OFFICIAL GML Updates in 2019 (Follow up AMA)

Status
Not open for further replies.

drandula

Member
Just wanted to drop an update on 2.3. We're working hard to reach open beta but have nothing new to share at this time other than to say we still looking to hit Q1. Thanks.
This made me smile :)
Keep spirit high yyg-team! And don't let it be too much of crunch for you.
 

Lukan

Gay Wizard Freak
Thanks for the update.
Just release the beta, I promise we'll scream at y'all about any horrific bugs we find. <3
 

gnysek

Member
That's the worst part of open beta. I'm nearly sure, that 99% of reports/comments on forum will be about why they made something this way and not another, or that they can change something else too, instead of reporting real bugs. So it will be mainly screaming :p
 
That's the worst part of open beta. I'm nearly sure, that 99% of reports/comments on forum will be about why they made something this way and not another, or that they can change something else too, instead of reporting real bugs. So it will be mainly screaming :p
Sadly this may be true :(
But luckily they don't look at forums that much for suggestions and bug reports.
You have to file a bug/request for that :)
 

Tsa05

Member
I do wish there was a chance of YYG managing new features through an end-user accessible view for "changes going into a release."
I know there's a zillion reported things being crunched through in Mantis, but new, breaking changes are the anxiety, and also the dark spot.

The vibe I've been getting around the forums:
  • New format? Will that break how I'm saving/collaborating/backup/svn setup/custom tools?
  • New language features? Should I wait and structure differently?
  • Dynamically working with assets under these new changes?
Yes, people want to know about the shiny new things a bit, but "change" is the pain-point to manage and transparency in that area is--I think--more valuable even than hitting various hoped-for dates etc.
 

gnysek

Member
I had a moment of epiphany now. They not only adding new GML functionality and Sequences, but... they also need to do all of that for Drag and Drop too! Totally forgot, that GML also have that functionality and compared to GM8 and GMS1 it's not covering only a small part of GML, but most of it. That's a lot of work then, probably required to rewrite some of DND engine for new functionalities too.
 

Zhanghua

Member
I had a moment of epiphany now. They not only adding new GML functionality and Sequences, but... they also need to do all of that for Drag and Drop too! Totally forgot, that GML also have that functionality and compared to GM8 and GMS1 it's not covering only a small part of GML, but most of it. That's a lot of work then, probably required to rewrite some of DND engine for new functionalities too.
Just give up the useless DND
 

Cpaz

Member
Just give up the useless DND
There's probably why we haven't been given any public updates yet.

Something that I've been trying to remind myself with this update cycle, is that, for better or worse, yoyo games is still treating Game Maker Studio as a hobbyist tool that can be used commercially.

This update is a massive step in the right direction, but YYG as a company needs to be allowed to split the "hobbyist" from the "professional".

If that means splitting the application in two, or what, I dunno. But I do think that it's going to be worth keeping an eye on as they introduce more industry standard and industry catering features.
 
Last edited:
There's probably why we haven't been given any public updates yet.

Something that I've been trying to remind myself with this update cycle, is that, for better or worse, yoyo games is still treating Game Maker Studio as a hobbyist tool that can be used commercially.

This update is a massive step in tye right direction, but YYG as a company needs to be allowed to split the "hobbyist" from the "professional".

If that means splitting the application in two, or what, I dunno. But I do think that it's going to be worth keeping an eye on as they introduce more industry standard and industry catering features.
I think I have to support you there in your thought...

Currently if WE (users) suggest a feature that would not be that difficult to add in GML but would be a pain to make compatible with DnD probably YYG would be reluctant to add this news feature in.
And I understand the reason... their software started as a no-need-to-program DnD solution that would be great for non-programmers.
But as I also said the software STARTED like that... right now almost all the posts in the forums are related to GML questions and in the Discord forums it is the same thing. People even get encouraged to switch to GML because it is more powerful.

So I think the way we could go over this would be:

- NOT to create to separate softwares (this would be a pain to manage)
- BUT to make some things GML only. For example the new features they are called specifically "GML Update" so I think it should stay in the GML side.

I say this because sometimes the front-end stuff is harder to implement than the back-end stuff.
You have to deal with screen space, IDE design, graphics... the back-end is just (almost) reusable.
 
