OFFICIAL 2.3.7 Stable Release Thread

[Dan]

That's an updated 2.3.7 release rolling out now - IDE v2.3.7.606 Runtime v2.3.7.476

This one is a collection of the stability and performance fixes which have been in the Beta releases for the last week or so, plus the fixes for some smaller-and-often-reported issues in the first release - for example, a bunch of broken hotkeys. Note that there are a lot more fixes in the Betas already, but the emphasis here was on safely resolving top issues.

See the release notes for the full info (per usual, all the changes are at the top):
IDE Notes
Runtime Notes

 

[Cameron]

That's an updated 2.3.7 release rolling out now - IDE v2.3.7.606 Runtime v2.3.7.476

This one is a collection of the stability and performance fixes which have been in the Beta releases for the last week or so, plus the fixes for some smaller-and-often-reported issues in the first release - for example, a bunch of broken hotkeys. Note that there are a lot more fixes in the Betas already, but the emphasis here was on safely resolving top issues.

See the release notes for the full info (per usual, all the changes are at the top):
IDE Notes
Runtime Notes

Am installing it now but it keeps getting stuck saying that the selected runtime could not fully install. Going to restart my computer and see if that helps.

Edit: Restarting computer did the trick and the runtime was able to finish installing.

 

[Terminator_Pony]

I can't use number keys or numpad 1-3 in the sprite editor without it cycling images/going crazy. Affects everything from inputting grid size, brush size, color picker window...

Anyway to make it stop? I looked in the key definitions and they aren't even listed.

 

[gnysek]

In-Game: Nullish operator now works with struct/map/list accessors

Wow, this stable has fixed even more bugs than beta (but as I understand all of them will be in tomorrow/Friday beta too). Nice, I got lot of place where I could use above feature, so I will optimize lot of my code now

 

[Panda_Iro]

That's an updated 2.3.7 release rolling out now - IDE v2.3.7.606 Runtime v2.3.7.476
...

Definitely fixed the performance issue for my M1 Airbook, it went from choppy to smoother than my Mac Pro. Builds seem to be taking longer but otherwise it's running great. Thanks team!

EDIT - WHOA! The sequence editor (which, as an animator, I love) has always felt really sluggish on my Mac Pro, but it is like butter now. That alone will save me a lotl of time, Kudos!

 

[peardox]

I thought I was going crazy earlier when my script wasn't acknowledging that jpegs are a file format (apparently already reported but not mentioned in the fix list)

606 landed and everything started working again

 

[8BitWarrior]

I don't know if it's an issue with my Windows 10 install, but I can't get the GameMaker runtimes to install without an "Exception:System.IO.PathTooLongException" error which shows up in the ui.log. I updated my Windows registry to allow for longer filenames, but it didn't resolve this issue. I do not run any anti-virus software, other than default Windows 10 stuff.
This is happening with the latest stable and beta.

Update 1: I found that I am able to install the default runtimes with a different GameMaker account using a free license. But when I log back into my main GameMaker account, it fails to install the remaining runtimes and won't let me past. From the logs, it looks like the issue might have to do with the tvOS runtime failing to install.

Update 2: After a fresh Windows install, it is now working. I'll leave the log output below just in case this issues shows for anyone else.

I'll share a bit of the log output here:

