Fusion 3 blog any time soon?

Well, I said "might not be" because that's a whole other discussion, and I don't have all the facts.

I totally agree about web technologies being the way forward though - if you ever read one of those "MMF3 wishlist" threads from 15 years ago, you'll see C3 ended up being pretty much exactly what I was wishing for even back then (I've been using Clickteam products since Klik n Play).
I have absolutely no doubt that sooner or later, most software will be built using them (cutting edge 3D games being one of the few exceptions). The web has better GPU hardware acceleration than CF2.5 (WebGPU being based on DX12/Vulkan rather than DX11). Ignoring WebAssembly etc for now, Javascript is undeniably slower than a compiled language like C++, but only about 8 times slower, and modern hardware is so fast that it really doesn't matter (my 2020 smartphone is 4-5 times more powerful than the 2015 laptop) - even if it's not the fastest, it's fast enough. Being able to leverage the Javascript knowledge that I already possess would be huge - whereas extensions in CF2.5 are a mess, and have to be rewritten for every runtime.

Many popular productivity apps like the Google Office suite and Canva are already built as web apps. I think the reason that games of the kind made with CF2.5 (casual, 2D) haven't followed suit is that they are only really played on mobile devices these days, not PCs - and Apple have done absolutely everything they could (including breaking any and all regulations) to kill web apps. In fact, Apple are back in the news again this week, for this very reason - Please login to see this link.

I understand the reason for making C3 subscription based, but it doesn't mean I have to like it! There was a similar backlash against Adobe's Creative Cloud. In fairness, a £105 per year subscription, while expensive compared to CF2.5, is still pretty cheap as hobbies go - I spend more than that in a month, playing badminton.

Will F3 be subscribtion based, does anyone knows this for sure? I hope not!! I would rather pay $300 for it. If it will be subscribtion then I will am out.

Quote from andy

Will F3 be subscribtion based, does anyone knows this for sure? I hope not!! I would rather pay $300 for it. If it will be subscribtion then I will am out.

No announcement has been made or hints given, and I doubt any decision has been made either (since there's nothing to suggest F3 will be coming any time soon, it would seem a bit premature). I think people just started to speculate based on the subscription model becoming increasingly widespread in recent years.

And of course, $300 is only equivalent to 3 years subscription to C3, which is far less than the product's lifespan, so if it is sold as a one-off purchase, then it's entirely possible that it might cost considerably more than that - but of course, that's also purely speculation.

Do you need to run HTML5 games using a browser? If so, I don't understand why that would be desirable. It makes sense for a cloud-based productivity app like Google Office, but for a game, it just seems like an unnecessary bit of faff to impose on the player, doesn't it? Dx12/vulkan might be a good thing of course, but that doesn't require HTML5 per se.

I don't even think web-based is a good fit for many non-game types of software. I once bought a program that plays seamlessly randomised ambience sounds (forest, rain, ocean, etc.) It did a great job of that, and was hugely beneficial to someone like me who tries to work with noisy kids in the house.

But it was a web-based app that I had to run through Chrome, which gave it a number of usability flaws: it took up space in the task bar, I couldn't control it via global hotkeys or a system tray menu, its volume was inextricably linked to that of Chrome, and its stability was at the mercy of other tabs not crashing Chrome.

Eventually, all the extra steps required of me to operate this little program became so annoying that I uninstalled it and built my own version in Fusion. I now have my own program that does more or less the same thing, yet is markedly superior. It's a lot quicker to operate via global hotkeys, it has an entirely independent volume, it can startup automatically with Windows, and it sits quietly hidden away in the system tray. It's superior precisely because it's not a Web app.

Maybe I'm misunderstanding what an HTML5-based world would actually entail, but all I can see are various downsides without much real end-user benefit, so I don't understand why one would be excited by it.

Quote from Volnaiskra

Do you need to run HTML5 games using a browser?