Last edited:

Cameron

Member
I agree with @xDGameStudios, and others, although I don't think this will ever happen. To give an analogy, it's like riding a bike versus riding a bike with training wheels. No one in their right mind expects you to be able to do everything you can with training wheel that you can with a bike that doesn't have training wheels, like going mountain biking, going off jumps, going faster, going around turns better, etc., The drag and drop (training wheels) should be there for newbies but at a certain point they should be expected to start using gml and any advanced features and/or larger project features should really be left for gml users exclusively.
 

gnysek

Member
The drag and drop (training wheels) should be there for newbies but at a certain point they should be expected to start using gml and any advanced features and/or larger project features should really be left for gml users exclusively.
That's how it was before GMS2, and that was good. And GMS2 allows to preview DND as code, which makes even easier to switch to real code.
Of course covering everything by DND is a very good thing, but with the manpower they have... Maybe it shouldn't be a priority.
 

Nocturne

Friendly Tyrant
Forum Staff
Admin
Would just like to say that with GMS2, adding new DnD isn't as much as a pain as it used to be in the days of 8.1 and 1.4 and it's fairly simple now (of course it all still needs specific QA testing, but that's another matter)... and contrary to popular belief it IS used quite a lot! And no, not all of the new GML features will be translated into DnD directly, but most of them will be useable in DnD... believe it or not, it's quite powerful especially with the "Function Call" node which lets DnD users call any GML function in Drag and Drop easily. ;)
 

gnysek

Member
as much as a pain as it used to be in the days of 8.1 and 1.4
Action Library Maker wasn't that bad, it was allowing adding custom code, passing arguments from textboxes, etc. :D

1583745651470.png

In both cases all is possible by "execute code" action, but it's good that for users who like DND, actions for 2.3 features will be also added, they will fill more comfortable.
 

gnysek

Member
Yeah, there will be a beta, but a closed one 🤪
I think it will be rather somewhere between 23rd and 31st of March, as "trying to hit Q1" sounds like they not sure if they will be able to fit in that date. And I hope that there will be no epidemic in Dundee, so they can work with no obstacles (at least programmers can work remotely, but it's always better to be in office).
 

drandula

Member
Yeah, there will be a beta, but a closed one 🤪
Nooo 🤣

Well it is just speculation when beta would start, though if they are still targeting Q1, then 31th March would be last day. This is where I have put my expectations, and earlier is just positive surprise :)
And if they don't get it in Q1, hopefully they mention it bit earlier than on 31th 😝 otherwise I would be bit frustrated
 

gnysek

Member
Of course, beta time is speculation as all depeneds on time needed to fix bugs and tests. So even YYG don't know where it will be. Probably with every closed beta release they are praying "please, no new critical bugs, we just fixed all we known!" :D They must feel more frustrated than us, as they believed it will take much less time.
 
I'm hoping we do get an ETA on this soon. Is this the thread to be for updates on 2.3? This is the first time I've actually been very excited for a program update. The functions are a game changer for me. I'm really excited to go through my project and neaten everything up.
 
Yeah, there will be a beta, but a closed one 🤪
I think it will be rather somewhere between 23rd and 31st of March, as "trying to hit Q1" sounds like they not sure if they will be able to fit in that date. And I hope that there will be no epidemic in Dundee, so they can work with no obstacles (at least programmers can work remotely, but it's always better to be in office).
My question still stands! They are trying to keep with the 2020 Q1 for the 2.3 beta release or for the 2.3 final release?
 

drandula

Member
I would assume it would be open beta, but that is only my guess.
Anyhow, eventually when they release beta or actual build, you finally get to tinker around with 2.3 : )
 

gnysek

Member
I'm just wondering what they do with Marketplace extensions. As all 1.x and 2.x extensions should work without problems in 2.3, but if you create extension in 2.3, there will be no way to run it in previous versions. The simplest solution for now seems to be to force users to use 2.3, or don't allow them to use marketplace in 2.2.x anymore.
But if they wan't to leave 2.2.x active too (after stable 2.3 release), that might be hard to maintain. There for sure will be persons that will complain about 2.3 (as it rises complexity) and will want to stick with 2.2 for now. I feel lot of complains.
 