Spoiler: ui log

[16:03:51:222(0407)] verifying file C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\download\cb523c98-9ce3-4421-8070-a53958e5ecff.zip matches signature MCwCFFTbGm6rUrx8Z8HP1CmL/4gp5mt9AhQ986h7x2YaV7sGG272wy1S/+w/Pw==
[16:03:51:916(0407)] verified file C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\download\cb523c98-9ce3-4421-8070-a53958e5ecff.zip matches signature MCwCFFTbGm6rUrx8Z8HP1CmL/4gp5mt9AhQ986h7x2YaV7sGG272wy1S/+w/Pw==
[16:03:51:917(0407)] unpacking zip to temporary directory C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\hz3vtbmh.mqo
[16:03:56:174(0407)] Unzip [zip] Exception:System.IO.PathTooLongException: The specified path, file name, or both are too long. The fully qualified file name must be less than 260 characters, and the directory name must be less than 248 characters.
at System.IO.PathHelper.GetFullPathName()
at System.IO.Path.LegacyNormalizePath(String path, Boolean fullCheck, Int32 maxPathLength, Boolean expandShortPaths)
at System.IO.FileStream.Init(String path, FileMode mode, FileAccess access, Int32 rights, Boolean useRights, FileShare share, Int32 bufferSize, FileOptions options, SECURITY_ATTRIBUTES secAttrs, String msgPath, Boolean bFromProxy, Boolean useLongPath, Boolean checkHost)
at System.IO.FileStream..ctor(String path, FileMode mode, FileAccess access, FileShare share)
at SharpCompress.Archive.IArchiveEntryExtensions.WriteToFile(IArchiveEntry entry, String destinationFileName, ExtractOptions options)
at SharpCompress.Archive.IArchiveEntryExtensions.WriteToDirectory(IArchiveEntry entry, String destinationDirectory, ExtractOptions options)
at SharpCompress.Archive.IArchiveExtensions.WriteToDirectory(IArchive archive, String destinationDirectory, ExtractOptions options)
at YoYoStudio.Core.Utils.ZipUtil.ExtractZipArchive(String _archiveFullName, String _extractDir, String _password, String& _outErrorMsg, ProgressCallback _progressCallback)
[16:03:56:175(0407)] FAIL - unable to extract the downloaded zip file C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\download\cb523c98-9ce3-4421-8070-a53958e5ecff.zip to C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\hz3vtbmh.mqo error was The specified path, file name, or both are too long. The fully qualified file name must be less than 260 characters, and the directory name must be less than 248 characters.
[16:03:56:175(0407)] Deleting Temporary Directory C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\hz3vtbmh.mqo
[16:03:56:208(0407)] Finished Wait for all Modules
[16:03:56:209(0407)] failed to download runtime module tvos
[16:03:56:209(0407)] moduleName = tvos, HasModule( tvos ) == True
[16:03:56:210(0407)] start installation for Module tvos
[16:03:56:210(0407)] DownloadInstallandVerify tvos to C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476
[16:03:56:211(0407)] Wait for all Modules
[16:04:01:196(0407)] verifying file C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\download\cb523c98-9ce3-4421-8070-a53958e5ecff.zip matches signature MCwCFFTbGm6rUrx8Z8HP1CmL/4gp5mt9AhQ986h7x2YaV7sGG272wy1S/+w/Pw==
[16:04:01:880(0407)] verified file C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\download\cb523c98-9ce3-4421-8070-a53958e5ecff.zip matches signature MCwCFFTbGm6rUrx8Z8HP1CmL/4gp5mt9AhQ986h7x2YaV7sGG272wy1S/+w/Pw==
[16:04:01:881(0407)] unpacking zip to temporary directory C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\5dacgkj1.t40
[16:04:06:35(0407)] Unzip [zip] Exception:System.IO.PathTooLongException: The specified path, file name, or both are too long. The fully qualified file name must be less than 260 characters, and the directory name must be less than 248 characters.
at System.IO.PathHelper.GetFullPathName()
at System.IO.Path.LegacyNormalizePath(String path, Boolean fullCheck, Int32 maxPathLength, Boolean expandShortPaths)
at System.IO.FileStream.Init(String path, FileMode mode, FileAccess access, Int32 rights, Boolean useRights, FileShare share, Int32 bufferSize, FileOptions options, SECURITY_ATTRIBUTES secAttrs, String msgPath, Boolean bFromProxy, Boolean useLongPath, Boolean checkHost)
at System.IO.FileStream..ctor(String path, FileMode mode, FileAccess access, FileShare share)
at SharpCompress.Archive.IArchiveEntryExtensions.WriteToFile(IArchiveEntry entry, String destinationFileName, ExtractOptions options)
at SharpCompress.Archive.IArchiveEntryExtensions.WriteToDirectory(IArchiveEntry entry, String destinationDirectory, ExtractOptions options)
at SharpCompress.Archive.IArchiveExtensions.WriteToDirectory(IArchive archive, String destinationDirectory, ExtractOptions options)
at YoYoStudio.Core.Utils.ZipUtil.ExtractZipArchive(String _archiveFullName, String _extractDir, String _password, String& _outErrorMsg, ProgressCallback _progressCallback)
[16:04:06:36(0407)] FAIL - unable to extract the downloaded zip file C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\download\cb523c98-9ce3-4421-8070-a53958e5ecff.zip to C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\5dacgkj1.t40 error was The specified path, file name, or both are too long. The fully qualified file name must be less than 260 characters, and the directory name must be less than 248 characters.
[16:04:06:36(0407)] Deleting Temporary Directory C:\ProgramData\GameMakerStudio2\Cache\runtimes\runtime-2.3.7.476\5dacgkj1.t40
[16:04:06:71(0407)] Finished Wait for all Modules
[16:04:06:71(0407)] failed to download runtime module tvos
[16:04:06:111(0407)] validate failed for tvos it is in _runtime not in json
[16:04:06:112(0407)] starting cmd /c subst C: /d
[16:04:06:152(0407)] elapsed time 00:00:00.0409929s for command "cmd" /c subst C: /d started at 12/09/2021 16:04:06
[16:04:06:191(0407)] ShowInfoBox :: title="RuntimeDownload_Failed", details=""