That's actually a complicated question to answer, as they can work in different ways (hybrid apps, PWAs, electron, neutralino/tauri, etc) but essentially, yes, behind the scenes, they are running in a browser - although they have access to lower level functionality that normal websites don't, which means a well made example won't have the issues you experienced. And obviously, as they become more widely adopted, they're only going to become more powerful. While Apple may be actively trying to kill PWAs now, it was they who first came up with the idea to begin with - originally, the plan was for all iPhone apps to be web apps, so even they can see the potential (that's what scares them, in fact).

When Java (not Javascript) was first created, it was marketed as offering the ability to "write once, run anywhere", but it never actually delivered on that promise. This is exactly what Web Apps could offer though - applications that can be written once, and will then simply work on anything, whether it's a smartphone, tablet, PC, smartwatch, etc and regardless of operating system. Developers don't need to learn a million different languages. Businesses don't need to pay people to remake the same project 5 times. In the case of CF2.5, it means no separate exporters that take Clickteam years to develop, cost users extra to buy, and then quickly become obsolete (XNA? OUYA?). No need to port extensions to every different runtime either (which usually doesn't happen, because the original developer doesn't know the required programming language).

This is what has allowed C3 to progress so rapidly, while CF3 has languished in development hell, year after year.

For me personally, the tight integration of Javascript in C3 is huge. CF2.5 events are very user friendly, which is great for beginners, but it's an incredibly slow and clunky way of generating code - particularly for anything moderately complex, like pathfinding or AI. It's much, much quicker and easier to write actual code to do that. The only way to do that in CF2.5 is by creating an extension, and like I already said, CF2.5 extensions are a total mess. Related to that, you might have noticed a number of threads recently, asking about the best way to save a game in CF2.5, and the honest answer is that there is no good way. It's always "You could use INI file, BUT..." or "You could use an array, BUT..." . In C3, every object has an expression "asJSON" which converts its entire state into a single JSON string - or even simpler, there's a function to automatically save the entire gamestate as a JSON string.

Quote from MuddyMole

whereas extensions in CF2.5 are a mess, and have to be rewritten for every runtime

As long as you're not using platform-specific tech in non-portable libraries, DarkEdif extension SDK that supports Windows, Android, iOS, Mac. You write code once and build for all CPU types, simulator/device, is Unicode and 64-bit compatible, etc. Same C/C++ code lines will run on all of them. No rewrites!

The Please login to see this link. opened access to runtime variables and object selection too. So hopefully this encourages more cross-platform exts.

Weakness is displaying exts, since each OS uses different graphics engines, and you'd have to go through each Fusion runtime's differently tuned graphics library, but it could be in a future update. Not all internal variables are accessible, but there are examples for ext devs who know which one they want to expand it, and of course, it's open source, you can request edits to it.

Quote from MuddyMole

When Java (not Javascript) was first created, it was marketed as offering the ability to "write once, run anywhere", but it never actually delivered on that promise. This is exactly what Web Apps could offer though - applications that can be written once, and will then simply work on anything, whether it's a smartphone, tablet, PC, smartwatch, etc and regardless of operating system. Developers don't need to learn a million different languages. Businesses don't need to pay people to remake the same project 5 times. In the case of CF2.5, it means no separate exporters that take Clickteam years to develop, cost users extra to buy, and then quickly become obsolete (XNA? OUYA?). No need to port extensions to every different runtime either (which usually doesn't happen, because the original developer doesn't know the required programming language).

I appreciate the principle. Though the "write once, run anywhere" idea is just wishful thinking. In the real-world, an application that was written just once for phone, tablet, PC and smartwatch would be an absolutely awful application: it would likely be well-designed for 1 or perhaps 2 of those platforms, and would be horribly ill-suited to the rest of them. The vast differences between screen sizes and input devices alone means that apps for different platforms need to be designed and even conceived differently from the ground-up, regardless of what language they happen to be programmed in. Surely you've experienced the frustration of clunky, dumbed-down websites with enormous buttons and barebones features that were clearly designed for a touchscreen phone, but require annoyingly massive mouse gestures to navigate on PC (or conversely, tried to navigate a desktop-centric website on a phone). Or seen how compromised FPS or RTS games are when ported to console and played without the unique benefits of a keyboard and mouse. Or how unsatisfying racing games or twin-stick shooters are without analog gamepads. Or experienced the disappointment of playing a PC game that you know could have looked 5 times better, and had a x10 bigger map, had it not been developed with outdated console hardware in mind. The day we accept companies taking a one-size-fits-all approach to all app development is the day we settle for a world of lowest common denominator mediocrity.

