• 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.

Question - IDE Has GM2 finally caught up to the stability and performance of GM 1.4?

E

elsi11

Guest
Hello friends!

It has been some time, and I have my annual GM itch again. Last time I was here, I made a nice script for inputting and converting strings or something, and we discussed how a bug regarding resizing windows got re-introduced in (then) latest version of 1.4. Good times.

Since then, I had my account deleted because I didn't use it for a long time. But I support the initiative to push the forums towards GM2, and clear people's old private data and whatnot. Though, from a design perspective, I would like to see a "please don't delete my account" toggle in the settings.


All this being said, let's get on with the question at hand. When GM2 came out, people said that it had some quirks and bugs, and that it was inferior to GM 1.4. From the performance perspective, it was also vastly heavier on the hardware. It even missed or changed some things that were better in GM 1.4.

It is the end of 2019 now, and I think sufficient time has passed to fix all the stuff and improve the program. I would like you guys, especially people advocating 1.4, to tell me whether anything has changed and which version do you prefer and why.

For me, I am seriously considering getting the GM2, but if it's not up to par, I will be just as happy to make my next project in 1.4.

Please advise!

I'll just tag some people I remember :p
@Yal @FrostyCat @TheouAegis @Simon Gust @RangerX @Nocturne

Tnx!
 

Simon Gust

Member
I don't have GM2, I would make any new Projects with GM2 though (or the latest software in general). Say if new pc software Comes out and it isn't compatible with gm1.4 games, then you're screwed. You could make the game with 1.4 as Long as you port it over to GM2 before release I guess.
 

GMWolf

aka fel666
Since then, I had my account deleted because I didn't use it for a long time. But I support the initiative to push the forums towards GM2, and clear people's old private data and whatnot. Though, from a design perspective, I would like to see a "please don't delete my account" toggle in the settings.
That was because of gdpr or something...

I would like you guys, especially people advocating 1.4, to tell me whether anything has changed and which version do you prefer and why.
I prefer GMS2 because of the tilemaps and better GML.

But I still do prefer the GMS1.4 IDE.

That pretty much sums it up. Is it worth getting over 1.4, yes!
Is it worth the upgrade if you already have 1.4? Maybe. I don't know. If you are doing this professionally then definitely yes.
 

JeffJ

Member
I prefer GMS2 because of the tilemaps and better GML.

But I still do prefer the GMS1.4 IDE.
This is pretty much exactly how I feel.

Some genuinely cool features in GMS2 that I would miss going back to GMS1.
The IDE, however, remains atrocious and a daily pain to work with. Yes, even compared to GMS1. Even after 2 years of constant use. It's not something you get used to, it's something you realize you just have to live with.
 

FrostyCat

Redemption Seeker
On the IDE front, I can attest that GMS 2's stability is absolutely not comparable to GMS 1.4.

I've been using GMS 2 for the GMC Jam on 4 occasions, and each time I had to give the GMS 2 IDE "hourly breaks" because it was leaking RAM during regular use. If I don't, it fires up my MacBook Pro's cooling fans at 1GB and crashes circa 2GB-3GB. I also noticed that projects checked out as Git working copies sometimes fire up the hard drive but not unversioned projects, presumably from the automated file scanner trying to scan the entirety of the .git directory. And when I close the lid with the IDE on and put the system to sleep (a very common use case for me as I do some of my development on public transit), the IDE is gone by the time it wakes up. None of this was a thing when I was using GMS 1.4.

I also have a huge issue with the waste of screen real estate in the new workspaces system. It was designed with 4K monitors in mind and then imposed on everyone else. I could not use it without excessive scrolling on my 1280x800 laptop screen on the go, nor on my 1920x1080 side monitor at home. This could be ameliorated by altering the workflow to rely more on tabs and keyboard shortcuts instead of workspace windows and clicks, but I'm still investigating the details of that.

While the GMS 1.4 IDE has the benefit of familiarity and stability, starting new projects on GMS 1.4 may soon be a preference you can not practically afford. While I would expect the plain Windows export to continue working for some time, the other exports have either been left behind already or will be within the next 1-2 years. In particular, if your project has to do with iOS or Android, starting with GMS 2 is a must because of new app store policies that have disqualified GMS 1.4's exports.
 
H

Homunculus

Guest
The IDE, however, remains atrocious and a daily pain to work with. Yes, even compared to GMS1. Even after 2 years of constant use. It's not something you get used to, it's something you realize you just have to live with.
I'm happy to know that I'm not the only one who didn't experience almost any improvement in terms of usability and workflow even after a couple of years with the new IDE (in fact, it's the opposite). I started with an open mind thinking I'd get used to it, but so far that isn't true.
As FrostyCat pointed out it's a pain to use on a laptop or an average sized monitor, but that's only part of the problem for me. I just don't get why having a workspace with tens of windows connected with lines all over the place is considered good, we are not building a flowchart. Also, the little animations when opening / closing panels (like events or variable definitions), with the other panels shifting following the animation, and possibly the workspace position moving as well, is something I just can't get used to. It's just eye candy that I could very well do without most of the time (thank god scripts and pieces of code can be set to open on their own tab).

That being said, I still recommend GMS2 as in all other aspects I consider it to be an improvement over GMS1, especially the changes and additions to GML, room editor and tileset management. Moreover, even though I don't like the IDE workflow in general, there are A LOT of customization options, which is definitely great.
 

JeffJ

Member
You're far from alone. Just take a quick look at this thread as an example:
https://forum.yoyogames.com/index.php?threads/classic-style-ide.25206/

And yeah, it's the same for me. My workflow has directly suffered, and even several years later, I am still slower and less productive (by a significant amount) in GMS2 than I ever was in GMS1. It's downright crippled my workflow and the joy of usage. For me it's exactly the same stuff you mention; the chains, the automatic movement of workspace, the animations... All of it is just in the way, taking up too much space etc. - horribly inefficient and works against the user rather than with or for the user.

I would love to have all the features and such of GMS2, but with GMS1's IDE. Even the room editor of GMS2 still has some disadvantages over GMS1, particularly in terms of ergonomics (such as forced usage of both mouse AND keyboard for operations where mouse alone was previously sufficient).
 
E

elsi11

Guest
So, basically, we all prefer 1.4, but you suggest switching to GM2 regardless because it's newer and will be more supported.
I dunno, I think I'll google a bit more about pros and cons, and decide eventually...

Can you tell me whether GM2 has the option to change savegame location? I always despised when games made poo in some random hidden directories. I understand it's a matter of permissions, but it still rubs me the wrong way that it couldn't be done in the game directory so you could make portable games for UBSs.
 

GMWolf

aka fel666
So, basically, we all prefer 1.4, but you suggest switching to GM2 regardless because it's newer and will be more supported.
If you read our comments again:
We prefer the 1.4 IDE, but overall GMS2 is a better engine because of the new room editor, layers, TILEMAPS (I love those) and the new GML features.

My view is that, if you are just gonna mess around for a bit and you already have 1.4, GMs 2 might not be a good investment.
If you do a lot of gamedev and want tilemaps and layers, then GMS2 could be worth it.

