• 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.

GMS 2.3+ GameMaker needs "stable" releases that don't get new features but do get bug fixes

SIG.

Member
Unity has "long term support," or LTS, releases. For example, right now, 2019.4 LTS is the stable release. What this means is that new features will not be added to the LTS release, but big fixes will be back-ported into it. New features instead go into the newer 2020.x releases. People can use these releases knowing that certain of the new features may change, sometimes significantly depending on the particular feature-set and its designation (some are "experimental," etc.).

Right now? I really wish GMS had something similar. As a mobile dev, I want basic bug fixes and platform updates. I don't care about the new features at all; I'm never making anything in GMS again, I'm just stuck with it for maintenance right now. But if I want the mobile platform updates, I'm forced to update both the IDE and runner to 2.3, which have a lot of bugs. Some of those bugs have workarounds, some don't, and the mystery 2.3 bug(s) crashing my game are going to steal a lot of time figuring out whether and how they can be worked around, and whether it's better for me to try to work with 2.3 or else figure out how to manually graft the new iOS requirements onto my 2.2.5 build. (And, of course, even if I bite the bullet with 2.3 right now and can make my game work again, there's a whole new universe of potential bugs injected by the new 2.3 features.)

YYG thought they could get around this by making things modular, but that hasn't worked out. The IDE and runner were supposed to be separate, but we've had at least two forced, linked updates. YYG somehow intended to separate the IDE and runner for GMS2 while also baking GML functionality directly into the IDE. Mobile packages were supposed to be at least somewhat separate, but they're not. Basic export functionality has always been hardwired into both the IDE and runner. The IAP updates required updating the runner and rewriting the entire IAP system, as well as importing a new package. (I'm really not clear why they couldn't make the updates behind the scenes and map the existing API onto it. Someone's making an asset to streamline into a single API, seems obvious YYG should have.)

I don't think only mobile devs could use this. I expect that many people with a complex project that's past the initial stages of development would choose stability over new features.
 

kburkhart84

Firehammer Games
The idea makes sense...the issue is the difficulty in actually packporting code like that. Depending on what it actually is, it isn't always just a simple copy/paste to the old project. Unity AFAIK has a much bigger team behind them so it is feasible for them, while it may not be an option that works for Yoyo.
 

Non.A

Member
I was so excited to get this after researching. it seems perfect for what I need but I didn't realize it was so buggy. it's a shame because it seems like such a good product.
 
I was so excited to get this after researching. it seems perfect for what I need but I didn't realize it was so buggy. it's a shame because it seems like such a good product.
Just remember that people always talk more when they have a problem. There's literally thousands of very satisfied users of GMS, but they're not constantly posting topics saying "I have no bugs!" On the other hand, whenever a random person does encounter a bug, very often the first thought is to search for the community around the product and post about the bug, either to get help or to complain. It's a classic case of the vocal minority. You can rest assured that most users have absolutely no problem working within GMS without encountering bugs (I've used it for many years and very rarely run into any bugs and the rare ones I have encountered are not impossible to solve).
 

kburkhart84

Firehammer Games
Just remember that people always talk more when they have a problem. There's literally thousands of very satisfied users of GMS, but they're not constantly posting topics saying "I have no bugs!" On the other hand, whenever a random person does encounter a bug, very often the first thought is to search for the community around the product and post about the bug, either to get help or to complain. It's a classic case of the vocal minority. You can rest assured that most users have absolutely no problem working within GMS without encountering bugs (I've used it for many years and very rarely run into any bugs and the rare ones I have encountered are not impossible to solve).
This, 100%. I will say that version 2.3 is buggier than the version 2.2.5, but they are fixing those soon as I understand it. If it is that big of an issue you can use that version(just remember the new features like sequences and structs won't be there until you upgrade again).
 

