What if you're really playing the super high res version of ProgressQuest?
cue illusions of free will

As I already stated, I simply explain the concepts and offer advice.
You could debate endlessly what I mean, but thats better left to after I'm dead, since I'm right here

Yeah pretty much, I never really make a point I just explain stuff :3

It does seem as soon as you mention Minecraft a lot of people take it as an oppurtunity just to make it clear how much they don't buy into it and think it's overrated. Never enough just to enjoy and promote the things you do like, gotta let everyone else know they're wrong for having fun. Haters gotta hate.

Or do one better and say something constructive that could actually be helpful to the game, instead of just saying "I LIEK THIS" or "I HAET SHIT". Its well and good to have people advocating a game- that is how grass roots exposure is built and how indie games publicize. But you can do one better by being constructive.

I do understand the concept, What I don't understand is your particular argument. As a matter of fact, Minecraft does have goals and structure, and not just "ostensibly". Sure, you could argue that the game could be "tighter" and guide the player more (which is, as you said, a forthcoming feature), but saying it completely lacks any structure is just plain false, sorry.

I say ostensibly with much emphasis. If a game has a definable goal, then Minecraft's "Goal" right now is "Build stuff". Not "survive a zombie onslaught by building stuff". Those elements could be built upon a bit, and don't need to be the emphasis, but they're sure as heck not developed. Imagine it having a campaign where you have to build your base to survive and/or beat back hordes of enemies.

I never said the game doesn't have any structure to it- I said it has underdeveloped structure that can be built upon. And to say that the foundations it has right now qualify as a developed game would be false- ostensible.

I think gilbert timmy has a good evaluation of it.

Not to be a dick, but what was the point of your post? Everybody knows it's an elaborate sandbox at the moment, and that's what a lot of people thoroughly enjoy. It's not a "complete" game yet, but that doesn't mean people can't enjoy it...

The point is that much like a sandbox, the game itself can be built into something greater.
Theres no law stating indie games have to be *worse* than their large title equivalents. If you take the raw clay and apply some game design and build something solid out of it, you can make something just as good as any quality title. Of course, this is tigsource the internet and its filled with a load of unthinking wankers that couldn't tell the difference between derision and constructive criticism and would jump on any comment not groveling praise to the almighty as heresy.

Or in more metaphoric parlance, its the distinction between building your castle out of tinkertoys and giving it life.

And the developer- not wearing his pants on his head- already acknowledges this and as previously mentioned plans to add some kind of narrative to give the game focus. And it will be great to see what can be built out of it.


If you're still not understanding the concept, look at Scribblenauts. Its a wonderful engine with amazing scope and sandbox potential. But as a game its driven by progression and loose storyline. The developers intentionally chose to focus on the emergent gameplay and sandbox appeal over strict wiring in keeping with the engine and gameplay. But they knew well enough to give it proper structure to guide the player and give the game focus. Because as many would argue- a game without a definable goal is not a game. And Minecraft already has the tools and base elements in place to implement something of a structure. Again, it will be nice to see whats built out of it in the end- right now those structures could only ostensibly be called a 'narrative'

yeah, ostensibly so.
if that was developed on into an actual story arc and game you might have something
right now its just elaborate sandbox.

Game design has long since known that pure sandboxing does not actually work. It does not stay enjoyable. When you play the sims, you have an objective. You need to keep your sim alive- it is a simulation (duh). This follows in the same spirit of older concepts such as tamogachi. If the sims just had you building your house and leaving it cold and unpopulated, it would never have flown off as a massive hit. Grand Theft Auto 3 is a sandbox- but it guides you through a story arc and linear gameplay just as well as any other game. It is open ended, but inexorably follows a linear path. Again, herding the player along.

This is a simple concept; restriction and authority breed creativity and choice. When your players go to a Baskin Robbins and hear there are 500+ ice cream flavours, which one will they choose? They will stand there for 10 minutes with no idea what to do. But if you put a giant sign in front of them listing the 5 best selling flavours, quickly they are drawn towards those, and the rest stay as esoteric choices to give expansion appeal. This carries over as well in game design.

Minecraft is raw potential. It is not a competent game. But it could definitely serve as the basis for one.