Can you tell me whether GM2 has the option to change savegame location?
You could even in 1.4
But 2 now has the option to turn off the file system sandbox so you can do it better (without asking for a save location from the user)
 

rIKmAN

Member
I'll have to go against the general feel of this thread and say that for me it's the opposite: I hate the GMS1.4 IDE and don't miss it a single bit, in fact having to go back to it to test my extension a couple of weeks ago just made me realise how archaic the thing is.

I wouldn't say I "love" the GMS2 IDE and there are definitely some improvements to be made, but I don't have any major issues with it and I have no real problems with the workspaces at all - I'd take it over the 1.4 IDE 100/100 times without hesitation.

I must say though that 2 things that most certainly play into my feelings being so different to those in this thread:

1) I have a 4K main screen and 2 x 1920x1080 (one either side) - so screen real estate isn't an issue for me personally so I don't find the workspaces a problem.
Before I got the 4K screen it also felt fine on the 1920x1080 screen, but I can see how going lower (ie. FrostyCats 1280x800) would feel really cramped, but then I think I'd personally find most things really cramped at that resolution and not just GMS2.

2) I only used 1.4 for ~2mths before GMS2 came out so I have no long term history with it: no ingrained habits, no nostalgia for it and no fondness for how it did things built over years of growing up with it or using it for a long time before switching to GMS2.

I realise that of course not everyone is in the same situation, and the workspaces really are like marmite both here and in previous threads discussing them all the way from when it first came out. I'd have no problem with YYG adding functions to appease people who hate workspaces but it's been said that that's not going to happen so I won't flog a dead horse - just wanted to give another viewpoint from someone who actually enjoys using GMS2.

@elsi11
Download the 30 day trial and try it for yourself to see how you like it - there is no better opinion than your own via hands on testing.
 

Mike

nobody important
GMC Elder
While I'm obviously going to be biased, I have to say, the vast majority have none of these issues, and have found GMS2 to be far more productive.
I've said it before, GMS1.x was a horrible place to be. Massive amounts of overlapping windows meaning you simply couldn't have that much open at once without losing track of things.

GMS2 allows you to have open many things at once, without the mess. Yes, we should have tested more on laptops, but there are loads of shortcut keys to help you navigate, and you can scale the whole UI down using the DPI override.

On top of all this, the new UI allows it to be easily ported to other platforms - like OSX, or even the Raspberry Pi for that matter. Something that was utterly impossible with the old UI. That was a key design choice.

Still, none of this matters. The old UI won't be coming back. Before I left, there was talk of taking a pass over the UI and doing a "thinner" skin. I've no idea if that's been dropped, delayed or what. But really, that's all it needs.

@JeffJ, you've spent years moaning about the UI. There's always gonna be folk like yourself who don't like change, but I hate to break it to you, they/you are in the minority.... Most find it a nice place to be, and monumentally better than 1.x.

Personally, I can't go back to 1.x... it's horrible, clunky, and a mess. While there's certainly stuff in the image editor I'd love to have, those actually take a long time to do, and when I was there, were never given the time to do it. But everything else is so much better. The text editor is brilliant, the full screen tabbed editors. I honestly can't understand why anyone would want to stay in 1.x.

Embrace change! Once you actually do, the workflow in 2.x is Sooooo much better.

So, basically, we all prefer 1.4, but you suggest switching to GM2 regardless because it's newer and will be more supported..
This isn't strictly true. A handful on this forum don't like 2.x, but this forum is a tiny percentage of actual users. While some louder folk on here can make it look like it's everyone, it's really not.

The only way to find out is try the demo for yourself. Make something in it, see how it goes...
 

JeffJ

Member
@JeffJ, you've spent years moaning about the UI. There's always gonna be folk like yourself who don't like change, but I hate to break it to you, they/you are in the minority.... Most find it a nice place to be, and monumentally better than 1.x.
EDIT: Retracted some petty personal stuff. Want to focus on the actual important things here.

Every time someone has any complaints about the UI of GMS2, you claim that it's a vocal minority, but what do you actually base that off? Is it telemetry? Surveys? What actual data ensures you to this level, even 10 months after having left?

All I have is my own experience and satisfaction/dissatisfaction comparatively to GMS1, and then the rest of the forum members. Clearly, the vocal minority would like to be enlightened, but then show us the data.

I would also love to hear you actually address some of the complaints rather than just brush them off as moaning.

Specifically:
- Forced chaining. Why not allow us to open/close specific parts of the chain? I asked this several times way back on the first Zeus stream, and never got a reply. What is the reasoning or benefit from this?

- Why are there vertical lines between each chain to resize? Why not simply allow resizing by dragging the edge of the window itself? And why have those vertical lines go across elements below and above?

- Why not allow the user to drag a single self contained code editor to another display without having the entire app wrapper around it?

Everything about the GMS2 interface just takes up much more screen space than it needs to.
We can already enable overlapping, which means we are right back to what you loathe about GMS1, but with all the disadvantages of the chains and everything else. What is gained here? I really want to see what you see.
 
Last edited:

curato

Member
GMS2 allows you to have open many things at once, without the mess. Yes, we should have tested more on laptops, but there are loads of shortcut keys to help you navigate, and you can scale the whole UI down using the DPI override.
As a laptop user, once you get used to the shortcuts GMS2 is better and everything stays open. The only knit I would pick is that if you double click on an object it centers on the main window for the object even if you have your code tabs opened up. I wish if the code widows where open it would try to center on that. A lot of times I have to drag the view over to see long lines of code or if I want to see the find/replace interface.
 

Cameron

Member
Guys lets not turn this into a flame war.

@elsi11 GMS2 is the way to go now a days and this is coming from someone who spent a lot of time with GMS1.4 and doesn't have a 4k monitor. Is GMS2 perfect, absolutely not, but it's alive and GMS1 is dead in that it is not being updated. GMS2 however is still being improved and while yoyo was being rather stubborn for a great while in listening to their base they are doing that now and I feel they have corrected course and right the ship.

There is a very useful workflow setup you can have for yourself that no one is talking about and that is to use gmedit by YellowAfterLife. This allows you to avoid the gms2 ide for coding, where a lot of peoples gripes are and still take advantage of the gms2 room editor and layering system. I use gmedit for all of my coding, it's lightweight and elegant and is absolutely the way to go.

Another thing that hasn't been mentioned yet but in a few months, for version 2.3 on the roadmap, yoyo is vastly updating gml to have some very nice improvements. I'm not going to go into detail on them, there is a blog on yoyo's website about it and a link to it in the announcement section of this forum but basically it's bringing a lot of useful changes that are going to be a paradigm shift in terms of managing large scale projects, which was probably gamemakers biggest weakness.

Basically, gms1.4 is the way to go if you are feeling nostalgic and refuse to move on and gms2 is the way to go otherwise.
 