Non.A

Member
Just remember that people always talk more when they have a problem. There's literally thousands of very satisfied users of GMS, but they're not constantly posting topics saying "I have no bugs!" On the other hand, whenever a random person does encounter a bug, very often the first thought is to search for the community around the product and post about the bug, either to get help or to complain. It's a classic case of the vocal minority. You can rest assured that most users have absolutely no problem working within GMS without encountering bugs (I've used it for many years and very rarely run into any bugs and the rare ones I have encountered are not impossible to solve).
thanks for the perspective. maybe if I can get it running I'll see things differently. it's just I've hit a roadblock or a bug at every step since installing it. after days of trying I finally got the install to work, only to be immediately presented with an error message when I try to run the game:
"WARNING: The Carbon driver has not been ported to 64bits, and very few parts of Windows.Forms will work properly, or at all."
I've already spent 2 hours trying to figure that one out. it just seems like a lot of troubleshooting before I even get to use the program.
 

gnysek

Member
I think that without subscription instead of lifetime licence that would never happen, as they don't have enough developers to keep so many branches alive at same time. That would make even less updates and even less bug fixes.
 

SIG.

Member
How about pulling out the platform-specific functionality, to be maintained and updated separately as modules? Presumably it's already separate within the engine... Or does the YYC compiler have what's essentially a bunch of buried checks that say, "if Windows do this, if HTML5 this, etc."?
 

Plura

Member
I would say that the gms 2.3 is not ready for the stable version as it has too many changes.
Garbage collector makes us headaches as it crashes our game without any explanation.
New features are welcome but stability has to be #1 priority.

I know that the yoyo staff is small compared to other popular engines and they can't afford a stable and rapid development at the same time. They sacrifice stability in order to get new tools faster. On paper, it looks perfect but in reality, it hurts some developers as they catch some yoyo bugs besides their own bugs.

We all need to work together tighter, yoyo needs extreme good game developer teams to stay on the track of top game engines.
This is an ecosystem where yoyo needs us and we need yoyo and stability should be the foundation of this cooperation so we can focus more on making games.
Yoyo send us some polls, so we can upvote what we need. Maybe most of us need stability over new features. Let's find out!
Maybe the subscription plan is the way to go for the pro users.

I was shocked by how many changes and new tools gms 2.3 offers.
If there was no version number I would say that this is a 3.0 beta version.
For god sake, gives us stability and long-term SDK support over new tools.
New tools are always welcome but not at the cost of stability.
 

FoxyOfJungle

Kazan Games
I agree, and they are taking too long to release a new version in which fix several bugs and especially the annoying code folding thing 🙄
GMS 2 could change its versioning using the year (like 2020.3), which even Unity did, like you say. It makes even more sense, so the engine would only be updated over time. If there should be GMS 3.0 I don't know, but if it is the case to implement a better 3D API, they should do it using the version number with "years".
 

Tornado

Member
@RefresherTowel & @kburkhart84 I don't agree with you.

I guess the vast majority of GMS users are using GMS more or less as a toy tool. That's ok, nothing bad with it.
We (very small team) invested 6 years of OUR LIFES developing our game with GMS.

Of course GMS is great product, but the users have rights to complain and they should complain, otherwise how would GMS improve if not listening to users' needs.

2.3 should not have been 2.3! It seems to me like a mistake. But maybe they had no other choice. I don't know.
I just know that developers who are serious about their product(s) and have releases in production are in certain circumstances pretty fuc*ed with GMS 2.3. It is like spit in the face.

At one hand I was very pleased that after migration to 2.3 we had to do just minor changes and the game ran almost without problems.
But on the other hand why Yoyo released a version with major GML changes and marked it as stable where for example garbage collector is in some "experimental" state?

We lost 3 full days (2 people each) chasing a bug where garbage collector crashes the game!
(I posted the problem in another thread.)
We still couldn't find the problem. This isn't funny for us, this isn't funny for our investor. We have a game in production on Google Play, Apple App Store and Huawei App Gallery and soon we will have to deliver next update of our game, which is very problematic now with GMS 2.3.

I'm affraid in long term Yoyo will lose more of the serious users. I don't know exactly what is their goal, which developers they are aiming at to keep.
We like Game Maker very much, we invested so much in it and we hope that some day it can raise from the shadows of other game engines.
But strategy like this makes me ask myself if I would do my next project with GMS.

This is just my opinion. If I'm the minority of GMS users, then my voice won't count anyway.