I hear Braid was pretty well known for its soundtrack whether its really indie or not

Mathias announced at the last klik conference that there is Python support for MM2, too, as well as the already existing lua interface. The HWA is long since stable and finished. Theres no understating the value of having multiple compilers- I've taken the same games I created as downloadable .exe's, and rereleased them on newgrounds as flash, and embedded them as javascript in a webpage, and even run them on Mac & Linux machines natively with HWA.

Construct holds a lot of promise and Klik holds a lot of already existing content. And I respect both of them for the hard work thats been put into it. You're really bogging down with the whole contrarian oneupsmanship attitude. Theres a lot of code thats less of a headache to write in Construct. Theres a lot of code thats impossible to write in construct that can be easily done in MMF2. Theres plenty of community go around for them both, and its foolish for anyone to naysay one to eschew the other- theres more common ground than not and loads of functionality to both.

They're both growing projects and always getting new tools and compilers and editor reworks. You can't just point out the way one is getting better without the other.

Actually MMF2 has a handy ForEach object now, which cuts down many previously O(N^2) algorithms to O(N). Its true that it was very late in the coming, and was dead necessary. And I'm sure theres lots of things Construct does better- nobody doubts that. I'd probably use it too if it had the support for backwards compatibility with what I've already created, the vast third party extensions and multiple compilers. Its not at that point yet- thats simply something that takes more time. And just as Construct is improving, MMF is improving too- likely having all those features in MMF3.

They're both equally valid ways to work, and theres definitely things better suited to both of them.

Well, I try to be reasonable, but you two are just leading this argument in circles from topic to topic and acting surprised when that gets pointed out. So I'm done. I guess I'll just go make some games or something. Since I know Construct and flash, I can make both flash and non-flash games without beating my head against MMF2/TGF2 and all the fun limitations that totally aren't there, you guys.

Also, Pixelthief, you could have been a bit more subtle. Arriving right on time to take up a position that matches up with that of another person already knee-deep in an incredibly fruitless argument.
Not at all suspicious.

Well I have been active in the community for over a decade since long before tigsource existed, but to 2-for-1 that bit about 'limitations', please refer to this:
http://forums.tigsource.com/index.php?topic=14574.0



Unlike our resident trawls and misanthropes, I'm not here to argue. I'm a professional, a computer scientist, a software engineer and foremost a game designer. I'm here to explain. If you don't understand the context of software development and the thought that goes into language choice and modern development cycles- I can help explain them. If you don't understand concepts like abstraction and turing-completness, I can help you understand. But in general the best way to illustrate a point is by example.



You know, I've never understood this bitter Construct divisiveness. Am I the only one who appreciates the hard work that both Tigerworks/Ashley and crew have been doing, as well as people on the other side like Yves, Jeff and major 3rd parties like Mathias & Greyhill & so on. The whole contrarian attitude is extremely juvenile. Not cutesy adam being intentionally juvenile-esque, not Phizzy being a wanker style, but a rather undignified and unmerited bitterness that is unbecoming of what should be professional attitudes. Or in short: You turds.

Not going to read most of what you said because you're talking about standalone MMF2 while the rest of us have been talking about the MMF2 (or TGF2 in this case I guess) flash interpreter.

Obviously a multi-platform renderer/VM trades off performance. What the hell made you think anyone was questioning that? Or you you just trying to word bomb us? Seriously man, you're in the wrong thread if you think this is a code vs authoring tools discussion.

Oh, now thats curious with all the "Construct Comparison" going on, considering construct can't export .SWF files. Flash is just one of many compilers for MMF2, many existing and more to come that I've been helping with. When you create a Flash in MMF2, you can later port it to every other format instead of being stuck in just one- that alone is worth something whether you plan to use it or not.

You haven't changed all that much :S

[single sitting]

MMF2 *is* a programming language, one that comes encased in an IDE.
Its just a high level interpreted language, just like Java or Actionscript or yes, C.
Its just one more layer of abstraction. This is how computer science works.


I don't think you understand the tradeoff that abstraction makes in computer science. Each level just increases the speed of development while decreasing the overall efficiency. That is the tradeoff at any and all layers of programming, and is not unique to MMF2. MMF2 lets you make applications faster than C++ while your applications will be *weaker* than C++, in the same way C++ is faster/weaker than C, and C is faster/weaker than machine code, and so on.