 

[Abaloneater]

Anyone knows that why the "Download" button leads to release notes while updating?

 

[BaBiA Game Studio]

Anyone knows that why the "Download" button leads to release notes while updating?

So you know what is in the version you are downloading, and it gives you something to read instead of staring at the download progress . I believe it has always done this (at least it has for me for as long as I can remember).

 

[Dan]

I believe it has always done this

Indeed it has. Every version of GMS2


Thanks for all the immediate feedback about performance being good again - lovely to read that the update is working well for you all.

 

[Deleted User]

Indeed it has. Every version of GMS2


Thanks for all the immediate feedback about performance being good again - lovely to read that the update is working well for you all.

works good on my end too, and no crashes on any systems we have (sofar) great work, keep it up

2.3.8 stable coming on january?

 

[gnysek]

2.3.8 stable coming on january?

If it will contain only Font Inspector as per current roadmap, I believe they might be able to release it in December, even when having traditional 1-2 week holidays around Christmas

 

[Alice]

If it will contain only Font Inspector as per current roadmap, I believe they might be able to release it in December, even when having traditional 1-2 week holidays around Christmas

Actually, it wasn't uncommon for Roadmap to show still in-development features associated with a specific beta. So just because Font Inspector is the only feature In Beta at the moment doesn't mean this is all that's planned for the 2.3.8 release. ^^'
(at any rate, nice to see we're about to get Intellisense Improvements and Find References in 2.3.8/2.3.9 release)

 

[gnysek]

we're about to get Intellisense Improvements and Find References in 2.3.8/2.3.9 release

oh, yeah, it actually moved up on priority list! Must be day or two ago, as google cache still had previous version.

So seems it's for sure Q1 2022. From all of above features, intellisense, triggers, and UI are on top of my list (intelissense is first, as it's the only here which cannot be done using current features of GM). Even if that would require writing a lot of JSDoc in every event, to mark variables as instances of some struct/constructor, I would be happy if there will be any way to get hints about constructor methods, so I'm eagerly waiting to see what they cooked there

 

[Mehdi]

"Triggers" are a MUST for my project. I wish it could go up in the list.
Right now I use objects as triggers which is not an optimum way at all.

