Should I use C2 for a large game?

I'm gonna get a lot of flack for this but..

It's not that those in this community don't have talent, skills, good ideas etc. or never made a large-scale game before. It's that C2 isn't designed for large-scale game development by any means. It's designed with the single hobby developer in mind.

Everything about C2 is internal. The level editor, the event editor, the objects, the animations, the textures, the sounds. It's practically impossible to use C2 in a team and you know what teams offer? Large-scale games.

We don't have massive areas and entire worlds or even (good) tile-based games because you can't use external level editors.

We don't have heavy dialogue,inventories,world-maps,customizable characters, etc. because you can't really use external files. Project files are a temporary workaround at best.

We don't have lots of sound effects and music because you can only import .wav PCM which must then be converted into an unorganized project-bloating super folder of sounds.

We don't have solid, bug-free, ready-to-distribute games because all of that relies on 3rd party software seemingly shoehorned into C2 and forgotten about.

"Oh, well just use a project folder instead!"

Nope. You still have to add everything to your game inside the C2 editor else it's not recognized. So if you're in a team of 5, all building levels - each member will need a C2 license - all of the project's plugins - and the latest build, just to put his or her levels into the game. That applies to *everything*. There's SVN but that sounds like even more of a nightmare and I still don't see how C2 will pick up anything if it is added outside of the editor.

That's not to say it isn't currently possible to make a large-scale game in C2 - I'm working on 2 right now - but it's a dreadful process and I'm not sure we're going to get much farther without a number of changes and additions to C2. I've already had to cancel two others.

I agree.

I have followed construct from nearly the beginning. I loved construct when it first came out, and to be honest I really enjoy construct 2. I have discussed using C2 for some projects with the team I am working with and what it really comes down to is that it is not possible for us.

I still make heavy use of C2 for prototyping ideas and throwing together examples to help explain my thinking on our projects. After we knew that XNA is essentially dead we decided to move on to a technology that we knew would be supported for quite some time.

We even checked out the game making software including C2 and game maker. C2 in our opinion shines for hobby games, and pretty much rapid game creation. Game maker... heh, why in the world would we pay 500+ for a terrible GUI that gave us a sub-par scripting language. We ended up setting on libGDX, and have not looked back. However, C2 is still a very valuable tool for prototyping and even far more fantastic for creating interactive multimedia advertisements and videos.

It's because you're thinking as "a team", what i can understand of course.

But for a single developper, as I am, which not means a non-professional project, Construct 2 is very good to my opinion, for a 2D game.

It's really open, and you can easaly do all the "classics" you can encounter in a 2d game.

I'm more a designer than a programmer, and i'm on my own, and even if i can deal with some code, i'm not at all able to write one from A to Z, and that's not what i want to do.

For people like me , C2 is a very good solution, and will definitvely be when the webgl will be supported on mobile devices.

After all, it's also a question of "philosohy" : i tried GameSalad, Strencyl, GameMaker, and others, but the only one that talks to me is C2, and I think it's because it's just in the same "logic" as a software that i use everyday for my work as a designer : photoshop. (simple work on layers, for example)

Otherwise, I would have turn to a "real" developp program such as Corona. But in this style of software, easy to program, C2 is the best one i found, for me.

I love C2, I know its limits, and more than everything, i know the limits of the devices i'm developping for (mobiles).

When the dynamic memory usage will be supported, with for exemple CocoonJs (before the end of the month it seems, or with Ejecta), C2 will be a very very good option to develop for mobiles, and so other platforms, to my opinion.

C2 is growing up every month, as the HTML5 games, and I think it will be very intersting now, and in the future.

It's a kind of bet, the same as the team of C2 did.

I bet on it too.

(sorry for my bad english lol)

Tokinsom - I haven't tried using c2 in a team environment, so I can't comment on that, but actually I agree with you about some of the problems you mentioned.

Keeping the wavs in the project seems to have no purpose and just wastes space, as they don't seem to be used for anything after conversion to ogg and aac (unless I'm missing something they're used for).

The level editor process is a somewhat cumsy in comparison to how smooth it was in CC. In CC, I could make the level editor save and then the level can then be immediately loaded. In c2, Saving the level results in it getting downloaded to my downloads folder, I need to reimport it into c2, make a new event for loading it because ajax can only select project files from a drop down menu and not a string, and then repreview to be able to access it.

