MMF3 legacy support - yes or no?

Welcome to our brand new Clickteam Community Hub! We hope you will enjoy using the new features, which we will be further expanding in the coming months.

A few features including Passport are unavailable initially whilst we monitor stability of the new platform, we hope to bring these online very soon. Small issues will crop up following the import from our old system, including some message formatting, translation accuracy and other things.

Thank you for your patience whilst we've worked on this and we look forward to more exciting community developments soon!

Clickteam.
  • Note - this is nothing to do with Clickteam and is just out of curiosity. What does the community favour?

    Yes - compatibility with MMF2 is more important than new features

    No - new features are more important than compatibility with MMF2

    Please think carefully before you vote - UBB doesn't support changing it afterwards.

  • MMF has to move on in many ways. After all, in my opinion, it should be completely redesigned - which leads to a problem: the runtimes. I think it would be too much of an effort to port all the runtimes again... but hey, if MMF3 was designed from scratch, it could be lighter and more efficient, and it would be less work to port the runtimes.

    Please login to see this link.

  • Quote from Looki

    MMF has to move on in many ways. After all, in my opinion, it should be completely redesigned - which leads to a problem: the runtimes. I think it would be too much of an effort to port all the runtimes again... but hey, if MMF3 was designed from scratch, it could be lighter and more efficient, and it would be less work to port the runtimes.


    yeah i think this too

    mmf2 still seems like 1.5 but with more features and is actually stable... but there are a ton of stupid bugs and limitations that are apparently really difficult to get rid of with the current code; redesigning it from the ground up to be a lot easier to modify and work with would probably be a lot better in the long run, and make porting the runtime engine and/or editor easier. apparently the extension api is quite limited in what it can access (for example i hear that quite a few things can't be done or are hacky in xlua [pretty much anything involving collision masks and image surfaces] because mmf hides necessary information from extensions)

    even if this trashes BC, though if it's designed well enough it should be relatively trivial to make a converter

  • Well my understanding from Yves is that that's how MMF2 works; by converting the file old files into MMF2's language and that's why the file suffix changes and the conversion is irreversable.

    Provided they do something along the same lines in MMF3 that'll be great. Unless I can port pre-existing projects into MMF3 I don't think I'll have any reason to buy the product and I think a lot of users would feel the same way.

    Please login to see this link.

  • I've recently started to unlock to the true power of MMF, and I got to say, it is impressive.

    On the other hand, I would have found a lot of that power a lot faster if the basic movements worked properly in the first place :/

    I love how easy it is to use and I want to keep that, but those default movements have GOT to go. I was hoping someone would have made some new ones by now using the clickteam movement controller, but it seems that idea was a bust, which is a shame.

    As long as it's as easy to use and customizable as it was in the past, I'm all for it, especially if will still support the java and upcoming flash runtimes

  • One thing does worry me, though. as with most things like this, changing the runtime completely in such a way that past applications would no longer be compatible would just be opening another can of worms. Try as hard as you can; there will ALWAYS be problems.

  • Quote from Crash86

    Well my understanding from Yves is that that's how MMF2 works; by converting the file old files into MMF2's language and that's why the file suffix changes and the conversion is irreversable.

    Provided they do something along the same lines in MMF3 that'll be great. Unless I can port pre-existing projects into MMF3 I don't think I'll have any reason to buy the product and I think a lot of users would feel the same way.


    Thing is, there are so many design related things that will never be fixed, optimized, or made more efficient if things in the existing system can't be removed or changed into stuff that old projects wouldn't load with. Think about it, so many features of the product HAVE to work the way it did in k&p for the sake of backwards compatibility to the previous product, and CT are better now than they were back then but are kind of stuck with 15 year old design decisions.

    I purchased MMF2 to make MMF2 games, not to make games to carry over to MMF3. And really, how many of the TGF1.5 projects was successfully carried over to MMF2 anyway? I heard a lot of people wanted to do it, but I can't recall having actually seen it happen. So many extensions could never be converted because it's authors were gone and CT didn't have the source, and even when they did, I juts don't know what happened with the projects themselves. If you start a game out in MMF2 knowing you'll need MMF3 to finish it, you kind of made the wrong decision.

  • I don't even know what Legacy is, however I think that what you can do is significantly more important than what it can do it with, there's no point in having it run on every operating system in the world if it can't even move this one pixel here to over there. ->

    I'm the Original Dragonguy! It's true, I really am the Original Dragonguy bigger and better than ever!
    Please login to see this link.

  • Quote from Crash86

    Well my understanding from Yves is that that's how MMF2 works; by converting the file old files into MMF2's language and that's why the file suffix changes and the conversion is irreversable.

    Provided they do something along the same lines in MMF3 that'll be great. Unless I can port pre-existing projects into MMF3 I don't think I'll have any reason to buy the product and I think a lot of users would feel the same way.

    Yes, that's true, it's not exactly the same, but very close.
    The whole storage system was kept, and most of the data structures (How to store a transition, frame, object...), too, only slightly modified and with new parameters and functionality.
    Porting KNP data to MMF data is much, much easier than porting game maker data for example.
    It would be possible, but very hard and utopian.