Mike

nobody important
GMC Elder
What a lovely reminder of how you used to interact with customers and community members alike. It's no secret that a lot of your personal decisions were, and still are, very unpopular, not only with the users, but even with your own old team. Some of them have already been undone after you left, since you were the only one keeping those decisions from being made, and now, thankfully, the team are actually free to make properly thought out decisions that tends to land positively with the userbase.
Ha! You have no idea how much tongue biting was going on while I was there - not just by me. Something I no longer need to do. There were lots of changes in the pipe before I left, and you seem to think I decided everything on my own. YoYo is a team, and things were discussed and agreed upon. We were always open to changing things after feed back from a reasonable number of users - just not those who continually bitched about the same thing over and over again, and even when they were in the minority, refused to accept things and move on.

You keep talking about the vast majority and such, yet we have never seen any numbers or any significant amount of posts backing up those claims.
This is because you think that everyone on this forum, is everyone using the product. This forum is a tiny percentage of people using the product, and even on here, there's only a small number that really want the 1.x IDE over 2.x...

You may claim minorities, you may claim we are just "moaning" (again, such a lovely way in which you interact),
Fortunately... I don't have to give a thundering [insert funny expletive of your choice here] what you think anymore. :rolleyes:

Maybe, just maybe, people don't see the same brilliance in chains that you - and only you - do. Show me one example of someone praising the chains. Because I have yet to find it.
Funnily, chains were up for discussion for changing, but James never gave us any time to even think about going back over old ground. He was constantly pushing for new features, not updating things. This is one of the (many) reasons I left.

Interestingly.... the "original" design of the chains made much more sense when seen in context. If you want to see what the "original" design was, go into preferences->"Text Editors" and untick "Open even scripts as tabs int he same window" and "Open linked scripts as tabs in same window" (so nothing is ticked).
Each event would then open in it's own window, and a link from the event to the script.... This was mainly for the long term view for the debugger, as you got a "tree" of scripts.
But... when folk started using it, it quickly became apparent that everyone preferred "tabs" to the multi-window approach (and we then added the full screen tabbed editor, but it was too late to actually change at that point. The debugger obviously changed as well. I've always been open to changing the IDE based on good feedback. When released, it was always supposed to be a starting point. While we spoke to some of the bigger devs, it's all bets were off as soon as users got their hands on it. But going back to 1.x style was never on the cards, but cause - and I may have to check my notes, it was awful.

I personally like having the chain of scripts open, and being able to pan through an entire script call chain with the mouse...*shurg*

But perhaps I should be careful with my moaning - wouldn't want you to go abusing those admin powers again and go against all protocols.
I'm no longer a admin or a mod... so do whatever you like, because I'm also not the one who has to deal with you. :D
 

JeffJ

Member
just not those who continually bitched about the same thing over and over again
Again... What a lovely way of putting it. And you wonder why they didn't want you at the conferences.
Could it be that people keep "bitching" about it because it needs to be improved?

This is because you think that everyone on this forum, is everyone using the product. This forum is a tiny percentage of people using the product, and even on here, there's only a small number that really want the 1.x IDE over 2.x
Again, based on what data? Telemetry? Surveys? Magic ball? Show us - then you would really be able to make it clear how much we are in the minority. Surely the data speaks for itself?

Fortunately... I don't have to give a thundering [insert funny expletive of your choice here] what you think anymore
For someone who doesn't give a thundering expletive, you sure seem to be spending an awful lot of energy explaining how little you care.

Each event would then open in it's own window, and a link from the event to the script
Honestly that sounds just as bad, just in a different way.
But you still haven't addressed some of the basics - why not simply allow closing of individual chain elements? Or opt out of chains altogether? Since you already allow overlapping, adding this would appease almost everyone - GMS1 and GMS2 people alike.
 

Mike

nobody important
GMC Elder
Again... What a lovely way of putting it. And you wonder why they didn't want you at the conferences.
Could it be that people keep "bitching" about it because it needs to be improved?
Really? You get offended by that? Wow... just as well I'm being nice. :p
Don't know where you're getting your info from - I was the one who was always stuck in a room talking to devs for the whole week, finding out how things were going, and what they needed. At no point did any of them say "I want 1.x back!"

Again, based on what data? Telemetry? Surveys? Magic ball? Show us - then you would really be able to make it clear how much we are in the minority. Surely the data speaks for itself?
Based on talking to people, and yes... the early questionnaires. Clearly, I can't show anything. I not only don't have access to the data, but it also belongs to YoYo. I don't really care if you believe me or not, I'm telling you the truth, it's totally up to you have you choose to think I'm lying - I really don't care.


For someone who doesn't give a thundering expletive, you sure seem to be spending an awful lot of energy explaining how little you care.
*shrug* It's a hobby..... besides, you shouldn't be left to rant without a counter point, and YoYo tries hard (too hard really) to be nice to everyone, no matter who that is.

Honestly that sounds just as bad, just in a different way.
But you still haven't addressed some of the basics - why not simply allow closing of individual chain elements? Or opt out of chains altogether? Since you already allow overlapping, adding this would appease almost everyone - GMS1 and GMS2 people alike.
No, it wasn't great in the end. It's something that sounded good on paper, and walk through's, but as soon as we used it in anger, it became obvious that it wouldn't work for more experienced devs. It was kept because we thought it "might" have worked well for beginners, especially with D&D.
In the end, no... it didn't work, which is why it wasn't the focus and has never been pushed.

As I said before, changing/removing chains was up for discussion - everything was, but we were never given the time. But to do it, it would need proper design, testing, focus groups - the works. That's a big investment in time and money. And while it "might" be the way to go, who knows if YoYo are ready to commit to that, or now view it as a minor issue - I've no idea now.

However... even if I'd gotten chains removed, I wouldn't have allowed massive window overlapping. That was the worst part of GMS1, and I'd never have brought that back.
And as to appeasing everyone.... no... only you. This is a huge amount of work/effort/money, just to keep you and a few others from bitching.

I'd also just like to say on the stability of GMS2. I've not had any issues, either with memory or performance. I've been running the same IDE for a few weeks now. No issues at all. GMS1.x still has stability issues...

So clearly your mileage may vary.....
 

JeffJ

Member
Really? You get offended by that? Wow... just as well I'm being nice. :p
Don't know where you're getting your info from - I was the one who was always stuck in a room talking to devs for the whole week, finding out how things were going, and what they needed. At no point did any of them say "I want 1.x back!"
I love how you genuinely perceive it as being nice continually saying that people with complaints about the UI are "bitching" or "moaning" or other such lovely terms.
Alright, so no data, but anecdotal evidence based on developers you talked to. How is that any different from developers I have talked to? We could always compare numbers, but at that point, neither of our findings seem very scientific or accurate. Again, the word anecdotal comes to mind. But you claim yours as majority and fact.

Based on talking to people, and yes... the early questionnaires. Clearly, I can't show anything. I not only don't have access to the data, but it also belongs to YoYo. I don't really care if you believe me or not, I'm telling you the truth, it's totally up to you have you choose to think I'm lying - I really don't care.
So... Again... Just as above. It would be another matter if you said that it's your impression based on the specific developers you spoke with, but there really is no hard data, then. So we're both guessing or feeling?