But obviously it would be nice to be able to overcome some of the fragmentisation in the gaming landscape - particularly between Android-iOS and between PC-consoles. As long as it's truly seamless and doesn't require any lousy compromises. With Windows still being the largest gaming platform in the world, I don't think it's too much to ask that we be able to develop games in Windows that run natively in Windows, without people having to jump through annoying hoops like browsers.

From what I remember of the old Fusion 3 blogs, Clickteam are making Fusion platform-independent, with any extension written for F3 automatically working on all exporters. From memory, they were achieving this with C++, not with HTML5.

Quote from MuddyMole

For me personally, the tight integration of Javascript in C3 is huge. CF2.5 events are very user friendly, which is great for beginners, but it's an incredibly slow and clunky way of generating code - particularly for anything moderately complex, like pathfinding or AI. It's much, much quicker and easier to write actual code to do that. The only way to do that in CF2.5 is by creating an extension, and like I already said, CF2.5 extensions are a total mess

Each to their own. I personally detest syntax-based programming, and I always will. I used Actionscript2 a lot back in my Flash days, and assumed that I hated it simply because I was too much of a noob. But after years of using Fusion, I've realised that I've actually become a pretty half-decent programmer. I'm comfortable with the logic, and enjoy the problem-solving. And because I'm a productivity-obsessive - with my 15-button mouse, extensive macros, and various customised utility programs - I work pretty fast too. Yet I still have to dabble in syntax-based code from time to time (whether writing an Autohotkey script, or some C-based keyboard firmware, or a Javascript-based Toon Boom Harmony script), and I despise it as much as I ever did. In fact, my loathing of syntax-based programming has only increased as I've gotten better at Fusion.

I hate how finicky and brittle programming languages are, how reliant they are on memorisation, and how they seem to go out of their way to obfuscate important information with structural scaffolding and busywork. It takes me much longer to think through logical problems when I have to keep half my mind occupied by where to place the round bracket and where the square one, which line to end with a comma and which with a semicolon, or which of two near-identical items is a class and which is a function. It all strikes me as an incredibly inefficient and poorly-designed system, and it's hard for me to believe that people accept this type of programming for any reason other than that they simply don't know any different. Though the existence of people like you and Phi obviously proves that wrong. You've experienced the syntax-free world of Fusion, and yet you still feel very comfortable in the syntax-based world. It probably comes down to neurological differences, with some people's brains just being wired to process information differently to others.

Quote from Edel

I disagree with C3 not being the best choice for a fast-paced platformer. I've played fast-paced, HD metroidvanias/platformers made in C3 that run silky smooth on my 144hz monitor. The fact that it's an HTML5 game engine is trivial at this point. The tech is there! Not to mention WebGPU is here! It is superior to CF2.5 and it pains me to say that it's literally the F3 we should of had years ago. Still, my heart is with Cickteam!!!

On the whole subscription thing... I don't mind subscriptions(if fair). One reason we don't have F3 by now is a financial one. Not to mention all the pirates... I don't see how CT can survive with F3 if it's just buy once. I would gladly subscribe and support CT if it meant frequent updates and roadmaps. You have to support the developers that make the tools you use. Especially when it's a very small company.