Cpaz

Member
I'm just wondering what they do with Marketplace extensions. As all 1.x and 2.x extensions should work without problems in 2.3, but if you create extension in 2.3, there will be no way to run it in previous versions. The simplest solution for now seems to be to force users to use 2.3, or don't allow them to use marketplace in 2.2.x anymore.
But if they wan't to leave 2.2.x active too (after stable 2.3 release), that might be hard to maintain. There for sure will be persons that will complain about 2.3 (as it rises complexity) and will want to stick with 2.2 for now. I feel lot of complains.
I can see them working on conversion tools for things like scripts and make importing extensions a conversion process too.
Or, more likely, just tag them with versions and require all extensions to be manually updated to use 2.3 features. Kinda like how an extension can support GMS1.4 & 2?
 
I'm just wondering what they do with Marketplace extensions. As all 1.x and 2.x extensions should work without problems in 2.3, but if you create extension in 2.3, there will be no way to run it in previous versions. The simplest solution for now seems to be to force users to use 2.3, or don't allow them to use marketplace in 2.2.x anymore.
But if they wan't to leave 2.2.x active too (after stable 2.3 release), that might be hard to maintain. There for sure will be persons that will complain about 2.3 (as it rises complexity) and will want to stick with 2.2 for now. I feel lot of complains.
Right now market place offers 2.x and 1.x extensions as you said.. they could start to offer 1.x, 2.0-2.2.5 and 2.3 :) then it would be up to developers to port their assets to other versions. Like it is right now... if you make a extensions that uses 2.x features that won’t work with 1.x that’s just the way it is. Regarding backward compatibility, they said that old extensions will still work on 2.3 so or they will auto convert to a new format or there won’t be a new format at all (on the extensions back-end).
 

matharoo

Udemy Instructor
Cherish your days with GML1. You will remember them,
once they are in the past... How you used to create fake
classes, how you used to make fake structs with arrays
and enums, and oh how you used to make "functions"...

" Oh hey, I wanna create a function that should deplete an entity's HP. Let me create a new .yy and .gml file for that "
 

FrostyCat

Member
The size of the new features list should have told you a delay is inevitable. The more things go into a single update at once, the more ways things can go wrong in a way that sends it back to the workbench. I wouldn't be surprised if it takes until Q2 or Q3 for the changes to go live.

Let's look again at the list of additions:
  • Chained accessors
  • Inline functions
  • Multiple scripts in a single file
  • Lightweight objects and new operator
  • Exceptions handling
  • Upgraded garbage collector
Now consider places where things are most likely to go wrong (bolded in the list above):
  • Inline functions have a different scoping requirement than pre-update GML, where code inside runs under its own scoping bubble. This is different from the current "pull-up" behaviour that local variables have, so the assumptions that the current compiler's parser can use for this would no longer be valid.
  • Multiple functions in a script is a breaking change from GMS 2.2, and will require automated conversion for existing projects. It's easy to make mistakes here that would broadly vandalize existing GMS 2 projects everywhere.
  • Lightweight objects and inline functions both use the same bracing syntax as control structures (e.g. loop blocks). The IDE's syntax checker and the compiler's parser will need to be updated for this, especially all the corner cases where they're next to or mixed with existing control structures.
  • There are now 2 distinct behaviours for assignment operators in a function, one is running without new and the other is running with new. Again this is a source of potential confusion that the IDE's syntax checker and the compiler's parser will have to disambiguate.
  • There are lots of places in GML where a single cover-all error message would once have covered, but now require distinct and rerouted handling because of exceptions. Finding and testing all those interactions take time.
If even one of the above of the above goes pear-shaped, it's back to the drawing board. Is it all that unreasonable to expect that the internal builds are sent back more often than not?

While hindsight is 20-20 and I think YoYo should have pushed it in smaller increments (just the chained accessors alone would have made me quite happy), what I think has been done is very likely to have been done. As a community, your choice is now between demanding that it come out early and vandalize your projects, or being an adult and let all the new features learn to behave.
 

SSJCoder

Member
The size of the new features list should have told you a delay is inevitable. The more things go into a single update at once, the more ways things can go wrong in a way that sends it back to the workbench. I wouldn't be surprised if it takes until Q2 or Q3 for the changes to go live.
yea same !!
I think YoYo is overwhelmed, got a lot of heat in their hands!! I'd say y'all gotta wait some time guys xD
(or the features will probably start out glitchy - but will get fixed I'm thinking)
 
The size of the new features list should have told you a delay is inevitable. The more things go into a single update at once, the more ways things can go wrong in a way that sends it back to the workbench. I wouldn't be surprised if it takes until Q2 or Q3 for the changes to go live.
I think this is flawed thinking. The end user shouldn't be responsible for predicting whether something is complex enough to cause delays. Yoyo are the ones that know their team's size and capacity, not the end user.