I could create the same projects in C++ as I have done in MMF2, and perhaps with more raw efficiency allowing me to do more features. But it might take me 10x as long to create those applications- 10 years instead of 1 year for a single person. That is why C++ is the language of choice for professional development in large software engineering, while something like MMF2 is more suited to indie development and personal projects. When you've got 100 coders working on a project, it can afford to be made in a lower abstraction language.

For example, a game developer at a professional level might prototype several parts of his game in MMF2, show these to management, and then flesh out the actual implementation in C++ with fellow coders. And yes, this really happens- MMF2 and similar tools are used in professional design. Once you've learned how to use MMF2, creating something in it might take the smallest fraction of the time it would take in C++; I can create an entire game in a single sitting instead of taking months of development with multiple programmers. And I've created *many* games in single sittings, mind you.


What tool you use in software engineering is like everything else in life- dependent on context. I would no more try to develop a largescale one-person indie game in C++ than I would try to cut down a tree with a herring wrapped in newspaper. And I would not try to create the next big multiconsole AAA title FPS in MMF2 with a team of a thousand.

And more importantly, the fact we're having this conversation at all sort of rules out you being in the latter category. So perhaps you should evaluate the kind of tools you are trying to use- all you're doing is getting the bark greasy.




Now I've got a platforming engine made in MMF2 that can do everything Braid did, everything And Yet it Moves did, everything every metroid and castlevania game ever did- and thats not even half of what it can do. The list of features stretches multiple pages, from a full scripting language to 360 degree platforming to object-independent time travel. And it runs comfortably in MMF2/HWA.

If you're a 1 person indie project evaluating "what the limit is on MMF2", well, the limit is a lot higher than anyone is giving it credit for. Rule of thumb is- if its in 2d, make it in MMF2, if its in 3d, make it in something else ie C++ (flash/construct/etc wouldn't be suitable anyway). Actually, if its in 3d, please don't attempt to make it as a 1-person team, that never works.

Funny, I see people in this thread praising construct, yet friends who have experienced it complain how it's still incredibly buggy and unrefined, and therefore somewhat untrustworthy.

Ive always said it's the skill of the person, not the tool that is usually the problem. So many people are quick to attack a product because of their own mistakes in the programming somewhere along the line, or lack of ability/knowledge. It sounds harsh, but I really think it's true. People also like to point out the negatives before doing their research and seeing the positive outcomes, like all the great games that have been made using click products.

If games like Noitu Love 2 and Knytt Stories can be made in MMF2, along with up coming games like Sky Invaders, Pitiri 1977 and a bunch more, then I see no reason why most of the users on TIGS would ever feel the need to complain.

Just because great things have been done with it doesn't mean the software is great. Gotta give some people credits for overcoming limitations, but wouldn't it be much better if there weren't limitations in the first place ?
Konjak here brings another point, wouldn't those games be even better if they were made using other software. My guess is that the answer for both questions is yes.

My project might be more powerful if it was written in C++
It would be even more powerful if it was written in C
It would be even more powerful if it was written in machine code.
It would also not exist.


Each level of abstraction sacrifices performance for development time. This is an importance concept in software engineering. Nobody programs in C- its like teenage sex, everyone thinks everyone else is programming in C, nobody really is, and the few that are, suck at it and think it will be better the next time. Thats why object oriented and 10x as usable C++ is an industry standard, not soul-crushing C code.

Software development is not about "What can I program?". It is about "What can I program on-time and on-budget, realistically?". Just because something could be more powerful in another language doesn't mean it should be in a nother language- it probably wouldn't exist at all.

Yeah, see, in serious programming people don't describe languages as 'for newbies'.
Ironically, only one demographic really hops to that....


Yah its perhaps tidier and faster to use subevents but so nominally so in the latter thats its really just syntactic sugar. And I've definitely espoused over and over again that MMF should have subevents, but its really not a dealbreaker. Maybe I'm just so used to coding without them, but I've never really found myself wishing I had them.

Id easily say that in terms of efficiency MMF2 can easily be well up beyond what say Flash itself is capable of

You really, really need to elaborate on what you mean by "efficiency".