Having been with CT from early CnP TGF days and seeing how the software has evolved.
I love to use it on occasion but now I'm a little more travelled as far as software building goes there are many solutions out there.
No one solution will work for everything/everyone. Clickteam have a learning curve that is second to none and allow multi platform builds. I love making software and Fusion 2.X has its place in development.
They give your own rudimentary 3D gaming engine to build on. IE make your own 3d game with your own programmed engine. (Firefly was a good idea but lack of good documentation and an experienced based examples meant it was difficult to make anything usable, like with the standard engine)

What everyone also knows is that during this time many new extensions and features have been released into the now Fusion 2.X builds.
Regular updates are still being given to these current builds and they are not just bug fixes so you can't work on 3 and 2 if you don't have the resourcing.

The development team is evolving so need to bring on new blood to cater for the ever changing development landscape.
Developers tend to flit from job to job (From personal experience) depending on how committed or comfortable the role is will depend on how long they stay.

You can speculate all you want but just don't know what is going on. Everyone knows its being developed.
For all we know F3 could just round the corner. The problem with major software releases on a new platform/product is sales.

There is a risk that there could be bugs in the software which can not be addressed before release so risk losing sales as a result and causing bad press to reduce sales even further.
Alternatively the software is being stringently tested for all manner of bugs before release to ensure less volatile sales.

I agree piracy hurts sales but too many people expect something for nothing and this software deserves the money asked. Don't mistake Software applications for Software Games. What makes me laugh is people will fritter their money away on a game without any thought for the improved innovation of that game / genre (if any).

Quote from Volnaiskra

Surely you've experienced the frustration of clunky, dumbed-down websites with enormous buttons and barebones features that were clearly designed for a touchscreen phone, but require annoyingly massive mouse gestures to navigate on PC (or conversely, tried to navigate a desktop-centric website on a phone).

Not in the last decade or so, I haven't. In fact, this only helps to demonstrate my point.
In the early days, all websites were optimized for desktop browsers, and if you tried viewing them on a mobile device, you were going to have a terrible time (and before that even, there was WAP, which everyone here has probably already forgotten, because it was so bad).

Then eventually, websites started to detect the visitor's browser, and would redirect visitors on mobile devices to an entirely different version of the website, which had been made specifically for them (at additional cost, of course). That was when web designers were still using tables for layout, and specifying the size of elements in pixels or other non-density-independent units. This is pretty much the stage we're at now with most other software - different versions having to be made for different devices.

Nowadays though, we have this thing called "responsive web design", where websites are dynamic, and a single design can automatically adjust itself to look good and function smoothly whatever device and browser the visitor is using, and whatever additional accessibility requirements they might have. It's very rare that you see a website that isn't made this way now - not least because it's a key factor that Google uses when ranking search results. This could be the future of other types of software too, if Web Apps take off.

I don't really understand your point about input peripherals on consoles vs PC vs mobile. Console hardware supports the use of a mouse and keyboard, and smartphones and browsers support the use of a gamepad - it's up to the consumer which peripherals they purchase, and up to the developer which they want to support. Even if a developer decides that a console gamer probably won't have a mouse, and they need to implement an additional control system, that's still far less work than porting the entire game to a completely different system.

Incidentally, I'm not saying Web Apps are going to be suitable for everything - big AAA games are one of the very few cases where ultimate performance really does matter a lot, so that's where "proper" programming still has its place.

Quote from DJFuego

Having been with CT from early CnP TGF days and seeing how the software has evolved.
I love to use it on occasion but now I'm a little more travelled as far as software building goes there are many solutions out there.

I hope that's still true, but I sometimes wonder. You and I both came in at the beginning, when Clickteam software was far less capable, but also far simpler, and we've been able to learn all the new features gradually, as and when they were introduced. I can't help thinking that for someone just getting started now, it must be pretty overwhelming.
It's the same with everything though. Every now and then I think "maybe I'll just give that other game engine another try", but then I do, and I'm greeted with something like this:

Please login to see this picture.

...and I just think "Nope!", and immediately uninstall it again. I bet if I'd started using it 15 years ago though, I'd have been fine.