YoYo tries hard (too hard really) to be nice to everyone, no matter who that is.
Something you never seemed to grasp.
It says a lot that all the countless hours of my life I've spent trying to put the right words on what exactly is wrong with the UI is literally just "ranting", "bitching" and "moaning" to you. Just because it's not what you want to hear doesn't make me a troll - as you will notice, I am not the only one here with these complaints. The fact that they date back to the very beginning of GMS2 and are still here should say a lot. They would be even if I wasn't around.

However... even if I'd gotten chains removed, I wouldn't have allowed massive window overlapping. That was the worst part of GMS1, and I'd never have brought that back.
And as to appeasing everyone.... no... only you. This is a huge amount of work/effort/money, just to keep you and a few others from bitching.
Again, this says a lot about your mentality. You don't like it, so you don't care how many other people do. And especially with no data to back it up, it makes it even more out to look like you're just dead set on your way, and nothing else. I love the fact that I can at the very least have the windows overlap, and I would hate working in GMS2 even more than I already do if that wasn't an option.
You keep going on about how it's only a few, but there's a lot more here (not just in this thread) speaking against GMS2's IDE than for. Has been that way for years now, and it's not really changing.

Look, GMS2 has some really cool features, and like I said, there's a bunch of stuff I would sorely miss if I went back to GMS1. But the IDE has some inherently bad decisions that directly impacts workflow in a negative way for a lot of people. If it's too late to change that, then that's one thing, but brushing it off as just a minority with no data to back it up, when there's a clear majority of people arguing against GMS2's IDE on the forums is kind of bold - just be up front about it.

Giving us an option to opt out of the chains would go a LONG way for a lot of us, especially combined with the ability to overlap. Even just the ability to close specific parts of a chain only would go a long way. Add then the ability to disable auto panning of the workspace, and you've actually got something that could work. It doesn't take a lot.
 

GMWolf

aka fel666
Eeehhh....
Why does this always happen? (Is that what I used to be like?)

Back on track:

GMS2 will also be getting new GML features soon. So if you are still on the fence, perhaps wait until these get added and see if they make it worthwhile the upgrade.
 

Mike

nobody important
GMC Elder
It says a lot that all the countless hours of my life I've spent trying to put the right words on what exactly is wrong with the UI is literally just "ranting", "bitching" and "moaning" to you. Just because it's not what you want to hear doesn't make me a troll - as you will notice, I am not the only one here with these complaints. The fact that they date back to the very beginning of GMS2 and are still here should say a lot. They would be even if I wasn't around.
No... I've never said anyone who has constructive criticism about the IDE was bitching or moaning - I said you were, because you've always gone over the same points, over and over and over again. Even when the company said it wasn't going to happen, you refuse point blank to let it go. You assume "your" always right, and you're not.

Again, this says a lot about your mentality. You don't like it, so you don't care how many other people do. And especially with no data to back it up, it makes it even more out to look like you're just dead set on your way, and nothing else. I love the fact that I can at the very least have the windows overlap, and I would hate working in GMS2 even more than I already do if that wasn't an option.
You keep going on about how it's only a few, but there's a lot more here (not just in this thread) speaking against GMS2's IDE than for. Has been that way for years now, and it's not really changing.
LOL - why do you insist on thinking it's just me? That I controlled everything that happened there? I expressed my opinions, as did everyone else there.
The overlapping was added because a lot of folk asked for it, it wasn't a huge task, and it didn't cripple the usability of the IDE. And sorry... saying that there are more folk who are against GMS2 than for it, is just a lie.... you live in this "little world" and don't know anything outside of here - but hay, whatever. I really don't care.

If it's too late to change that, then that's one thing, but brushing it off as just a minority with no data to back it up, when there's a clear majority of people arguing against GMS2's IDE on the forums is kind of bold - just be up front about it.
Just because you don't like the answer... hell, you've never liked the answer - hence the "keep on bitching" comment, doesn't mean there isn't evidence, your just refusing to accept it. That's your right if that's what you want to do. By all means, think I'l lying to you, I don't really care.

Giving us an option to opt out of the chains would go a LONG way for a lot of us, especially combined with the ability to overlap. Even just the ability to close specific parts of a chain only would go a long way. Add then the ability to disable auto panning of the workspace, and you've actually got something that could work. It doesn't take a lot.
You're also assuming this isn't a huge task... the "just give us an option". I said before, doing so is a massive commitment, not just a case of throwing something in an hoping for the best.

Anyways.... I've yet again, had to repeat myself multiple times - just in the past few posts, as I've always seemed to have to do with you. But hey. Believe whatever you like. Go file a suggestion. I'm rooting for you!! :)




As to the GML changes... I don't think you should wait for this before trying the demo, there's way more to GMS2 than that, and as has been said... GMS1 will eventually break - Apple and Android may already be broken (I've no idea). For anything new, you really do have to switch.
But the only way to know for sure is to try it on for size.....
 

Kezarus

Endless Game Maker
I am interested in the new GM2 and there are some cool new features on the roadmap (link). =]

If you will start a project now and don't have GM1.4 then GM2 is a way to go as other people here said too.

What @FrostyCat said got me thinking about if the windows version have the same issues or not.

I am looking on other dev tools too. Maybe they are better, who knows?


p.s.: it's unfortunate that some comments have to be so ill-intended and don't add a thing to the discussion but bitterness. shame.
 

JeffJ

Member
you've always gone over the same points, over and over and over again
Actually, in this case, someone was asking for opinions on how GMS2 is now. Someone mentioned the IDE, and I backed it up, and other people joined in. That's why this was brought up. And I wasn't even the first (or the last) one to do it. If people ask, they get an answer.

You assume "your" always right, and you're not.
And sorry... saying that there are more folk who are against GMS2 than for it, is just a lie.... you live in this "little world" and don't know anything outside of here
doesn't mean there isn't evidence, your just refusing to accept it
For the last time - show us the evidence you keep claiming exists. Because until you (or someone) does, it will always remain a he said / she said deal. Put up or shut up - it's really easy to say "I'm right, you're wrong" with absolutely no evidence. I could do the same, yet you take issue with that.

Anyways.... I've yet again, had to repeat myself multiple times - just in the past few posts, as I've always seemed to have to do with you
No, you've really never had to, and you really don't have to now. Again, another user asked about experiences and opinions specifically comparing GMS1 and GMS2, and I answered with what I thought was relevant (as did other people). You chose to start all this by calling me out directly.
 
H

Homunculus

Guest
If only I knew the direction this topic was heading into... Let me just say I appreciate the fact that you are still here engaging in conversations (with your tongue biting limiter removed) Mike.

I wanted to add that, despite my earlier comment may seem to point out the contrary, I don't want the old IDE back. I had an experience similar to the one @rIKmAN described, so I still think GMS2 and its IDE is the way forward, but I can't deny that the current experience is, at least for me, far less than ideal. It may be a personal thing, I don't really know the numbers involved here, everyone talks about majority of X or minority of Y, but let's not forget that it's not black and white, there is people like me who don't want to go back but still want to see some important improvements to the current state of the IDE.
 