I don't want to attack anyone here, everyone fights with their own problems and Yoyo is not an exception here.
I'm just reflecting on our current situation here and giving my opinion.
If some important pieces are missing (not tested well), then it is not suitable for professional use.
 

Toque

Member
@RefresherTowel & @kburkhart84 I don't agree with you.

I guess the vast majority of GMS users are using GMS more or less as a toy tool. That's ok, nothing bad with it.
We (very small team) invested 6 years of OUR LIFES developing our game with GMS.

Of course GMS is great product, but the users have rights to complain and they should complain, otherwise how would GMS improve if not listening to users' needs.

2.3 should not have been 2.3! It seems to me like a mistake. But maybe they had no other choice. I don't know.
I just know that developers who are serious about their product(s) and have releases in production are in certain circumstances pretty fuc*ed with GMS 2.3. It is like spit in the face.

At one hand I was very pleased that after migration to 2.3 we had to do just minor changes and the game ran almost without problems.
But on the other hand why Yoyo released a version with major GML changes and marked it as stable where for example garbage collector is in some "experimental" state?

We lost 3 full days (2 people each) chasing a bug where garbage collector crashes the game!
(I posted the problem in another thread.)
We still couldn't find the problem. This isn't funny for us, this isn't funny for our investor. We have a game in production on Google Play, Apple App Store and Huawei App Gallery and soon we will have to deliver next update of our game, which is very problematic now with GMS 2.3.

I'm affraid in long term Yoyo will lose more of the serious users. I don't know exactly what is their goal, which developers they are aiming at to keep.
We like Game Maker very much, we invested so much in it and we hope that some day it can raise from the shadows of other game engines.
But strategy like this makes me ask myself if I would do my next project with GMS.

This is just my opinion. If I'm the minority of GMS users, then my voice won't count anyway.

I don't want to attack anyone here, everyone fights with their own problems and Yoyo is not an exception here.
I'm just reflecting on our current situation here and giving my opinion.
If some important pieces are missing (not tested well), then it is not suitable for professional use.
Sorry to hear that. 6 days of lost labour isn’t a joke. I don’t think us hobby users respect what that means. We can just wait until things fix themselves out. Time and money doesn’t matter. But if I lost 6 days of real work production that would be serious.

Mobile publishers seem to be a minority and it changes the most often.
 
@RefresherTowel & @kburkhart84 I don't agree with you.

I guess the vast majority of GMS users are using GMS more or less as a toy tool. That's ok, nothing bad with it.
We (very small team) invested 6 years of OUR LIFES developing our game with GMS.

Of course GMS is great product, but the users have rights to complain and they should complain, otherwise how would GMS improve if not listening to users' needs.

2.3 should not have been 2.3! It seems to me like a mistake. But maybe they had no other choice. I don't know.
I just know that developers who are serious about their product(s) and have releases in production are in certain circumstances pretty fuc*ed with GMS 2.3. It is like spit in the face.

At one hand I was very pleased that after migration to 2.3 we had to do just minor changes and the game ran almost without problems.
But on the other hand why Yoyo released a version with major GML changes and marked it as stable where for example garbage collector is in some "experimental" state?

We lost 3 full days (2 people each) chasing a bug where garbage collector crashes the game!
(I posted the problem in another thread.)
We still couldn't find the problem. This isn't funny for us, this isn't funny for our investor. We have a game in production on Google Play, Apple App Store and Huawei App Gallery and soon we will have to deliver next update of our game, which is very problematic now with GMS 2.3.

I'm affraid in long term Yoyo will lose more of the serious users. I don't know exactly what is their goal, which developers they are aiming at to keep.
We like Game Maker very much, we invested so much in it and we hope that some day it can raise from the shadows of other game engines.
But strategy like this makes me ask myself if I would do my next project with GMS.

This is just my opinion. If I'm the minority of GMS users, then my voice won't count anyway.

I don't want to attack anyone here, everyone fights with their own problems and Yoyo is not an exception here.
I'm just reflecting on our current situation here and giving my opinion.
If some important pieces are missing (not tested well), then it is not suitable for professional use.
Well, firstly I didn't say that people couldn't complain. I simply pointed out that the ratio between satisfied users and dissatisfied users is skewed towards dissatisfied when you read the forums, because dissatisfied users are the most likely to comment. And my comment was aimed at initial users who might seem put off by seeing complaints on the forum, not people in the middle of year long development cycles.