With improving capabilities and expanding AI implementations in various software, I think F3 should take an interest in this to create games using prompts

Quote from MuddyMole

Not in the last decade or so, I haven't. In fact, this only helps to demonstrate my point.
In the early days, all websites were optimized for desktop browsers, and if you tried viewing them on a mobile device, you were going to have a terrible time (and before that even, there was WAP, which everyone here has probably already forgotten, because it was so bad).

Then eventually, websites started to detect the visitor's browser, and would redirect visitors on mobile devices to an entirely different version of the website, which had been made specifically for them (at additional cost, of course). That was when web designers were still using tables for layout, and specifying the size of elements in pixels or other non-density-independent units. This is pretty much the stage we're at now with most other software - different versions having to be made for different devices.

Nowadays though, we have this thing called "responsive web design", where websites are dynamic, and a single design can automatically adjust itself to look good and function smoothly whatever device and browser the visitor is using, and whatever additional accessibility requirements they might have. It's very rare that you see a website that isn't made this way now - not least because it's a key factor that Google uses when ranking search results. This could be the future of other types of software too, if Web Apps take off.

I don't really understand your point about input peripherals on consoles vs PC vs mobile. Console hardware supports the use of a mouse and keyboard, and smartphones and browsers support the use of a gamepad - it's up to the consumer which peripherals they purchase, and up to the developer which they want to support. Even if a developer decides that a console gamer probably won't have a mouse, and they need to implement an additional control system, that's still far less work than porting the entire game to a completely different system.

Incidentally, I'm not saying Web Apps are going to be suitable for everything - big AAA games are one of the very few cases where ultimate performance really does matter a lot, so that's where "proper" programming still has its place.

I used to be a professional web designer, so I know about responsive web design. I'm also aware that much of the time, people design things with one platform primarily in mind, and derive the other form factors from that as an afterthought.

A case in point is this very forum, actually. Whoever designed the template that Clickteam uses for this forum clearly designed it with phones in mind. As a result, it looks very good on a phone: the font is crisp and highly readable on the high-pixel-density screen, and the small purple bars around the avatars elegantly separate the posts. On desktop, the same font renders poorly, and becomes thin and harder to read. And since having a big purple bar over each post would no longer seem fitting in the landscape orientation of a desktop monitor, they moved it to the side of the posts. But now that bar must inherit the height of the entire post. The result is that what is an elegant little purple bar that stands out to helpfully demarcate and frame the content on phone, becomes an enormous dark purple eyesore on the side that needlessly draws the eye away from the content on a PC. I find it hard to read a post on desktop without that big purple patch continually pulling at my peripheral vision.

Of course, the template designer could have altered the CSS to give the desktop version a more appropriate font and a more bespoke header design that doesn't detract from the content. But we live in the real world, so they could have, but they didn't. Instead they focused on the phone format, which presumably is the most significant demographic for forums nowadays, and just shoehorned the rest. It just so happens that on this particular forum, 90% of the users probably view on PC, so get the shoehorned version.

Of course, the differences between a phone and a monitor are not so drastic, so though the forum is subpar, it's hardly the end of the world. But if smartwatches ever become the dominant form factor, and companies become more preoccupied with targeting that platform, you can expect your PC web browsing experience to become decidedly less satisfying.

I bring up this example, and my other pc/console-related ones, merely as a broad reminder that different platforms frequently have less-than-compatible needs, and that any hopes for a HTML5-driven one-size-fits all paradise should be tempered with a dose of pragmatism. Some of my examples were significant in the past but are less so today (eg consoles that can now play RTS games with kb+mouse) while some will remain relevant for the forseeable future (eg AAA games not taking advantage of the latest PC hardware because they're beholden to ageing consoles).

Some cases are not even technological, but cultural. The particularities of smartphone usage meant that phone gamers increasingly expected games to be free and gameplay sessions to be bite-sized, with developers responding by evolving free-to-play models that use addiction mechanics to entice spending or 'pay to win' mechanics to fastrack progress. These mechanics have sometimes crept into the PC gaming space, where I would argue they don't belong, and are detrimental.