I also agree that C2 should automatically add everything that's in the project's files folder when the project is loaded and previewed instead of needing to import them in c2's editor (or even maybe have a 'refresh' option for the right click menu of the filers folder as well). I think it's an unnecessary step, and it's obviously affecting you and your team.

Files should be able to be loaded with ajax via a string rather than a drop down menu. As it is every time a level is created another event needs to be made for it. This way we wouldn't need 100 events for load level1.json, level2.json, etc. We could just do load "level"&variable&".json" instead.

Some way to do file organization, a 'for each project file' condition and 'current file name' expression would also be nice. With file I/O like CC has we can know everything in the images folder on disk is an image, and in CC I can loop through with 'for each file' (in c2 there isn't any way I'm aware of to organize/loop through files like that) and load all the images into a palette that can be used to build levels from, without having to adjust the loading code every time a new image is added. Even just a way to loop through the project files in c2 and get the name of the current file to be able to compare the file name to determine the file type and what to do with it would enable us to do something similar.

Arima Actually you can use "Request URL" to get a project file by string e.g. "Room_A_1.tmx"

Ah, cool, I didn't know that. Thanks for the heads up.

NP. You still have to add the file in the editor once but you can overwrite it outside of C2 afterwards.

I used to import with .wavs, but now just import .oggs. Since that's all that's needed for nodewebkit and the proper browsers.

It's true C2 lacks a lot of freedom for modular elements and teamwork. Other than a small team where people make assets and provide them for a developer, it would be rather difficult to work as a team. But, on the flip side, it is much easier to make big progress as a single person or small team than in a lot of other tools.

As far as finding a game to champion, I think multiple games is a better approach. I have full faith in Arima with lp, but it still may be a ways off until we really get to see it. I intend no selfishness, but I also would submit my own game, Courier, as an example of bigger things that C2 can do--particularly on the rpg front. There are some neat platformers already, and they'll only get better, but RPGs are lacking (for lots of good reasons!).

At initial release, Courier won't be gigantic. It will be larger than any other rpg project I've seen in C2, and most non-rpgs, but that isn't the point. The point is to treat the web as a unique platform, so that means early + often, ever-evolving and expanding, and freely available. So where bridges are knocked out and landslides have blocked paths, those paths will open up over time (and over development) to creat an ever-expanding world. And being web-based means changes can be made to existing areas organically and automatically, too. NPCs might not always say the same things or even have the same quests, buildings can be built, businesses can open and close (or move), and none of this is difficult to do in C2. Users also don't have to worry about installing updates since it will be the latest version each time they play.

It works now, the game has quite a lot of bigger systems, an inventory system (though admittedly much simpler than something like fallout), pleasing visuals, high-quality music, and is currently already larger than most C2 games and set to ever-expand. I say all this not to boast, but to explain that things like this are very doable in C2 and it handles them quite well. The ONLY problem I've had with C2 is it sometimes throws errors (or crashes) after previewing repeatedly. But Courier certainly seems to fit many of the qualities of a showcase for C2, and it isn't alone. I just have the benefit of releasing progressively, so it can be played sooner than a game targeting the end credits before release. Though I doubt I would ever have end credits... Just keep adding content even after the main story...

Well, define large... because "Baxter" is at 82 MB right now, but the majority of that comes from the soundtrack, and another chunk of it from large PNG files. The entire soundtrack is embedded in the demo, but the PNGs really beefed it up. The final product should be a little over 100 MB.

Baxter takes the level design from platforming with some puzzles added, and the overworld is something you see off of Angry Birds. 5 Acts (10-15 levels each), 4 special acts (with 3-4 super hard levels each) so you're looking at about 55+ levels for the completed product.

C2 seems to handle it, but I have noticed that it takes longer to save progress and test it via browser now that I've gotten so far compared to just starting out.

It's true C2 lacks a lot of freedom for modular elements and teamwork. Other than a small team where people make assets and provide them for a developer, it would be rather difficult to work as a team. But, on the flip side, it is much easier to make big progress as a single person or small team than in a lot of other tools.

You know, I haven't exactly tested this myself yet, but I'm pretty sure C2 is actually quite capable of being used for a more collaborative type of project, especially compared to Construct Classic, old versions of Game Maker, and any ClickTeam product. The event sheets and layouts are all XML so that should have no trouble working with something like subversion and you should even be able to work collaboratively with multiple developers being able to handle merges of eachother's work in place as long as the placement of blocks in the XML don't change arbitrarily (and I don't see any reason that they would).

Obligatory link to How to collaborate on projects with SVN