Secondly, I'm in the middle of a commercial project. I haven't upgraded to 2.3 and I won't until I'm finished. Precisely because it was completely obvious that the upgrade is a huge change and there's probably going to be some bugs and there's also probably going to be compatibility issues with the way I've implemented certain things, so there's literally no reason to risk it. To pretend that this is some sort of Achilles heel that only GMS suffers from is a bit disingenuous, at the very least. Unity often breaks things with their updates as well: https://lmgtfy.app/?q=unity+update+broke+project

I'm also not saying GMS is perfect and there's no way things could have been handled better. But in general it's a really nice development tool, there's no subscription cost, it's the best in class for quick 2D development and, unfortunately, it's worked on by a relatively small team. So I can cut them some slack and mitigate the problems for myself by doing things like limiting myself to proven stable updates.
 

Plura

Member
Well, firstly I didn't say that people couldn't complain. I simply pointed out that the ratio between satisfied users and dissatisfied users is skewed towards dissatisfied when you read the forums, because dissatisfied users are the most likely to comment. And my comment was aimed at initial users who might seem put off by seeing complaints on the forum, not people in the middle of year long development cycles.

Secondly, I'm in the middle of a commercial project. I haven't upgraded to 2.3 and I won't until I'm finished. Precisely because it was completely obvious that the upgrade is a huge change and there's probably going to be some bugs and there's also probably going to be compatibility issues with the way I've implemented certain things, so there's literally no reason to risk it. To pretend that this is some sort of Achilles heel that only GMS suffers from is a bit disingenuous, at the very least. Unity often breaks things with their updates as well: https://lmgtfy.app/?q=unity+update+broke+project

I'm also not saying GMS is perfect and there's no way things could have been handled better. But in general it's a really nice development tool, there's no subscription cost, it's the best in class for quick 2D development and, unfortunately, it's worked on by a relatively small team. So I can cut them some slack and mitigate the problems for myself by doing things like limiting myself to proven stable updates.
When developing for mobile platforms then you are heavily dependent on new SDK upgrades.
We are forced in that case to go ahead or we will get stuck with a version that does not support the new ios or android.
There is a significant difference between developing for mobile and developing for the desktop.
 
When developing for mobile platforms then you are heavily dependent on new SDK upgrades.
We are forced in that case to go ahead or we will get stuck with a version that does not support the new ios or android.
There is a significant difference between developing for mobile and developing for the desktop.
Yes, that is understandable. However mobile games made on 2.2 still work fine right? If so, then again, if you're in the middle of a commercial project, wait until 2.3 becomes more stable before upgrading, as I said before. This doesn't mean stay on 2.2 forever and have your game become obsolete due to changing SDK requirements. Just hold off on the upgrade for a bit.

(Of course, that point is moot if 2.2 does not actually run on mobiles anymore, but I don't believe that's the case).
 

SIG.

Member
(Of course, that point is moot if 2.2 does not actually run on mobiles anymore, but I don't believe that's the case).
For iOS, 2.2.5 doesn't work. For Android,...
Edit: It works for now, I believe (I'm not using a lot of Google services precisely because YYG is too unreliable) but that will definitely change. Whether it changes before or after YYG cleans up its mess is unknown.

Desktop users of GMS2 have a very different experience with the software.
 
Last edited:

Tornado

Member
@RefresherTowel: I won't "cut them some slack", they are here for us as we are here for them.
And in no sentence i was "disingenuous".
I already wrote that GMS is great product.
And for the first time I hear that the purpose of a Tech-Support forum is for serving "satisfied users"!

You have absolutely no idea what hell we went through to be able to publish on three mobile platforms (including Ads and IAP) and keep them running despite the fact that GMS has fantastic platforms export support.
There are times we sleep only couple of hours a day, because we have full daily jobs too and families.
You have to be insane to pull this off in your free time with a minimal team. (including weekends)

Our migration to 2.3 was planed for longer time now. We thought we can switch now to 2.3 stable(!) release as the pressure from mobile platforms is enormous at this moment!
(The bug with SSH to MacBook was fixed longer time ago, but we waited for the stable 2.3 release)
We had to find the sweet spot in timing by taking into account many factors.