 

[Stra]

 

[Sean Catherine]

First, the two machines I'm using:


Mac mini (M1, 2020) Mac OS Monterey, Version 12.0.1
Chip: Apple M1
Memory: 16 GB
&
Mac Pro (Late 2013) Mac OS Monterey, Version 12.0.1
Processor: 3.5 GHz 6-core Intel Xeon E5
Memory: 64 GB 1866 MHz DDR3
Graphics: x2 AMD FirePro D500 3 GB

I've just downloaded the latest GMS2 update and the experience has blown me away: I have a very massive project with an enormous amount of graphical assets; so much so that I've had to split up the project into 625 separate projects from which I'll create local packages... So I do my major project editing on the M1, and I have a map of the game as a whole in a room on the Mac Pro. This map is 18,200 x 18,200 consisting of 625 asset thumbnails that are each 728 x 728. Those thumbnail png.s each have file sizes ranging from 1.3 to 1.6 MB. It wasn't until this morning after this update that I was able to work on that map with anything other than a very sluggish pace (on both machines). Now, I'm able to zoom in and out of the map/ edit it without ANY lag that I can sense. There had to have been some very major optimization that had taken place because I've never had such ease with this in the over 4 yrs that I've been working with GMS2. This update has made things so much easier for the way I work in the IDE.

 

[Dan]

View attachment 44825

Hehe, I did also say that in the 2.3.8 Beta thread's announcement post: https://forum.yoyogames.com/index.php?threads/2-3-8-betas-thread.91485/post-549132

In fact, I have forgotten to edit those lines to remove the bit about if you let us know about the stability we will release a 2.3.7 update, so will go do that now that update release is out


https://www.yoyogames.com/en/blog/gamemaker-version-236 did also announce a while back that it's 10 releases a year, not 12, but I see now that we forget to state explicitly that Dec was one of those "missing" two.

 

[Dan]

There had to have been some very major optimization that had taken place because I've never had such ease with this in the over 4 yrs that I've been working with GMS2. This update has made things so much easier for the way I work in the IDE.

Glad to hear it's made such an improvement!

 

[Panda_Iro]

Is anyone else getting issues where duplicated sequences "revert" to their original state?

The sequence duplicating issues seemed to have mostly resolved but when I make duplicates now, and then make minor adjustments to one (ie move some key frame positions), when I run the game or maybe at random times, the duplicate will entirely default back to the original state. Doesn't always happen, but sometimes.

 

[kingyo]

Font editor > Add > FromFile > I can not load a csv file.

 

[Dan]

Font editor > Add > FromFile > I can not load a csv file.

That one is a deliberate change (you can see it in the release notes), as people were trying to pass in .rtf files and other forms of text file which were not going to work and then sending us bug reports that the IDE allowed the file. However, we are going to review that change further and see if more filetypes can/should be allowed again.

 

[kingyo]

That one is a deliberate change (you can see it in the release notes), as people were trying to pass in .rtf files and other forms of text file which were not going to work and then sending us bug reports that the IDE allowed the file. However, we are going to review that change further and see if more filetypes can/should be allowed again.

I keep all my in-game text in csv files, so I'm at a loss if I can't handle csv files.

 

[Nehemek]

Not exactly a bug but at least on my M1 Mac, Mono didn't install automatically and took me a bit to figure it why builds failed. Sure it is listed in the SDK Requirements page but not on the Troubleshooting MacOSX one which is an article that hasn't been updated in 8 years. Other than that everything else seems to be going smoothly~

 

[kupo15]

I keep all my in-game text in csv files, so I'm at a loss if I can't handle csv files.

This is very confusing change for me to hear, I thought the community advice for using external text was to use .csv format. Dan, are you saying that including the csv format also allowed .rtf and other files to be selected causing issues as a consequence?

 

[rIKmAN]

but not on the Troubleshooting MacOSX one which is an article that hasn't been updated in 8 years.

That's because that article is for GMS 1.4, not GMS2.

This is very confusing change for me to hear, I thought the community advice for using external text was to use .csv format.

The manual says GMS2 will load the file and "create character ranges for the text found within it."
An untested suggestion: can you not make a copy, change the extension to .txt and load it that way?