Last edited by a moderator:

Mike

nobody important
GMC Elder
For the last time - show us the evidence..
Already said that wasn't possible....try and pay attention. And YoYo doesn't have to either if they don't want to, they're a commercial company and can run it however they see fit.

If only I knew the direction this topic was heading into... Let me just say I appreciate the fact that you are still here engaging in conversations (with your tongue biting limiter removed) Mike..
:D

I just wanted to say that, despite my earlier comment may seem to point out the contrary, I don't want the old IDE back. I had an experience similar to the one @rIKmAN described, so I still think GMS2 and its IDE is the way forward, but I can't deny that the current experience is, at least for me, far less than ideal. It may be a personal thing, I don't really know the numbers involved here, everyone talks about majority of X or minority of Y, but let's not forget that it's not black and white, there is people like me who don't want to go back but still want to see some important improvements to the current state of the IDE.
There's always things that can - and should, be improved. The text editors are great- especially the full screen ones, but the work space ones could certainly be improved in the initial size/position of them.
The image editor "selection" as a brush is a brill feature, but shouldn't be the core copy/paste mechanic. Placing large images is just too finiky. Again, this was something that was going to get looked at, but I guess other things came along....
The room editor seems solid enough, but then there's always things in there you could add - but it's monumentally better than the 1.x one...

I do wish the API had been opened up so others could write plugins... but that's probably still MILES away. There's SO much to document for this to happen....
 

JeffJ

Member
Already said that wasn't possible....try and pay attention. And YoYo doesn't have to either if they don't want to, they're a commercial company and can run it however they see fit.
That goes without saying, they don't have to do anything. But you can appreciate that it is a bit hard to argue with "I am right, I have the data, but you will never see it, just trust me" when you have very differing experiences and a lot of other people sharing said experiences. That doesn't really allow any discussion to be had other than "you're wrong, I'm right". Which is why we're going in circles.

Anyway, to get back on topic:

I guess a lot of it depends on what features you want, and what platforms you aim for.

If you could benefit from tile collision (which can be a huge performance boost) or some of the upcoming (or already existing) GML additions (some of my own personal favorites right now are the variable functions), then those are cases to be made for GMS2. The room editor also has some big improvements in certain areas, and makes designing big levels much more doable, plus you have features like room inheritance. You also have instance creation code in the room editor, something I personally use a lot for getting much more mileage out of one single object that can have much more unique behavior on an instance level within the same room, all very easy to set up from a level editing perspective.

From a code perspective, I also enjoy QOL features like regions, and the newer features like syntax checking (that will tell you about stuff like unused variables)

Last but not least there's the ongoing support for the ever changing platforms such as mobile and console (and native Switch isn't even a thing in 1.4).

So if you can live with the IDE changes, and/or if any of the above sounds like something you would want or need, then it definitely seems like a no brainer.

If, on the other hand there's nothing of interest mentioned above, and you primarily intend to target Windows, then you may still get a lot of mileage out of 1.4.

Ultimately, like others have mentioned, you could at least try the demo and get a feel for yourself.
 

rIKmAN

Member
I do wish the API had been opened up so others could write plugins... but that's probably still MILES away. There's SO much to document for this to happen....
~3yrs ago we were told it just needed to be documented, this bad boy is gonna give the National Archives a run for it's money! :D
Well, either that or nobody has started to document it yet - GMS3 maybe?
 

Mike

nobody important
GMC Elder
~3yrs ago we were told it just needed to be documented, this bad boy is gonna give the National Archives a run for it's money! :D
Well, either that or nobody has started to document it yet - GMS3 maybe?
The API is huge.... initially I'd though we'd document "bits" and release a few sample plugins, then let folk figure stuff out. But James muted that... Perhaps now it's Stuart running things that could appear again. But you're probably right... GMS3 at least.....
 
Last edited:

samspade

Member
As someone who had used GMS 1 for only a few months before GMS 2 came out for real and was using both simultaneously pretty much from the start and therefore without the baggage of being familiar with one of the other I can tell you that it isn't even close for me. I switched to GMS 2 exclusively after the first year and every time I've gone back to GMS 1 to look through an old project I wonder how I made it that long.

While there was a period for me where GMS 2 did freeze up at lot, that seems to have stopped after some of the recent updates. I've had no issues with it stability wise for the last few months.

IDE wise, GMS2 is leaps and bounds better than GMS 1 in my opinion. I agree about the chained windows thing. I think it is terrible and never use it (I routinely close close and reopen the workspace tab so it gets wiped). Everything I do is full screen and in tabs in the IDE. This is infinitely preferable to the windowed version of GMS 1 which you can still basically do in GMS2 if you want and I sometimes do if I want to display two sections of code in split screen. You can even book mark tabs with keyboard shortcuts if you want. And that doesn't take into account all the other quality of life improvements that others have mentioned.

While I would agree that the laptop version could be slightly 'thinner' I worked on GMS 2 on a laptop exclusively for over a year and really had no problems (but again I work exclusively in the 'tabs' veiw).
 

FrostyCat

Redemption Seeker
Allow me to chime in on the conversation on both sides.


As for you, @JeffJ, I decided to take some lunch break time to investigate GMS 2 Preferences for people who don't like workspaces and chains. It turns out there are a number of options for those of us who are less inclined towards workspaces and have limited screen space.
  • General Settings
    • Disable IDE transition animations: CHECK
  • General Settings > Workspace
    • Workspace keyboard navigation degree range: 360
    • Workspace chain column padding: 1
    • Workspace chain row padding: 1
    • Workspace chain rendering segment count: 1
  • Text Editors
    • Open scripts in a full screen editor
    • Open shaders in a full screen editor
    • Open event scripts as tabs in the same window
    • Open linked scripts as tabs in the same window
  • Resource Tree
    • Sprite double click opens image editor
    • Object double click opens all object events
Credits: @Mick for recommending the options in Text Editors and Resource Tree (https://forum.yoyogames.com/index.p...-the-old-interface-instead.44509/#post-276751)
Add to these the keyboard shortcuts Ctrl+T, Ctrl+W, Ctrl+Tab, Ctrl+PgUp and Ctrl+PgDn, and the navigability of the IDE improves drastically. I'll admit, there are gaps in this "low-mouse-utilization" workflow (e.g. adding events or accessing Edit from Ctrl+T when "Object double click opens all object events" is on). But first we should get familiar with effective use of the GMS 2 IDE (not just default or popular usage), and then we can propose low-cost ways of making it better in the short term.


As for you, @Mike, allow me to give you a glimpse of what annoyances laptop users have with workspaces. This screenshot is perhaps as representative of our complaint as I can think of (as demonstrated on a 1280x800 MacBook Pro screen):

covered_workspace.png

Above is the default positioning of an object properties window after double-clicking on it. As you can see, this window is placed in a way that needs to be dragged back into view first, and that's pointless busy work that anyone without a 4K monitor now contends with. In contrast, a tabbed layout doesn't need this kind of busy work. You just need to click the right tab and it's ready to go.