While I agree that we shouldn't be annoyed with YoYo games, and I really like that the community is patient and understanding rather than demanding of them, remember this isn't a free product. I think it's ok to want occasional release window updates when something is delayed substantially.
 

gnysek

Member
Yeah, the end user doesn't know what they finally included in this version, so can't tell what is delayed and what not.

Remember, they not only added GML changes (those mentioned in this topic and not mentioned), there are also Sequences, resource tree upgraded, data format changes, DND actions for nearly everything new, gml functions for sequences too, and there might be other small IDE improvements/addons (I don't have access to closed beta, but a friendly squirrel :squirrel: said, that "I didn't noticed some changes, until get back to 2.2.5, and started to miss them. Nothing big, but sometimes a small change make a big difference.". My hype just raised, so sad that I don't know the details :/ ).

So, even if it seems that they have some delay, it's delay for GML updates only, and for whole 2.3 there's probably no delay at all, as they decided to put much more that they initially thought year ago, and that of course moved the deadline. The issue here is that they didn't updated roadmap to include those "not mentioned" updates too, with new dates.

Also, take in mind, that even if YYG completed features for 2.3, if closed beta users said that something isn't intuitive, and need changes, they even could rewrite some part of sequences IDE or GML changes syntax, or even add something new not planned before - as that's for what closed beta was, and why there's a small amount of most experienced users. Delay might be not because of bug fixes, but because of adjustments and new ideas, so we will be more happy with what we get in stable version. And even leaking any info (for example sometimes they hint something in comments on bugtracker, mentioning something is fixed in 2.3, when bug was for 2.2.x), might not be accurate, as they might have changed their mind even today.

I believe that after open beta will start, we will understand why it taken so long, and we gonna find some other changes/fixes not mentioned anywhere :) There will be for sure a long "what's new" changelist.
 

FrostyCat

Member
I think this is flawed thinking. The end user shouldn't be responsible for predicting whether something is complex enough to cause delays. Yoyo are the ones that know their team's size and capacity, not the end user.


While I agree that we shouldn't be annoyed with YoYo games, and I really like that the community is patient and understanding rather than demanding of them, remember this isn't a free product. I think it's ok to want occasional release window updates when something is delayed substantially.
How is having a sense of self preservation and street smarts "flawed thinking"?

I never said to let YoYo off the hook for overloading GMS 2.3 with new features. In fact, from what I've said thus far, I pinned the responsibility of that squarely on them:
I think YoYo should have pushed it in smaller increments (just the chained accessors alone would have made me quite happy)
GMS 2 is a paid product, and I too expect more professional handling of this. On that point, I agree.

But while it is YoYo's responsibility to make sure GMS 2.3 pans out as they promised (time-wise and features-wise), it is YOUR responsibility to decide how you'll act based on how you think things will pan out. It's the end user's responsibility to decide whether they should continue to wear that hat the same way.

Initially I too had high hopes for GML 2020, and arguably tried to explore it harder than anyone else outside YoYo's core dev team (given my lion's share of the AMA). But the facts have become apparent that neither Q4 2019 nor Q1 2020 are likely release dates for GMS 2.3. I've stopped theoretical exploration of the new elements, and redirected short-term efforts into other areas of my off-work development life. Everything is still sailing smoothly for me thus far. That's how I chose to respond.

Living in this world means making choices based on the interests of yourself and things under your custody. You can either let yourself be a victim first and try to litigate later, or you can decide to protect yourself pre-emptively. And when a default is likely based on observable facts, not protecting yourself is YOUR FAULT.

If you buy an investment, it's the vendor's responsibility to disclose the details as honestly and accurately they can, but it's YOUR responsibility to decide whether to buy, sell, or stay out. Ultimately both parties are human and will make mistakes, intentionally or not, especially when predictions are involved.

Whether your investment is in units of money, time, emotion or work, the dynamics are still the same.
 

