I want to make something, instead of thinking about how to make it.

[troshinsky]

I´ve been using Multimedia Fusion 2 for a year and it dosen´t have that magical feature of Click and Play you are talking about. However, I have strictly no programming knowledge and it allows me to make the games I want. It´s pretty visual, very intuitive, and can get powerful if you are skilled enough, as Nifflas proves. It could be better for sure, even more visual and intuitive, but they are getting there, Multimedia Fusion 3 is on the works so we´ll see. I would love if you could edit the code in real time like in Unity...

[Hugo Bille]

I´ve been using Multimedia Fusion 2 for a year and it dosen´t have that magical feature of Click and Play you are talking about. However, I have strictly no programming knowledge and it allows me to make the games I want. It´s pretty visual, very intuitive, and can get powerful if you are skilled enough, as Nifflas proves. It could be better for sure, even more visual and intuitive, but they are getting there, Multimedia Fusion 3 is on the works so we´ll see. I would love if you could edit the code in real time like in Unity...

Kay, that's sad. The Games Factory may be better than Klik & Play and still do these things, but I doubt anyone would call it a worthy game development tool today. So yeah, these functions would be more of an inspiration than something to actually make games with.

[Michaël Samyn]

I would love if you could edit the code in real time like in Unity...

That only works in very rare circumstances. Most of the time a running game starts misbehaving if you change the code.

In Quest3D, on the other hand, this works perfectly! In earlier versions there was no separation between edit and play mode, even. That combined with its visual programming (which I do find essential and a million times more efficient than scripting -even if its is not necessarily easier) allowed us to make our two most complex games to date (The Endless Forest and The Path) and an alternative AI system (Drama Princess).

A new plugin for Unity, called Antares Universe, allows for real-time visual editing as well. It's still in beta but it works better than the scripting for me. We're making the prototype for the new version of 8 with it.

[QXD-me]

It's probably worth noting that the only reason the "shoot laser" and "destroy object" things worked (at least based on my experience using the Games Factory) is that they were considered standard things to do by the editor. All the assets that came with it had their own death/explosion animation and I think you may even have been able to assign "shoot" from the right-click context menu (or I could just be misremembering that). Doing anything other than shoot/destroy/change variable essentially required you to program it yourself (admittedly in words).

Also, all the programming logic for a given level was just stored as one big list and I'm not sure how well something like this would translate to a more structured context, e.g if you implemented something similar in Unity where would the logic go? You could probably make something like that work, but it's not quite as good as it sounds (though I haven't used it for years so I may not be remembering it very well).

[György Dudas]

I tried the lite version of Games Factory (which is coming from the Multimedia Fusion company). I like it, but I don't think that it speaks to the creative side of the brain... Instead of using a programming language, you use a GUI, but you are still trying to solve logical problems. You are still coding. I would think that there is not much difference to programming.

Actually, I have the feeling that I can be more creative when programming. Also when programming there are way more possibilities for unknown things happening.

Also, MMF and GF are great for platforming games (like the stuff what Nifflas did). I wonder if there are too many restrictions for doing other stuff...

I might need more practice to get fast with it... reminds me a lot of Wario DIY for the NDS. You could make very easy micro games. It got tedious when you wanted to implement a certain feature, but the components, the software bricks were not flexible enough, so you had to use a lot of invisible helping objects to get your logic done.

[axcho]

Ian Bogost's new project Game-O-Matic is a very interesting take on this problem. You make a concept map of things and verbs that connect them, and it randomly generates little games based on them.

You can see a video here:
http://www.youtube.com/watch?v=Y7KacgqA8cI

[axcho]

I finally got around to skimming through Bret Victor's Magic Ink, and it actually talks about the idea of "expressing general logic through specific examples" that I'd mentioned earlier.

There's even a name for it - it's called programming by demonstration. I'm imagining demonstration mostly by diagrams, but there could be other ways to do it.

We're getting closer...

[God at play]

It's well worth a thorough read.

Make sure to hit up Ladder of Abstraction if you get the chance. His famous video is simply logical conclusions of the principles he lays out in that article.

[QXD-me]

http://www.unity3d.com/ninjacamp/projects/10180

It would appear that someone at Unity is interested in Bret Victor, even if they did misspell his name. Although it dosen't seem that there's going to be anything close to his ideas in Unity any time soon, at least not officially.

[axcho]

Bret Victor has a new article up on his site, and it is so awesome:

Learnable Programming

YES!!

A lot of the stuff that we want, he talks about being a necessity. Now how do we make this actually happen...

[Kjell]

The solutions he proposes are still code-centric.

Making feedback instant & easier to "read" is great ( for programmers ), but you're still programming / writing syntax instead of working directly on the end-result.



Why type shape("rect",80,20,140,60); if you could draw a rectangle on the canvas instead.



People seem to have forgotten that "code" is ( binary ) data just like anything else.

[Michaël Samyn]

Some programmers -most?- are very hostile towards visual interfaces to programming. They keep finding better ways to make drawings by pressing the cursor keys on the keyboard, hoping that at some point the artists will admit that cursor keys are so much more convenient than brushes and pencils.

(while in fact it always feels like they are -consciously or not- excluding artists from the domain they believe is their own)

[vorvox]

I think that programming is fundamentally different from something like painting or music. I do both, and I feel a definite shift in my state-of-mind when switching between the artistic and engineering aspects of game-making. This is, however, during the development of a game engine. Once the game engine is completed, I'm going to try to make it so I can manipulate things using the mouse in real-time. From what I've tried of visual programming (A little bit of Puredata), it works better with simple relations, and it's very good for experimenting. I agree with what people have said earlier about real-time programming being a lot better than the process of changing something, compiling, and seeing what it does, like a film photographer. I think visual interfaces aren't more popular because they're pretty domain-specific -- good for visual or musical programs, but not for much else. I don't have much experience, but that's my feeling. I think a visual interface might be cumbersome for implementing something like AI, and probably wouldn't work too well for text-based games. As for the "pressing cursor keys to make paintings" idea, I'm not sure if you're referring to the procedural generation of art or something else. I think most big game studios have ways that artists can arrange things to their liking after the infrastructure is completed.

Edit:
By the way, there are a lot of neat real-time programming systems out there, many of them designed for audio/visual live-coding (programming as a performance art, a pretty new thing). The one I'm most familiar with is Supercollider, which I'm planning on integrating with a game at some point. Toplap.org has a lot of info on these.

[Michaël Samyn]

I have a lot of experience with visual programming. The Endless Forest and The Path are programmed visually. I also created an AI system visually -Drama Princess. Our newest Bientôt l'été is also programmed visually. The prejudices against it are wrong. Visual programming can be more powerful and more versatile, at least for some people. The only thing it is not is easy and quick. It has that in common with code-based programming. Programming is always complex. That's kind of the beauty of it.

[JRamon]

Hi all (first post), I have also much experience with Quest3d finding its visual programming a really powerful method, however it is actually abandoned by act3d:

(FerryAct3d) and it all comes down too..  Our efforts now go into Lumion marketing and Quest3D slowly transforms more into an in-house tool. In the future, only the hardcore Quest3D partners we have currently will keep access to it and we might stop releasing the program for individual sale completely.

http://lumionautics.com/interview-with-ferry-marcellis/

Michael, wondering what made you stay away from Q3d (you were right). Maybe the mono-platform?
What would be to your eyes perfect-visual programming? Something in between Q3d and Unity?