes crashing going INTO the game, sometimes when coming OUT, sometimes when simply moving a piece around the board (or not)....
I would suggest raising a bug through the bug tracker and attaching a link to your project that is crashing. If you give as much detail as you have here for the replication steps, then I am sure YYG will be able to track down what the problem is with the latest runtime and your project.
 

clee2005

Member
Has anyone tested iOS IAPs with the 2.1.4.288 release? I did a build with both 2.1.3.189 runtime and 2.1.4.203 runtime and the IAPs no longer work in TestFlight whereas they used to test fine in TestFlight before. Just wondering if anyone else has had any issues. Nothing has changed in the IAP code, so I'm looking at this new GMS update or possibly TestFlight working differently.

Cheers,
Chris
 

Japster

Member
I would suggest raising a bug through the bug tracker and attaching a link to your project that is crashing. If you give as much detail as you have here for the replication steps, then I am sure YYG will be able to track down what the problem is with the latest runtime and your project.
Cheers for the input and advice @BaBiA Game Studio - Actually, I'm turning in for the night, BUT...

...I've narrowed my main issues down to a LOADCSV command - worked perfectly in 2.1.3.x, toss a coin to see if it works on this version - seems VERY flakey - the CSV file couldn't be simpler, and checks out fine - ironically, I'm using the exact same logic to load the jukebox data, and that works fine.

"Tmp_Board" always equals 2 during testing.
"ww" always returns 22
"hh" always returns 16



If I replace my CSV command with this, it works *perfectly*, EVERY SINGLE TIME - no, it's not a fix (for a start, it makes all cells active, but it pretty much sums it up - I create the same size ds_grid, albeit not from the CSV, and populate it with active squares:-



(and it's honestly a very simple CSV file, here are the contents):-



So in short, I'm guessing something's possibly happened to the LOADCSV functionailty? - I've re-imported the file, and still no joy, nor my other layout files, which is why I feel that there is nothing I can do until this is investigated / resolved. My jukebox LOADSCSV'd ds_grid works fine, but it's a much smaller file (6 records, 3 fields, as opposed to 22 x 16, as in this case)...

I'll log a bug report tomorrow night, but some input or acknowledgement of this from YoYo in the meantime would be good...

Cheers....

-------------------------------
EDIT:- Okay, enough thread de-railing.... I've worked out a bodge for now - adding a carriage return at the end of my problem CSV makes it work PERFECTLY.... ?!? - Weird, as I tried other level files just now too, which don't have one, and end at the last line as above, and they work fine anyway, either as they are already, or with the added carriage return?... I'm happy to just have a fix! - I can work around the added line, although I'm a little concerned at how it will make the last row of the DS_Grid build, but it loads, doesn't crash, and seems to work around the issue - I'm very confused, but relieved....
 

Attachments

Last edited:
@Japster: I'm not 100% sure on this, but I'm almost positive you can't declare a global variable and then reference it as if it's a local variable. Change global.Board_Layout to just Board_Layout and see if that fixes it.
That's 100℅ correct for globals declared using "global." notation.

You used to be able to use globals like that when using the globalvar keyword, but it's not recommended to do that anymore as that method has been deprecated and may be removed in future.

Plus it's more obvious what are globals using the new method.
 
W

whale_cancer

Guest
Alright, whenever I have narrowed down my problem to load_csv(); calling this function always crashes gamemaker for me now, despite it working for me in previous versions without a problem.
 

rwkay

GameMaker Staff
GameMaker Dev.
Please file a bug with a project that shows the problem, it is most likely an issue with parsing that specific CSV file (so ensure the project has the file in it) - otherwise it is difficult to fix it (my force powers are not that good).

Russell
 

Ricardo

Member
Please file a bug with a project that shows the problem, it is most likely an issue with parsing that specific CSV file (so ensure the project has the file in it) - otherwise it is difficult to fix it (my force powers are not that good).