Flash compiles to bytecode within a sandbox that trades processing efficiency for massive exposure, being available on 98% of computers or whatnot. MMF2 compiles bytecode with higher overhead but no such VM in order to allow rapid prototyping. By extensions in MMF2, you can create just a large a scope projects as you could in any other language.

Or in short, "What could be done in flash" is a subset of "What could be done in MMF2" which is a subset of "What could be done in C++".


The biggest problem that newcomers to klik products run into is that they do not understand how the object scope list works, and wind up with high order polynomial time complexity algorithms where much more direct implementations might apply. And thats how you wind up with people going "BUT IT RUNS SLOOOOW"

Construct is really no different. The addition of subevents its only a minuscule drop in overhead and syntactic sugar over doing the same thing with multiple events. It would be great if MMF3 had them, but I've never found myself wishing too hard for them even in the most complex projects, and theres nothing you can do with subevents that couldn't be done with regular events. Now Ashley/Tigerworks taught me half the stuff I know, and I'm not one to favor MMF2 or Construct. Its simply that "one works and is stable and is backwards compatible with everything I've written", and "one is a WIP". It will be all well and good when it exists, but theres no particular features or efficiency it can promise that skyrockets it ahead- its just about equally powerful. And especially with the huuuuge amount of 3rd party (extension) resources available in klik already. You get into that situation where "Yeah, you could rebuild it from the ground up and have it nominally more powerful, but lose all the preexisting content", which applies to virtually all abstractions of software development.




Now I can jot down some pseudocode onto a piece of paper, and convert it into C++ code or MMF2 with about equal hassle. If the question is "what can be done in MMF2", well I've managed to make my project in it, and I'm no A+ computer scientist. Its foolish to nag on any of the languages people have mentioned (MMF2 / AS / Construct / Etc) as being "weak" or "for newbies" as anyone can do basically anything in any of them, provided they know what they are doing. I thought modern programmers had moved beyond that foolishness. Thats 1980-90's bickering

Id easily say that in terms of efficiency MMF2 can easily be well up beyond what say Flash itself is capable of, stronger than actionscripting for sure. Just because people aren't tapping into the full potential of a language doesn't reduce what it can do. We don't write programs in machine code with 100% efficiency- even C is a HLL by that comparison.

If I'm capable of creating an engine like this:
http://forums.tigsource.com/index.php?topic=14574.0

Entirely in MMF2, then you can be capable of doing anything.


This is yes just an elaborate demo- and a very useful one too, as it allows NG users to get access to a premier flash development program for *free* with full use as to NG. Those who are interested by it and learn to program in it should really look into getting MMF2 or MMF3 when that comes out, given that they can develop many more things than .SWF, from cross-system .exe to java to a few not even announced yet.


In terms of software design and computer science, any turing complete language is capable of anything. Now some IDE's are going to hold your hand and draw your graphics and animate everything for you (Adobe), some are going to ruthlessly punish your soul for even the slightest verge from perfection, and some are going to give you a good middle ground higher level language that allows you access to a good enough chunk of efficiency if you understand how the compiler works, while still streamlining development enough to let you prototype things completely implausible in more 'ruthless' languages.



To exploit any language to its fullest, you need the mastery of it, and thats something you'll only get by studying computer science. If you want a language that holds your hand and bakes the entire cake for you and creates the next big indie game with little to know coding expertise or graphics no-how; it doesn't exist. Unless you base said pre-generated game on the columbine massacre or something, ofc ofc ^___^

Nice engine! Its always great to see games based on gravity mechanics! The whole idea of using ray tracing to check the surface normal of a block is actually very similar to the way Super Mario Galaxy did it and the algorithm I'm using on one of my current

. If you're interested in reading up on it, you might want to check this out:
http://www.gamasutra.com/view/feature/3593/games_demystified_super_mario_.php?page=1

hey thank you all for comments and offers and all. Everythings put together now
Moving into main development now! May have a prototype out at some point as a demo

To be honest, what you really need is some kind of unique gameplay mechanic to create that upendedness in a zero sum game- something for players to exploit to give them an advantage. This could be anything. The best way to find the one that works is to try different ideas and see how they turn out.

It could be something as mundane as a double-production base (SC2 recently did this), or it could be something as complex as allowing you to draw lines between clumps of units, positioning them to enclose and destroy units trapped within the (polygon?).

Just play around with a few and see how they ring