A somewhat similar thing happened in the 80s, as you know. Most games had a '3 lives and you're out' structure that was born of the economics of coin op arcade machines rather than any inherent game design merit. Yet these games retained the life-based model when they migrated to Nintendos and Amigas, and it took many game developers an absurdly long time to unlearn it.

Retro gamers often praise the old arcade games' high difficulty. The irony is that some modern games like Super Meat Boy are much harder than anything from the 80s, precisely because you can die 1000 times without it costing you your mortgage. Dying almost becomes part of the fun, in a way that it couldn't be if you had actual money and the threat of "game over" riding on it. As a result, you can hone your skills in a latter level of Super Meat Boy to a much higher degree than you could in an 80s game where you only had 3 tries after a 30 minute slog to get there. In turn, the developers of Super Meat Boy could make its latter levels much more punishing than an 80s developer could have.

I'm aware that I'm really skirting the periphery of relevance to the original topic, but my point is just that we should be mindful of potential pitfalls before we embrace the hope of any universal single-system wonderland. Fragmentation is costly and often undesirable, but sometimes breaching those divides can also have undesirable consequences as people lose sight of the important differences, and as the idiosyncrasies of different platforms pollute each other. If HTML5 is the future, then so be it. But for now, I remain uncomfortable with the idea of throwing out the EXE format, and native OS integration to the world's biggest computing and gaming platform, merely for the promise of slightly easier integration with other platforms.

Quote from MuddyMole

I hope that's still true, but I sometimes wonder. You and I both came in at the beginning, when Clickteam software was far less capable, but also far simpler, and we've been able to learn all the new features gradually, as and when they were introduced. I can't help thinking that for someone just getting started now, it must be pretty overwhelming.
It's the same with everything though. Every now and then I think "maybe I'll just give that other game engine another try", but then I do, and I'm greeted with something like this:

Please login to see this picture.

...and I just think "Nope!", and immediately uninstall it again. I bet if I'd started using it 15 years ago though, I'd have been fine.

It's a bit of a pity that some of the DLC2.5+ features won't be seen by beginners, as some of them make working in Fusion easier and less confusing, such as Find All, Custom Qualifiers, or "Select all objects with a given qualifier". (Though I totally understand why they must be part of 2.5+...they were developed years later, and there's only so much work Clickteam can do for free).

But I agree with DJFuego that overall, Fusion's learning curve is terrific. The thing I find amazing is how much I dislike the Event Editor today - I find the way it obfuscates the actions makes it barely usable, and haven't touched it for over 5 years. Yet the Event Editor was a very big part of what attracted me to Fusion in the first place. I remember the Event List Editor looked dry and intimidating in the first year or two, and I'm sure I simply wouldn't have gotten into Fusion had the Event Editor not been there to visually break a project down for me into friendly, bite-size chunks. Needless to say, I would have run screaming at the sight of anything like what's in your screenshot.

Somehow the process of going from an Event Editor lover to an Event Editor hater (and Event List Editor lover) happened organically, with me barely noticing it - almost like some sort of magic trick. I think Fusion does a really good job of hiding extraneous information until you need it.

Quote from MuddyMole

Related to that, you might have noticed a number of threads recently, asking about the best way to save a game in CF2.5, and the honest answer is that there is no good way. It's always "You could use INI file, BUT..." or "You could use an array, BUT..." . In C3, every object has an expression "asJSON" which converts its entire state into a single JSON string - or even simpler, there's a function to automatically save the entire gamestate as a JSON string.

Actually it's not a clickteam thing. Serialization & Deserialization, aka save object to some format then recreate that object again, which is frequently used in web communication, so JavaScript, the script for webpage, has good support of this, also other VM based languages, as the type info & name info are carried to runtime. But fusion is built on C++, which has a long-time struggle about reflection, a language feature to make serialization easier, and that's one of the reason why fusion lacks this feature and only provides a bunch of callbacks for this feature.