SSJCoder

Member
I think this is flawed thinking. The end user shouldn't be responsible for predicting whether something is complex enough to cause delays. Yoyo are the ones that know their team's size and capacity, not the end user.
shouldn't be, but we know yoyo, not to mention this is already delayed from last year (2019).

there's a lot of things that "shouldn't be", but are. this includes many paid products out there, in fact, paid products tend to be worse than non-paid, yes that's right, because paid products work differently, they work more as a business rather than a product, so, even if their product isn't all that great but sells well - the business succeeds.

now, while I could go on all day talking about this, the point here is simple, given yoyo's history, and the current evidence presented, from someone's perspective who understands a little bit about that, this will probably just get delayed further out - no point expecting a full release with all features exactly at x date, unless you still want to believe in that, which is fine, of course.
 
Maybe we are also getting a few more features as well from the "Planned Features" section. HTML5 web sockets came out of nowhere in one of the previous updates. So maybe we will get a few more surprises.

I am not expecting April 1st or sooner for 2.3, just so I don't get disappointed. If it happens on or before that, then I will be surprised.

I have a feeling people are going to get super antsy though if it doesn't come out on April 1st/2nd.

Also, you have to consider that there is a virus going on that could severely slow down or halt development.
 

Toque

Member
Maybe we are also getting a few more features as well from the "Planned Features" section. HTML5 web sockets came out of nowhere in one of the previous updates. So maybe we will get a few more surprises.

I am not expecting April 1st or sooner for 2.3, just so I don't get disappointed. If it happens on or before that, then I will be surprised.

I have a feeling people are going to get super antsy though if it doesn't come out on April 1st/2nd.

Also, you have to consider that there is a virus going on that could severely slow down or halt development.
They might be able to get some staff to work from home but there will be some disruption as the pandemic evolves.

I can’t see them releasing until it’s ready.
this could delay things for quite some time asthe uk shuts down.
 

Mr. RPG

Member
It's disappointing that nothing really seems to have changed for the better for YoYo Games after the Playtech acquisition. :( YoYo Games is still really, really slow at releasing updates and consistently pushes them back months and months past schedule. Is their team still as small as it was during the Studio 1.x days?

I'm glad that 2.3 is coming, but it was first announced like almost a year ago now.
 
Last edited:

Mr. RPG

Member
Maybe we are also getting a few more features as well from the "Planned Features" section.
I really hope so. The room/code editors really need some updates. They haven't had any major changes since the release of 2.0 and that was several years ago.

That being said, everything they've announced for 2.3 is already amazing and more than I thought would happen. It's the biggest update to GameMaker since 2.0.
 

gnysek

Member
Looking at amount of changes they made in last year, I think they may complete roadmap and additional features in 1-2 years from now.
It's disappointing that nothing really seems to have changed for the better for YoYo Games after the Playtech acquisition.
I think that it was better before in some ways :p Sadly, that's how "big companies" which are on stock works - they can't talk much about internal stuff. But, at least they pumped enough money to drop Delphi IDE which was a big step! Looking to one of photos they poster last year, there's 2-3 times more people working in YYG than in 2012 (GMS 1.x release), but remember, that they also added a lot of exporters during this time, made a proper QA team, and helpdesk/support team is much much bigger now, as people are asking lot of requests :p So, lot of things changed for the better in fact.
 
Well I do completely understand the delays. Lots of features (even hidden/unknown ones) and the pandemic problem we are all facing... just wish we had some updated news!
 

matharoo

Udemy Instructor
Imagine if they kept us in the loop through regular news updates- even if it were just simple daily tweets- we'd be a lot more patient. 😾
Imagine...

Twitter | @YoYoGames

15th March 2020
11:44 AM:
Whoops, there's a bug in the Mac runtime! Gotta fix!
04:40 PM: Jerry fixed it!

16th March 2020
10:17 AM
: Wait that bug is back. Where the heck is Jerry

Everyone in this thread:

1584453485986.png
 
Last edited:
How is having a sense of self preservation and street smarts "flawed thinking"?
I don't think you were trying to make any profound point, other than to show everyone that you are quite a clever fellow, by demonstrating that you know all about the software life cycle/iteration planning etc.

All I did was point out that the average user might not have the same level of knowledge as you do, and shouldn't be expected to. I didn't say it was flawed thinking for you, THE FrostyCat to have predicted this. I think it was incorrect for you to imply that the average customer should have too.

I still have hope that the update arrives this month.
Me too brother, me too
 
Status
Not open for further replies.
Top