Furthermore, look at the size of the chains. This is a lot of space for something purely decorative. And given there are at least 2 of these before I could get at the code for an event, you can imagine why we aren't a fan of it.

Another bug is that when windows animations are on, sometimes a code window closes and the view doesn't pan back to the objects properties window. Instead it stays there and leaves us looking at a void, and then we have to manually pan back. More busy work.

Add to this the instability that the graphics-intensive GMS 2 IDE is adding by having more moving pieces. I have used GMS 2 in Jams on 4 different occasions, and every time I had to give the IDE "restart breaks" every 30-60 minutes. The RAM usage would just climb every time I change tabs or the syntax checker triggers, and it goes into the 2-3GB before it crashes outright. The "restart breaks" are ostrich fixes in anticipation of these. No other piece of software I own does this, not even GMS 1.4.

So here you go, this is the blunt perspective from someone who isn't a huge fan of workspaces.
 

Mike

nobody important
GMC Elder
Allow me to chime in on the conversation on both sides.
As for you, @Mike, allow me to give you a glimpse of what annoyances laptop users have with workspaces. This screenshot is perhaps as representative of our complaint as I can think of (as demonstrated on a 1280x800 MacBook Pro screen):
View attachment 26837
First, I have said laptops were definitely an issue, but reducing the DPI down will help that. Second, I'm not sure why you're running the windows version instead of the Mac one? Third... you're wasting a lot of space with the windows task bar - auto hide it if real estate is an issue. Also, pressing F12 will hide the resource tree and output window - or at the very least hide the output window. This would free up a lot of free space for you to use. You can also shrink the resource tree width down using the slider scale at the bottom, which would also let you halve the size of it.
I'm not saying it's not an issue, but I do think you're wasting a lot of space....

Furthermore, look at the size of the chains. This is a lot of space for something purely decorative. And given there are at least 2 of these before I could get at the code for an event, you can imagine why we aren't a fan of it.
I've also said this was something that was going to be looked at... but I've no idea if they still are. I don't think it's a massive issue, you can pan between text windows using the cursor keys, so how much fits on the screen is less important.

Another bug is that when windows animations are on, sometimes a code window closes and the view doesn't pan back to the objects properties window. Instead it stays there and leaves us looking at a void, and then we have to manually pan back. More busy work.
File it I guess... not much I can do about it.

Add to this the instability that the graphics-intensive GMS 2 IDE is adding by having more moving pieces. I have used GMS 2 in Jams on 4 different occasions, and every time I had to give the IDE "restart breaks" every 30-60 minutes. The RAM usage would just climb every time I change tabs or the syntax checker triggers, and it goes into the 2-3GB before it crashes outright. The "restart breaks" are ostrich fixes in anticipation of these. No other piece of software I own does this, not even GMS 1.4.
I've no idea what you're doing to get this. It's been run on "terrible" hardware, and it's more than fast enough. It's far from "graphically intensive". It's been run on a raspberry Pi 3.... on an old ATOM+crappy cheap 2011 Nvidia machine. Try running it natively on the Mac, and not using a VM... that would probably help l;oads.


So here you go, this is the blunt perspective from someone who isn't a huge fan of workspaces.
Yeah....I think we got that. But you're also clearly not making best use GMS2, your screen space, or actual machine. Try these suggestions, see if that helps.
 

TheouAegis

Member
Wow, this got far along quickly. I stopped reading at Mike's first post. lol

I only have a trial of GMS2. (Thanks again, YYG! :D ) I just use it to try to acquaint myself with some of its nuances and take notes for myself. I will still use GMS1 and GM8 (mostly GMS1 these days).

I try to make my code as flexible as I can, which I started doing shortly after using GMS1. Code I write in GM8 I use GM8 syntax for, but facets I know I would beed to change in GMS1 I put in scripts. Likewise, code in GMS1 I'd expect to have to change in GMS2, I also put in scripts. Even if it's something as simple as playing audio files. This definitely applies to tile collisions. I love GMS2's tile handling (so far). So even if you intend to use GMS1 exclusively, I'd advise making your code simple to edut for upgrading it to GMS2 code in case you need to have a friend port it over to GMS2 for exporting to other OSes.

I don't like the GMS2 IDE, but I'm getting slightly better using it. I still like GM8 and GMS1 IDE. I also like Windows XP...
 

FrostyCat

Redemption Seeker
First, I have said laptops were definitely an issue, but reducing the DPI down will help that.
I tried cutting down on the DPI, but that made much of the IDE text too small, and the dropdown carets became unreadable. The lowest I could feasibly do is 75% (72dpi). The only places I could see options for font size (for counteracting the scale-down) are the output windows and code editors.

Second, I'm not sure why you're running the windows version instead of the Mac one?
I dual-boot Windows 7 and Mac OS X Mojave on my MacBook Pro, and I have GMS 2 on both sides. This is the Windows side.

Third... you're wasting a lot of space with the windows task bar - auto hide it if real estate is an issue. Also, pressing F12 will hide the resource tree and output window - or at the very least hide the output window. This would free up a lot of free space for you to use. You can also shrink the resource tree width down using the slider scale at the bottom, which would also let you halve the size of it.
I'm not saying it's not an issue, but I do think you're wasting a lot of space....
I've been doing both the F12 and slider scales regularly already, the screenshot is for demonstrating what happens by default.

Also, hiding the task bar is a little better, but I also have another laptop whose screen is 1366x768. So the best I could do with the other laptop is that of my MacBook Pro with the task bar out.

I've also said this was something that was going to be looked at... but I've no idea if they still are. I don't think it's a massive issue, you can pan between text windows using the cursor keys, so how much fits on the screen is less important.
They aren't working on it. But now that I found the option for making the chains so small that they practically don't exist, I'll pass.

I've no idea what you're doing to get this. It's been run on "terrible" hardware, and it's more than fast enough. It's far from "graphically intensive". It's been run on a raspberry Pi 3.... on an old ATOM+crappy cheap 2011 Nvidia machine. Try running it natively on the Mac, and not using a VM... that would probably help l;oads.
I'm not running off a VM, this is Windows 7 over Boot Camp. So it's more or less native.

I also have another Windows laptop from 2010 (4GB RAM, upgraded to Windows 10) that does the exact same thing, so it's not exactly a Mac-exclusive issue. League of Legends could last all night on both of these, yet GMS 2 under a typical GMC Jam workload lasts at most an hour. I'm not doing any fancy here --- just changing tabs, typing regular code and building every now and then.

Yeah....I think we got that. But you're also clearly not making best use GMS2, your screen space, or actual machine. Try these suggestions, see if that helps.
I agree, but shouldn't defaults tend towards making better use of GMS 2 and screen space? Say, when "laptop mode" is on, the resource tree should be half the width as usual and the output window should be hidden by default.

In any case, what would your idea of "good use of GMS 2 and screen space" be under a laptop-based setup?
 