I wrote in another thread I'll try to strip down our huge project that way that I can send them a small project with the Garbage Collector crash still happening.
I'll need around one working day for that and I hope that Yoyo will be able to find the problem.
 

Cpaz

Member
@RefresherTowel: I won't "cut them some slack", they are here for us as we are here for them.
And in no sentence i was "disingenuous".
I already wrote that GMS is great product.
And for the first time I hear that the purpose of a Tech-Support forum is for serving "satisfied users"!
...
I want to hone in on this point, because it's incredibly important.

The reason I take as much issue with YYGs lack of communication, be it their fault or their parent company's fault, that isn't important to the point.
Is because people rely on this piece of software for their livelihood. It's important to acknowledge that, yes, the majority of users are likely hobbyists, but that isn't their only audience.
In fact, it's arguably their least important audience, if I'm being honest.

I think we definitely need to be critical of how these updates are being handled, especially where that lack of communication is concerned. Which, fortunately, this forums seems to be doing already.

Now that I think about, was the existence of 2.3.1 even officially announced? Or was that leaked by someone?
 

SnoutUp

Member
When someone says they have no issues with GM, I must assume they only ever used Windows export. I only switch versions when I'm being forced by other platforms and it's always an extremely stressful experience. GM 2.3 is giving me a massive headache right now and I would've postponed upgrade for months, but needed SSH fixes for remote Mac builds & Xcode 12 support.
 

kroart

Member
I remember the old days when updates to GameMaker were released quite often. And when you send a bug report it was possible to get a fix in one or two weeks when an update with fixes arrived. Now it's too long to wait for the next release when you are just waiting for some fixes that a crucial for you.

Maybe YYG staff can switch back to more frequent updates with bug fixes? I think it will make life easier for GM users who just want to get fixes. Why not release GM 2.3.0.xxx update without any new feature but just with fixes?
 

Tornado

Member
When someone says they have no issues with GM, I must assume they only ever used Windows export. I only switch versions when I'm being forced by other platforms and it's always an extremely stressful experience. GM 2.3 is giving me a massive headache right now and I would've postponed upgrade for months, but needed SSH fixes for remote Mac builds & Xcode 12 support.
This is exactly why we had to switch to 2.3 too. We are not switching for fun. We have to deliver update of our game in the App Store soon, but our next iOS deploy is very questionable now.
(We also have to buy new MacBook as ours won't be supported on the new MacOS anymore :-O :-( , but that's another story)
 

Toque

Member
I think if steam publishing or problematic this would be a different conversation.

The constant changes on  side increases the challenge. Changes in OS. Changes in xcode. Upgrading obsolescent hardware to meet updated software requirements. Getting GM to cooperate.........

I bow to commercial mobile devs for their resilience.
 

kburkhart84

Firehammer Games
@Tornado I'm not sure exactly what part of my thoughts you disagree on. We both agree that complaining is valid if there is an issue, and we both agree that 2.3 was not ready to be released as "stable", which is a valid issue. Besides that, you didn't really say what you disagreed with me on so I have nothing else to go on.

I agree that it sucks that you lost time due to upgrading...but as has been said all over, even though this is a "minor" upgrade it really isn't, and so projects really should be finished in the previous versions, as that version still works AFAIK with the mobile exports and doesn't have issues with SDKs, etc...
 

kburkhart84

Firehammer Games
You are incorrect.
Does the 2.3 version suddenly work where the 2.2.5? Or how far back do you have to go for it to work?

I haven't done mobile since GMS1 so I'm not fully familiar with recent versions not working...I haven't seen any massive influx of reports of things not working until the recent 2.3 release they called stable though.
 

Tornado

Member
@kburkhart84 you agreed with Refreshing Towel about this complaining stuff, but I don't want to go that road again. If I misunderstood, I apologize, no problem at all!