 

[kingyo]

An untested suggestion: can you not make a copy, change the extension to .txt and load it that way?

In my game, for example, all conversations are stored in a csv with various parameters, and a dedicated parsing function makes them available to the game. It's not impossible to rework all of this, but it would be a lot of work.
I don't understand why the IDE provides a load_csv but refuses to add characters from the csv.
If YoYoGames doesn't change this spec change, I will have to change it to be handled in a txt file, as you suggest.

 

[rIKmAN]

In my game, for example, all conversations are stored in a csv with various parameters, and a dedicated parsing function makes them available to the game. It's not impossible to rework all of this, but it would be a lot of work.
I don't understand why the IDE provides a load_csv but refuses to add characters from the csv.
If YoYoGames doesn't change this spec change, I will have to change it to be handled in a txt file, as you suggest.

Maybe I've misunderstood, but from reading the manual Font editor > Add > From File reads the characters within the file and adds the ranges to the font, but has been limiited recently to only load files with a .txt extension due to people trying to load files like .rtf etc which don't work.

.csv is plain text with a different extension, so making a copy of your .csv and renaming it to have a .txt extension should allow it to be loaded as a file in Font editor > Add > From File to add the characters as it seems to filter supported files by .txt extension - you can continue to use the .csv file as normal for reading the data into the game as you have been doing

Basically 2 copies of the same file:
mydata.txt - load into Font editor > Add > From File to add the character ranges.
mydata.csv - load as you have been doing to get your data in.

As I said I may have misunderstood and haven't tested it, but this was the first thing that came to mind as workaround until the Font Editor allows .csv as a valid extension in the file selector again.

 

[kingyo]

@rIKmAN
I understand what you are suggesting. If there was no other way, I would do the same as you. But it's a complete pain to have to do that every time you change the csv. That's why I'm asking YoYo to let me add the characters from the csv as before.

 

[rIKmAN]

@rIKmAN
I understand what you are suggesting. If there was no other way, I would do the same as you. But it's a complete pain to have to do that every time you change the csv. That's why I'm asking YoYo to let me add the characters from the csv as before.

I understand, I wasn't suggesting this to be the final solution but just a temporary workaround.

Dan said above they will review re-adding more supported filetypes at a later date, so my suggestion was because if it works then a couple of clicks and renaming a file extension will take a few seconds and allow you to continue working, rather than waiting an unknown amount of time for the update that adds .csv support back into the file selector.

 

[kupo15]

I'm noticing a bug in the draw_sprite_stretched_ext function, it doesn't appear to be taking the origin into account anymore and just makes the origin top left. Anyone else notice this?

 

[Terminator_Pony]

I'm noticing a bug in the draw_sprite_stretched_ext function, it doesn't appear to be taking the origin into account anymore and just makes the origin top left. Anyone else notice this?

That's how its been for a while, if not always: https://manual.yoyogames.com/GameMa...rites_And_Tiles/draw_sprite_stretched_ext.htm

 

[kupo15]

That's how its been for a while, if not always: https://manual.yoyogames.com/GameMa...rites_And_Tiles/draw_sprite_stretched_ext.htm

Wow, how long has this been the case? I upgraded from 2.3.6.464 and the sprites that draw using this function are not positioned correctly anymore so I have to adjust those. That's super weird!

 

[chamaeleon]

Wow, how long has this been the case? I upgraded from 2.3.6.464 and the sprites that draw using this function are not positioned correctly anymore so I have to adjust those. That's super weird!

At least since GMS 1.4.

This function ignores the sprite's origin.

 

[kupo15]

At least since GMS 1.4.

Well I am perplexed. It should not have been working the way it has been before the upgrade then, unless there was a bug with how it handled 9slices. Super weird

 

[FrostyCat]

After dealing with yet another OneDrive-related project loss topic today, I would like to know the status of my suggestion ticket #176690 from October 2020. For disclosure, this is what it reads:

Feature Type: feat_change_ide

Description: The IDE should show a warning when a project is saved somewhere containing these terms in the path:

- OneDrive
- Dropbox
- com~apple~CloudDocs
- Google Drive

The automatic sync services that these apps offer are known to fight with the IDE over file access, and cause unexpected changes in projects saved in these locations.