The latest news about C++ reflection is compile time static reflection, in C++ 26 draft. Don't know if it's possible to get passed this time. CT said they are working on C++ 17 in F3, IDK if it's easy to move to latest version, as complex projects usually rely on a specific compiler version.

For my very limited TypeScript experience, some times you also need to extend default toJson/stringify & parse for complex objects, especially the one including map, to create independent objects, as by default all objects will share the same reference and change one will affect another, but much more easier than compiled languages like C++

	async loadFile() {
		try {
			const path = basePath + "\\..\\DubMapping.map";
			const byteArray = await vscode.workspace.fs.readFile(vscode.Uri.file(path));

			const data = new TextEncodingPolyfill.TextDecoder().decode(byteArray);

			this.dubMap = JSON.parse(data, reviver);

			this.bUpdated = false;
		} catch (err) {
			console.log(err);
		}
	}
	async saveFile() {
		try {
			if (!fileListInitialized) { return; }

			if (!this.bUpdated) { return; }

			const data = JSON.stringify(this.dubMap, replacer);

			const path = basePath + "\\..\\DubMapping.map";
			const byteArray = new TextEncodingPolyfill.TextEncoder().encode(data);

			await vscode.workspace.fs.writeFile(vscode.Uri.file(path), byteArray);

			this.bUpdated = false;
		} catch (err) {
			console.log(err);
		}
	}
	
	// for map serialization
export function replacer(key: any, value: any) {
	if (value instanceof Map) {
		return {
			dataType: 'Map',
			value: Array.from(value.entries()), // or with spread: value: [...value]
		};
	} else {
		return value;
	}
}

// for map de-serialization
export function reviver(key: any, value: any) {
	if (typeof value === 'object' && value !== null) {
		if (value.dataType === 'Map') {
			return new Map(value.value);
		}
	}

	return value;
}

Quote from MuddyMole

I hope that's still true, but I sometimes wonder. You and I both came in at the beginning, when Clickteam software was far less capable, but also far simpler, and we've been able to learn all the new features gradually, as and when they were introduced. I can't help thinking that for someone just getting started now, it must be pretty overwhelming.
It's the same with everything though. Every now and then I think "maybe I'll just give that other game engine another try", but then I do, and I'm greeted with something like this:

Please login to see this picture.

...and I just think "Nope!", and immediately uninstall it again. I bet if I'd started using it 15 years ago though, I'd have been fine.

You think that's daunting, try and use Blender back in the 90/Nauti's. The UX design and objects where placed everywhere and you had to deal with it plus NO UNDO! of all things and crashing was a given. Although it worked much nicer under Windows NT and 2000.
Please login to see this attachment.

The Fusion GUI has evolved with the times quite nicely and yes; for anyone who has never used software building programs before or come from a different interface I'll agree it all looks alien because its not what you're used to.
This forum is a wealth of decades of experience so don't be afraid to ask for help. Practice searching for your answer or signing up to one of the many training sites there are or take the free tutorials on Youtube. You can't get from there to here overnight but practice daily first make very simple games with the pre-defined objects. This will help you understand the simplicity of the event driven coding. My first game was a breakout (bat and ball) style game. I loved the old scrolly games of the 80s and 90s with the paralax sscrolling scenery. This is not beyond a novice level either.

The only thing I'm crossing my fingers for is Fusion 3 to finally embrace 64bit builds so we can break through that 2GB RAM barrier. If that means dropping backwards compatibility I don't have a problem but it needs to be done.
What Clickteam has accomplished up to now is a miracle in programming and I'm proud to have been a part of it.

Quote from DJFuego

The only thing I'm crossing my fingers for is Fusion 3 to finally embrace 64bit builds so we can break through that 2GB RAM barrier. If that means dropping backwards compatibility I don't have a problem but it needs to be done.

The non-Windows are 64-bit compatible, and the limit was boosted to 4GB back in build 292.27.Please login to see this attachment.