    Why game maker? Well, it works completely different than MMF - and MMF3 should not be based on MMF2, it should follow the roots, I guess, but especially internally, it should differ a lot.

    Please login to see this link.

  • I think the results from this poll so far speak for itself.
    MMF is great. But I think its time that it gets rebuilt from the ground up. I don't care if we need to port over our MMF2 projects to make it compatible with the MMF3 runtime. If rebuilt completely from scratch MMF3 should be much more powerful and if it can give developers better rights and access to core functions, extensions should open up many more doors also.

    I know CT are busy with other runtimes right now and probably are not even thinking about MMF3 as yet, but I just hope they listen to what the community are requesting. Plus once all the other runtimes are finished it will probably make it easier to start from scratch as they can make any additional allowances they need to support multiple runtimes.

    previously know as bigredron

  • Quote from dragonguy

    I don't even know what Legacy is


    basically, outdated stuff that may still be in use (believe it or not some people still use knp)

    in fact windows is one example of software that's hugely dragged down by legacy software; some companies run really old software that may be too expensive or infeasible to upgrade, and that software sometimes relies on bugs/quirks in windows itself. microsoft doesn't exactly want to lose those customers so they're forced to keep certain kludges in place even for new releases of windows (i believe the desktop refresh that happens when you start an mmf2 app is one of them)

    mmf2 is similar in a way, because the userbase is used to having backwards compatibility. however this means some things can't be fixed without breaking things (like Francois mentioned in that string topic; strings aren't totally destroyed for BC reasons)

    breaking BC isn't entirely bad because, as mentioned, mmf2 doesn't have every extension that 1.5 did, so if your project relied on those extensions and there's no suitable replacement, you're screwed. Plus, like i mentioned, it doesn't rule out the possibility of having a file converter, but it will mean you will have to change stuff.

    A complete rewrite would also allow adding large features that mmf is missing already or doesn't have a lot of support for, such as everything being global, a cleaner and more powerful extension API (and getting rid of the double->float->double conversion i hear about), and possibly structuring the click event system to be more like a traditional scripting language (even if it isn't a textual one)

    it wouldn't hurt if extension developers also made sure clickteam always has the latest source code of their extensions, so at least SOMEBODY has it in case the original developer drops off the face of the internet

    unfortunately, as it seems right now, workarounds are almost like an idiom rather than a temporary solution; a lot of these could be fixed by a complete rewrite, taking the suggestion list in mind (so requests and stuff can be built in from the start instead of added later)

  • I think it's time to move forward, and if legacy support has to go, so be it. Despite all the effort and compromise behind making MMF2 backwards compatible with 1.5, there were still enough differences and problems that I had to trash my work and start over. I know others who still work in 1.5 because the conversion process isn't adequate, and they aren't willing to start over. I'm sure lots of people have had smooth transition, but then there's others were the end result is the same with or without the backwards compatibility support.

    To that end, I agree with the comment that people making MMF2 games will continue to develop those games in MMF2, and if/when they decide to start new projects, then they will start using a new product.

  • I'm a bit confused... I kinda feel MMF2 is *new*, what will be different in MMF3 then in MMF2? (other then improved runtime (something I know is better to have, but seems like is more like MMF2.5 or MMF2 (fixed))

    Not that MMF is *broken* at all, it just is a little frustrating that with each release, it gets better, but we lose some extensions and some of the work we've been doing.

    I was able to bring a lot of my original cca's into mfa, I just hope the mfa's will be able to be brought up when MMF3 comes out...

    Hope this wasn't taken the wrong way...!

  • Problem is, it IS broken, the organization of conditions, events, actions, and expressions needs a re-design to be both more user friendly and efficient (many of them could be made fewer while keeping the same functionality and yet provide more power and be easier for beginners), the software needs to be either 0 or 1 based in a consistent way, the built in movements needs to be replaced (I'd suggest an event-sheet based way with it's own internal values, so that the community can easily create movements with events and save them in a way that a beginner can assign to an object without needing the alterable values to be arranged a certain way), and the built in objects could use a re-think. It's the need to carry over all the old mfa's that have prevented all these important changes from being possible at all.

    Don't get me wrong, I'm not really after a whole bunch of new features here. I'm after organization, optimization and efficiency.

  • I agree. MMF2 projects can be finished with MMF2 - it's still a good program.
    MMF3 should not be burdened by support for old projects. Maybe being able to import the graphics from objects of old projects would be nice, but that's kinda all.

    I was recently been told about a new software for school kids to learn programming that works pretty much like MMF with events, conditions, actions and drag&drop etc. but writes actual script code under the hood which you can switch to at any time (back & forth) - so after starting with click-like programming kids can actually understand what the code does and continue from there. Pretty impressive. Forgot the name though :( But I'm sure we'll be hearing of it soon.

    Well. I wouldn't mind MMF writing actual script code that you can write instead of events ;)

    <span style="font-style: italic">&quot;I'm not saying you don't know what you are talking about, but I don't know what you're talking about.&quot;</span>

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!