Example Message:

This project is saved in a location managed by a cloud storage service:
C:\Users\dev\Dropbox\MyProject
Automatic sync with cloud storage is known to cause unexpected data loss. Please ensure automatic sync is OFF while the IDE is open, or move the project elsewhere.

Benefit: The GMC is still regularly getting complaints from users who lose data when these services update files unexpectedly and fight with the IDE over access. If the IDE can pre-emptively warn users against putting projects in these locations, this would help safeguard novices from unexpected data loss.

It has been marked as "solved", yet given the persistence of similar sob stories on the Q&A boards over the past year, I can safely say it has not been implemented.

Funny enough, this kind of message is already in the IDE for certain directories, such as %TEMP%:

What exactly is preventing the logic behind the message box above from applying to the likes of OneDrive and Dropbox? Especially with all the rookies flooding in from HB, at least a few will make the mistake of storing projects in those places and losing their early work. Anything that can discourage these bad choices early helps.

 

[Abaloneater]

Fixed random quit on most cases, somehow, it caused windows blue screen after clicked on stopping testing.

 

[gnysek]

unless there was a bug with how it handled 9slices. Super weird

Yes, there was bug with 9-slices, I've reported it during one of betas and they told me, that this bug was since 9-slices were introduced, which could lead to assumption that since it's there from beginning, origin isn't ignored for them - but it should be.

 

[COWCAT]

I'm using 2.3.7 now and I'm experiencing the same issue with scripts marked as "not saved" * constantly. It appeared with 2.3.4 and I notice it's still there and super annoying.

It's pretty easy to reproduce:
- Open a script
- Make any change (add a space for example)
- Save with CTRL + S
- Click anywhere else (another window like Notepad)
- The script has now been marked as "not saved" * even though I made no change after saving.

Also, pressing CTRL + S again in the script will NOT save (further proof that this is a bug), so the * is still there until you actually make more changes or close it down.

Has anyone reported this to support? Or should I do it?

 

[rIKmAN]

I'm using 2.3.7 now and I'm experiencing the same issue with scripts marked as "not saved" * constantly. It appeared with 2.3.4 and I notice it's still there and super annoying.

It's pretty easy to reproduce:
- Open a script
- Make any change (add a space for example)
- Save with CTRL + S
- Click anywhere else (another window like Notepad)
- The script has now been marked as "not saved" * even though I made no change after saving.

Also, pressing CTRL + S again in the script will NOT save (further proof that this is a bug), so the * is still there until you actually make more changes or close it down.

Has anyone reported this to support? Or should I do it?

This has been fixed in the beta so no need to report it, it'll be in the next stable release.
From the Beta Release Notes:

• Code Editor: Unsaved changes asterisk indicator on script tabs can return when GMS2 loses focus after a save operation

You can install the beta alongside Stable so it's useful to check if things have already been fixed before reporting them even if it's not an option to actually use the Beta itself. Alternatively keep an eye on the Beta Release Notes linked above for the same reason.

 

[COWCAT]

Ah, nice! Glad to hear!
Hopefully the "white cursor on white background" on text fields in light mode will be solved soon as well (I have reported that one a few versions ago)

 

[rIKmAN]

Hopefully the "white cursor on white background" on text fields in light mode will be solved soon as well (I have reported that one a few versions ago)

I think the problem here is........that you're using the light theme!

What did your eyes ever do to you to deserve that!?

 

[Dan]

That's the new "2.3.8" stable released - only, it's now called 2022.1.0! Its release thread is now available, so I will lock and move this one now.

As ever, a big thank-you from all of us to everyone who gave feedback and bug reports for 2.3.7. Hope you enjoy the new release!