Russell
It is really a shame that your force powers to fix already opened bugs (many of whom have been opened for several months) are also not that good. :(
 

Ricardo

Member
I'm sure Russel doesn't like that either, but he's not the only one developing this engine, and there are tons of other stakeholder and people to take into consideration when posting such a comment.
You are right, it was an inappropriate comment and unfair with Russell. I'm sorry Russell, it wasn't intended to be a personal offense. I got carried away by my frustration.
 

rmanthorp

GameMaker Staff
Admin
GameMaker Dev.
Another 2.1.4 Runtime release. This should fix the remaining collision issues. Please check again and let us know.

http://gms.yoyogames.com/release-notes-runtime.html
Update 2.1.4.212+
  • Fixed 0029284: General: Fast Collisions floor floating point values rather than ceil them which Slow Collisions do, causes changes in gameplay
  • Fixed 0029293: Rooms: [GMS2] Room Persistence causes collision issues with instances that don't exist in the new room
  • Fixed 0029280: Build Android: Android TV Banner doesn't get applied correctly, stops submissions from being successful
  • Fixed 0029282: Variables: GM_Version on Android is wrong and adds in .0 at the end even though version is only 3 numbers
  • Fixed 0028688: Code Editor: display_set_gui_maximize() recognised by auto-complete but not recognised by compiler
  • Fixed 0029299: Variables Window: 'image_blend' is 0 when assigned a colour from Variable Definitions
  • Fixed 0029257: HTML5: Mouse position not scaled to room view when using cameras
  • Fixed extensions on Mac, Ubuntu, and some other platforms not being able to call extension functions other than the first one in their list
 
W

whale_cancer

Guest
Please file a bug with a project that shows the problem, it is most likely an issue with parsing that specific CSV file (so ensure the project has the file in it) - otherwise it is difficult to fix it (my force powers are not that good).

Russell
Bug report submitted. I don't think it is the .csv file as, like I mentioned before, the file stayed the same but the project broke after the GM update. The included project is literally just a load_csv() call on an included .csv file. The project uploaded uses an included file, although in my project I was loading the .csv with code rather than as an included file. The crash looks exactly the same.

Edit: Oh, and I was using multiple different .csv files before, all of which were loading properly and all of which now cause a crash. These csv files were created using pyxel edits tilemap export function. Here is the one in the project I submitted in the bug report if anyone wants easy access:

Code:
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,2,3,3,3,3,4,21,21,21,21,21,
21,21,21,21,21,15,16,16,16,16,17,21,21,21,21,21,
21,21,21,21,21,15,16,16,16,16,17,21,21,21,21,21,
21,21,21,21,21,15,16,16,16,16,17,21,21,21,21,21,
21,21,21,21,21,15,16,16,16,16,17,21,21,21,21,21,
21,21,21,21,21,30,31,31,31,31,32,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,21,
 

kupo15

Member
So are they both ceiling now, or both flooring?
I first read it as both ceil but I think they both are floored which would make sense. If I were to rephrase it
" Fast Collisions floor floating point values rather instead of than ceil them."

The mention of Slow Collisions made it confusing. In any case even more reason for me to make sure I deal with fractional coors myself before doing collisions instead of having GM do it for me.

Glad to see the update, hopefully now it'll be safe for me to update
 
So are they both ceiling now, or both flooring?
I think you would need to look at the bug in the bug tracker for any more detail. The problem I find with the release notes is that the list of items being fixed is exactly the text that the person raising the bug put. So if it is someone that is unable to clearly define their problem in the title of the bug, it is always going to read as something strange in the release notes.
 

Japster

Member
@Japster: I'm not 100% sure on this, but I'm almost positive you can't declare a global variable and then reference it as if it's a local variable. Change global.Board_Layout to just Board_Layout and see if that fixes it.
@Thylacine Studios - Cheers for the input, appreciated! - TBH, the code above is just left over from my attempts to figure out the issue - I've declared it as a global in my controller create / game start code, and out of desperation when I was trying to figure out if I'd gotten mixed up in the same way you're suggesting, I started referencing it with "global." before it.

You're right though, when picking up GMS2 at first, I got really thrown by getting my globals mixed up - on the whole, it would *seem* to work, but yeah, I got weird spurious issues, argh!

Seems like myself and @whale_cancer (thanks for confirming that mate) are having the same issue - ALL of my project's CSV's worked prior to this, and now don't even work (ie. without messing with the CSV file to fudge it) even when trying to use older runtimes... weird...

I also removed all content at the end of my file to test, even deleting and re-entering the last data digit - broke my game again immediately. Only adding a carriage return worked, even changing the last digit didn't, so I really think it's a bug - will highlight it as such for @rwkay :) .

Obviously I can also check this with a hex editor to make triple sure... :)
 

rwkay

GameMaker Staff
GameMaker Dev.
@Andrey - good thing we never claimed that in the Release Notes - we are working on it, just going to take longer to fix properly
 
Status
Not open for further replies.
Top