Before migrating to 2.3 we were on the latest 2.2.x (I don't know the exact number by heart)
After migrating to 2.3 where new garbage collector is automatically enabled we got crashes on two places in our game. If we disable GC, then no crashes in 2.3.
But I have an extra thread for this particular problem so I don't want to duplicate the postings.
I'll just say we are still researching on this in our code, maybe we found a reason for one of the GC crashes. It maybe has something to do with calling instance_destroy() in Room End Event.
Can be that this is throwing GC off. But we are still researching on this.
Everyone interested in this particular problem and what exactly is garbage collected with new GC (Nocturne posted the exact list there) can follow that thread:
https://forum.yoyogames.com/index.p...-pointer-dereference-on-physics-object.79508/

It will be probably a long night again. I'll focus now more on that thread and on the solution. We are more under pressure as normal because our game is live on mobile platforms and at the moment big changes happen there.
I thank EVERYONE here for good discussion!
 
Last edited:

kburkhart84

Firehammer Games
you agreed with Refreshing Towel about this complaining stuff, but I don't want to go that road again. If I misunderstood, I apologize, no problem at all!
Yes, the post I agreed with was about the vocal minority being the ones to complain, while the majority of users are not here saying that they are in good shape. This in no fashion means that someone should not complain and search for help if they have issues(which obviously plenty of people do). The issue is that someone(Non .A) seemed to be getting the idea that the software is just buggy and unusable, and so Refreshing Towel mentioned the detail that just because its all complaints here doesn't mean that this is all there is, and then I agreed with that point. But of course, if you have issues, they should be brought up and hopefully resolved ASAP.

I'll post a thought I have in that other thread about your garbage collector issue. I'm guessing that maybe things get cleaned up differently than before and so you are trying to destroy an instance that already got cleaned up.
 
I recently took up trying to teach my little brother Game Maker and I actually haven't updated my GMS to the latest version. I didn't think our versions would be so different. Introducing all these new features in one massive update was a swing and a miss for me. Aside from the bugs (one time audio_play_sound just wouldn't work, it kept crashing and saying wrong number of arguments despite this not being the case) a lot of the changes are confusing and a lot to just get used to like the Asset Browser
 

SnoutUp

Member
While we're on this topic, I never understood why mobile exports are so neglected. Critical platform updates aside, we got (or had) outdated and poorly supported official extensions (things got better, but they still aren't ideal), lack of official ones for the biggest ad networks and important 3rd party services, somewhat lacking control of build settings and configs (best to just edit runtime files manually and update every time)... Main argument is that small team got no time for that (like it would take some intern more than 1h per extension a month to swap the new SDK in), but...

Imagine you have a great game development engine which is much easier to use than its competitors, has multi platform export capabilities and a low build size in a world dominated by mobile casual games... and you somehow don't market around the simplicity of building and MONETIZING your own Flappy Bird to attract people who are ready and able to invest in export extensions.

I would love to see more focus on exports and believe that everyone would benefit from it. #2cents
 

Toque

Member
While we're on this topic, I never understood why mobile exports are so neglected. Critical platform updates aside, we got (or had) outdated and poorly supported official extensions (things got better, but they still aren't ideal), lack of official ones for the biggest ad networks and important 3rd party services, somewhat lacking control of build settings and configs (best to just edit runtime files manually and update every time)... Main argument is that small team got no time for that (like it would take some intern more than 1h per extension a month to swap the new SDK in), but...

Imagine you have a great game development engine which is much easier to use than its competitors, has multi platform export capabilities and a low build size in a world dominated by mobile casual games... and you somehow don't market around the simplicity of building and MONETIZING your own Flappy Bird to attract people who are ready and able to invest in export extensions.

I would love to see more focus on exports and believe that everyone would benefit from it. #2cents
There doesn’t seem to be many mobile devs here. Once I looked at the featured game list for GM website. The first mobile game was 23 rd down on the list. Maybe 10% ( probably less) were mobile games. I should check again.
 

breakmt

Member
I have many troubles with 2.3 version and regret that I updated. I hope there will be bugfixes update soon! Just wish to finish current project and move out from GM.
 
Top