H

HW.

Guest
I'll have to go against the general feel of this thread and say that for me it's the opposite: I hate the GMS1.4 IDE and don't miss it a single bit, in fact having to go back to it to test my extension a couple of weeks ago just made me realise how archaic the thing is.

I wouldn't say I "love" the GMS2 IDE and there are definitely some improvements to be made, but I don't have any major issues with it and I have no real problems with the workspaces at all - I'd take it over the 1.4 IDE 100/100 times without hesitation.
This.
You are not alone. I don't miss GMS1 a single bit either! I still use GMS1 only when i am forced to use it (open old 1.4 assets from the marketplace).

While I'm obviously going to be biased, I have to say, the vast majority have none of these issues, and have found GMS2 to be far more productive.
I've said it before, GMS1.x was a horrible place to be. Massive amounts of overlapping windows meaning you simply couldn't have that much open at once without losing track of things.

GMS2 allows you to have open many things at once, without the mess. Yes, we should have tested more on laptops, but there are loads of shortcut keys to help you navigate, and you can scale the whole UI down using the DPI override.
And this.
Yes, a horrible place to be lol. (personal opinion to me also)
Personally, i can't agree more!
The Massive amounts of overlapping windows are really one of the downsides of the 1.4 that I really hate it, especially after i migrate to GMS2, after seeing the difference. In my case, there is impact on productivity.

That's why I rarely use GMS1.4, just only for opening 1.4 assets i bought from marketplace. The only reason i still use it. Yeah, GMS2 can import it by the way , but many times i don't want the compatibility script.

So , everytime i open GMS1, i will immediately copy paste all the codes there manually, one by one (really tiring), and open GMS2 IDE to paste all the codes from GMS1.

After finishing all the copy pasting of all the codes, i just start playing with the codes entirely in GMS2 without a mess lol.

I prefer being tired of copy pasting the codes, to seeing those overlapping windows (the mess).

I really hate those GMS1' overlapping windows and the need to "double clicking to open" everything.

GMS1 is more stable but "the mess" (overlapping windows) makes me feel uncomfortable using it for long.

GMS2 is much neater which makes me feel very comfortable using it for long. Only the crash that stops me from using it for a long time.

If only, all the marketplace assets are GMS2 assets, i already uninstalled GMS1 on my laptop lol.

I keep 1.4.1763, just only for 1.4 assets i buy on marketplace.

For anything else are GMS2 for sure!!
 

Kezarus

Endless Game Maker
I also have another Windows laptop from 2010 (4GB RAM, upgraded to Windows 10) that does the exact same thing, so it's not exactly a Mac-exclusive issue. League of Legends could last all night on both of these, yet GMS 2 under a typical GMC Jam workload lasts at most an hour. I'm not doing any fancy here --- just changing tabs, typing regular code and building every now and then.
Well, this always get me worried. I remember when GM1 was updating and then sudden crashes happens and all those shenanigans.

Now GM 1.4 is stable as ever. :D But not maintained anymore. :(

Will definetly give it a try on January 2020 when I will start another project cycle.
 

Mike

nobody important
GMC Elder
Also, hiding the task bar is a little better, but I also have another laptop whose screen is 1366x768. So the best I could do with the other laptop is that of my MacBook Pro with the task bar out.
Yeah.... that's pretty teeny...

They aren't working on it. But now that I found the option for making the chains so small that they practically don't exist, I'll pass.
That's a shame... a "thinner" skin would help a lot - but it's a HUGE amount of work. Ah..I thought there was an option for that.... but it's been a while so I couldn't remember

I'm not running off a VM, this is Windows 7 over Boot Camp. So it's more or less native.
okay.. no worries. It should be noted, that windows running on a Mac isn't that efficient. Apples drivers aren't great an definitely draw more power than under OSX. It might be worth trying it under OSX to see if it has a lighter load.....

I also have another Windows laptop from 2010 (4GB RAM, upgraded to Windows 10) that does the exact same thing, so it's not exactly a Mac-exclusive issue. League of Legends could last all night on both of these, yet GMS 2 under a typical GMC Jam workload lasts at most an hour. I'm not doing any fancy here --- just changing tabs, typing regular code and building every now and then.
It might be specific editors leaking I guess... I'm mainly on the workspace, text editors... and a "little" bit in the image editors and room editors. Do you use any other one heavily? I've really not seen any of these issues...

I agree, but shouldn't defaults tend towards making better use of GMS 2 and screen space? Say, when "laptop mode" is on, the resource tree should be half the width as usual and the output window should be hidden by default.
It depends what you think most folk will be using. I think when we started, desktops were still more the norm (on the whole).... but that's definitely shifted. Most folk these days sit on laptops, so I wouldn't have made the same choices now. Perhaps GMS3 will update that - whenever that appears...
EDIT: Also... education will still play a big part in this choice, and they are all defiantly on desktops. Up to YoYo I guess....

In any case, what would your idea of "good use of GMS 2 and screen space" be under a laptop-based setup?
More or less the things I said.... I use F12 a lot, I thin down the resource tree, and I use the full screen text editor... I do pan around a lot though, but that I don't mind, as I use it to follow script calls anyway.
 
Last edited:

kraifpatrik

(edited)
GameMaker Dev.
Most of the time I spend in GMS2 is in the code editor, so I can't really tell you about others like the new sprite and room editors, but the code editor is so much better compared to GMS1.4. The new variable name autocompletion, the ability to indent with tabs and set how much space one tab is, to edit multiple lines at once, to open multiple scripts and events in tabs, to create foldable code regions, to duplicate lines using Ctrl+D, to switch between resources quickly using Ctrl+T, the support for JSDoc to document scripts as well as the new GML features like macros (which were previously defined in GUI instead of with code), the new ternary operator etc. just make the programming experience so much more comfortable, that I literally hate when I have to go back to 1.4 to fix some older projects. I also really like that objects show events inherited from their parents, so I know that I'm overwriting something. Additionally, the YYC got a lot nicer and more performant. And GML is going to get even better in the future. So I would say yes, it only makes sense to upgrade, at least from the programming perspective.
 
Last edited:
E

elsi11

Guest
What do you guys say to the DRM? People seem to be really miffed by it because it inconveniences the paying customer, while the unpaying one gets the superior product.
Frequent logins are one of the top grievances people have on Steam, closely followed by memory leaks (will be supposedly fixed in 2019), unintuitive IDE with a big learning curve (we already talked about this), price, bugs, complicated Linux export and no support for the OS.

I remember someone saying that exported game performances also differ between 1.4 and 2. Is this because 2 uses newer dependencies? And if all the dependencies are bundled together into the game, why do you say that the game might stop working someday? Only problem with games, even ancient ones, is having the DirectDraw(somethingsomething) enabled in wondows. Everything else works nicely usually.

You know, I have this handicap because I have Linux. Do you suppose I could run GM2 in a VM? I'l probably have to dual boot or use my laptop...
 