(Although the editor can have issues with >2GB, that's Direct3D problem; fixed by turning off "use Direct3d in editor" in Fusion settings. Built apps still use it fine.)

The main issue is the extensions would need rebuilding; honestly I expect upgrading the runtime is pretty straightforward, only slowed by use of magic numbers based on how big pointers are, e.g. a 32-bit pointer takes up 4 bytes, so a list of 5 pointers is 20 bytes, so if you base other things on 20 rather than a sizeof(pointer), then it's gonna break.

The issue is that exts will be looking in the old positions, and can't store the new memory, so you'd need a 64-bit app that starts a 32-bit app for interacting with those 32-bit exts, then the two apps stay in sync. It's doable, but a pain. Better to port the exts than that.

But I could build my extensions for 64-bit Windows in a few minutes. But since runtime won't have 64-bit pointers, it wouldn't communicate properly, so it's a pointless gesture.

Maybe Fusion 2.5 could introduce a 64-bit module, much like the Unicode spin-off they had before they merged it into the main thing; and just limit the Windows building to the 64-bit exts. I know Firefly would get a big limiter removed from that.

Quote from Phi

But I could build my extensions for 64-bit Windows in a few minutes. But since runtime won't have 64-bit pointers, it wouldn't communicate properly, so it's a pointless gesture.

Maybe Fusion 2.5 could introduce a 64-bit module, much like the Unicode spin-off they had before they merged it into the main thing; and just limit the Windows building to the 64-bit exts. I know Firefly would get a big limiter removed from that.

What I meant was exactly this. We need to shed 32bit as we did 16bit. Windows still has the main share of the desktop market. The overall Windows market is at least 90% 64bit based. Weather Clickteam honestly want to keep legacy ability to run on a 32bit OS is beyond me in this day and age.

Fusion should be 64Bit or thinking about moving to 64bit; this in turn should provide a 64 bit build extension type for appropriate build types and you can then build a native 64bit application.
If this has not been done because of the technical debt you are still in then it should be on the roadmap by now.
There is an inherent hardware barrier between the 32bit and 64bit which you cannot overcome any other way until the software as a whole starts to utilise 64bit.

I believe that if you ran a proof of concept (rebuild a game with rebuilt fusion dependencies as 64bit)you would instantly see the benefits. Firefly cannot be of much use under 32bit limitations and this is one of the reasons for its lack of uptake.

If one thing happens in the next year or so let it be a move to 64bit.

Quote from Fadzior

With improving capabilities and expanding AI implementations in various software, I think F3 should take an interest in this to create games using prompts

Oh no, Prompt: Make me FnaF

Quote from Fadzior

With improving capabilities and expanding AI implementations in various software, I think F3 should take an interest in this to create games using prompts

That is the worst with AI going the way it is. There are good implementations and bad implementations. AI is an OK inspiration gather but not very creative.
The bad ones are usually the ones to make a quick buck by using generative AI. Its built into web platforms to help create websites. There is no regulatory body of any AI at the moment as its a free for all. Even making images with AI is considered a cheat. Its a time saver. Its not hard to spot so if you do want to use AI in any of your development projects and that includes code; you should really vet or QA it.
Someone will eventually process all apps and code and discover that its AI generated and if you have not stipulated where AI was used in your finished product, chances are you'll be ousted or worse, taken to court.

I have seen AI do a lot of good and in medical imaging it is a god send for fast initial diagnosis. You still needs a second set of human eyes to validate the result though. Because the end result is what happens when something goes wrong? Where is the blame put? You can't take an AI to court.

Quote from Yves

F3 blog should restart very soon, fingers crossed, another dev has been hired earlier this year and this has speeded it up significantly. The schedule is currently being updated and the blog should restart when it's done. PS: the other games we are publishing / busy with are not only FNAF!

It's almost a year since this message. What's going on? Where is any information about Fusion 3? Like, how long should we wait for anything related to it?