Mike

nobody important
GMC Elder
What do you guys say to the DRM? People seem to be really miffed by it because it inconveniences the paying customer, while the unpaying one gets the superior product.
Frequent logins are one of the top grievances people have on Steam, closely followed by memory leaks (will be supposedly fixed in 2019), unintuitive IDE with a big learning curve (we already talked about this), price, bugs, complicated Linux export and no support for the OS.
No one "likes" DRM, but I'm afraid it's a requirement on something that is so badly pirated. In 8.x it was so easy to copy and pirate, than sales were terrible, and most folk used a pirate copy. If someone doesn't "have" to pay for something, in general.... they won't. Most folk I know will have copy of "something", be that software of films/tv programs - this is a simple fact of life. After the DRM was introduced, piracy fell to almost nothing (by comparison). I've no idea what it is now, but I can't imagine that having changed much

I remember someone saying that exported game performances also differ between 1.4 and 2. Is this because 2 uses newer dependencies? And if all the dependencies are bundled together into the game, why do you say that the game might stop working someday? Only problem with games, even ancient ones, is having the DirectDraw(somethingsomething) enabled in wondows. Everything else works nicely usually.
GMS2 is much faster than 1.x games - if you use the new features (like tilemaps etc). There are also API speed ups (using DX11 etc), along with faster YYC integration. So in general, GMS2 should always be quicker than 1.x
Game will stop working in 1.x when Apple/Google update an API and make the one it's running on obsolete. At some point it will just no longer work. Windows is a slightly different target, because it'll work for some time. However, it's based on DX9, and driver writers no longer care about that, so drivers are becoming more and more flaky, meaning games "could" end up crashing or something on newer devices

You know, I have this handicap because I have Linux. Do you suppose I could run GM2 in a VM? I'l probably have to dual boot or use my laptop...
A few folk run GM in VMs... but it's never the best way, and your battery will take a pounding - even if you reduce the power settings in GMS2.
 
D

Dawn

Guest
I'm really curious what they were smoking when they decided to make GMS2 store assets in separate folders. I don't know if that's okay in other OS but that is likely the dumbest unfixable mistake they've made with GMS2. Also, talk about implementing local package after years since release.

And there are some very annoying UI issues like that windows can only exist in the workspace unlike 1.x but that's not a stability issue I guess.
 

Mike

nobody important
GMC Elder
It wasn't a complicated choice. In GMS1.x it stored everything (images etc) in one big ol' folder. It was an utter shambles. Trying to find anything was a nightmare. Add to that the layers for images etc and this folder would be utterly gigantic! Folders for each object - holding the individual scripts, or an image - holding each layer of the image, keeps the folder contents tidy.

It also meant deleting, renaming objects etc was more focused. This has definitely had issue though...

They may change it in the future - who knows.... but having tens of thousands of files in a single folder can kill windows explorer dead - I've had this before in another project.

*shug* Like it or not... that was the reason.
 
H

Homunculus

Guest
I'm really curious what they were smoking when they decided to make GMS2 store assets in separate folders. I don't know if that's okay in other OS but that is likely the dumbest unfixable mistake they've made with GMS2..
I don't get it, why is that a bad thing? And why should I care as an end user?
 

GMWolf

aka fel666
It wasn't a complicated choice. In GMS1.x it stored everything (images etc) in one big ol' folder. It was an utter shambles. Trying to find anything was a nightmare. Add to that the layers for images etc and this folder would be utterly gigantic! Folders for each object - holding the individual scripts, or an image - holding each layer of the image, keeps the folder contents tidy.

It also meant deleting, renaming objects etc was more focused. This has definitely had issue though...

They may change it in the future - who knows.... but having tens of thousands of files in a single folder can kill windows explorer dead - I've had this before in another project.

*shug* Like it or not... that was the reason.
Folders for individual scripts is ridiculous. You just swapped files for folders.

What would have been sensible is to use a file system based project structure.
 

FrostyCat

Redemption Seeker
Wasn't there news of a newer project format that would help address source control compatibility? Right now you can't even do a simple merge without a load of conflicts in the views directory, and a purely file-system-based project structure will achieve that.

Let's hear what YoYo has planned for that (if any), then comment on that.
 

chamaeleon

Member
Folders for individual scripts is ridiculous. You just swapped files for folders.

What would have been sensible is to use a file system based project structure.
I agree. Groups in GMS could then be folders in the filesystem, avoiding the explosion of files in single directories if the project is complicated enough. Instead of UUIDs, relative pathnames from the base of the project for unique identification in project files.
 

Mike

nobody important
GMC Elder
Folders for individual scripts is ridiculous. You just swapped files for folders.
What would have been sensible is to use a file system based project structure.
It's not quite that simple. Each resource gets its own folder. Objects - which hold a lot, have a folder. Sprites, which hold a lot, have a folder. Scripts do actually have settings with each file, so with every script, there are 2 files. There was also the possibility of project inheritance, but that got dropped as things got massively complex when editing and compiling. On top of this, it was to try and help plugin authors via an MVC model so they didn't mess up everything when things went wrong. So there is historical stuff in there as well, that's no longer required or desirable. But when you make a root choice - for whatever reason, you're certainly stuck with it for some time - not forever, but for a good while.

Wasn't there news of a newer project format that would help address source control compatibility? Right now you can't even do a simple merge without a load of conflicts in the views directory, and a purely file-system-based project structure will achieve that. Let's hear what YoYo has planned for that (if any), then comment on that.
Yeah, there was talk prior to me leaving, so I'm sure it'll change - I just don't know when. Time will tell....

Instead of UUIDs, relative pathnames from the base of the project for unique identification in project files.
It did used to have this in a few places, but as soon as you moved or renamed something, it had to rename/move loads of files, and it all seized up. UUIDs meant this was minimised, although it makes looking for files a pain in the arse.
 

chamaeleon

Member
It did used to have this in a few places, but as soon as you moved or renamed something, it had to rename/move loads of files, and it all seized up. UUIDs meant this was minimised, although it makes looking for files a pain in the arse.
In this day and age of object oriented programming, this seems like it should be a solved problem almost. An object knows about relevant filenames for it, and higher level objects like collection of assets or groups of assets know only about those things in memory as pointers or whatever, not the names. Making a filename change would be a matter of working with a property of said object only, followed with saving whatever project files reflect the hierarchy of affected things in memory by telling each object to seralize itself (or return a cached copy if it hasn't been modified, for premature optimization) to create the complete serialization of whatever project file is being written.
 

Mike

nobody important
GMC Elder
Yeah.... it's not as simple as it sounds, especially when you get into a hierarchy of inheritance... GMS base projects, company framework, game type framework, then actual game. All this causes a multitude of issues.

But whatever... My guess is that it's no longer needed. They'll probably clean up the format pretty soon. no idea when though.... perhaps even GMS3?
 

rIKmAN

Member
They'll probably clean up the format pretty soon. no idea when though.... perhaps even GMS3?
Hopefully GMS3 isn’t announced too soon, they’re still doing sales of GMS2 for hundreds of dollars so that would be a bit of a kick in the nuts.
 
Top