How Viable is Construct?

Hello, I'm Kayin, questionably famous for the questionably designed game, I Wanna Be the guy ( http://kayin.pyoko.org/iwbtg ). Now, I made this game in MMF2 as my first real game. MMF2 seemed like a good idea at the time -- until I reached a point of cursing and screaming at it and its anarchonistic bullshit. I use MMF2, but I HAEEET MMF2. The sheer lack of qualifiers in behaviors was more than enough to enrage me.

That said, I am very very willing to jump ship here. I have faith in my geek friends and their open source projects. But I'm also a bit nervous, after being 'burned' by MMF2, which held IWBTG back in a lot of ways (Basically by rendering it a crash prone, slow program). I googled around by construct and, to my horror, despite being elegent and progressive, has very little coverage to tell me how good/reliable it is. You'd think all the 10 year olds who want to make games would jump on this.

This is why, selfishly, I am making a thread here, to ask honestly to the users of this program... Is it viable? Is it ready? Should I wait for it to be refined more? Are their any scary/crippling bugs, oddities, important missing features or anything horrorfying I should know about? I hate to make a self centered thread like this, but theres a strong lack of information for what seems like a brilliant piece of software.

Again I apologize for this thread, but I am very excited and optimistic and am totally hoping I can jump ship for my next project. Thank you.

[quote:33ds749w]This is why, selfishly, I am making a thread here, to ask honestly to the users of this program... Is it viable? Is it ready? Should I wait for it to be refined more? Are their any scary/crippling bugs, oddities, important missing features or anything horrorfying I should know about? I hate to make a self centered thread like this, but theres a strong lack of information for what seems like a brilliant piece of software.

Current Vr. is 0.95.2 so its still Beta and you can sometime see minor bugs, and some features are still not added like Path Movement as other stuff but its under quick progress. BTW welcome to the Forum.

I develop Construct, and just to repeat what Doppel said a little, the most telling thing is it hasn't yet reached version 1.0 - we're still in beta. I have to admit there is still one known case (although hopefully rare) where your .cap file can become corrupted and you can't open the file any more - although autobackup should help stop this from ruining everything. The aim is to have construct 100% stable and completely viable for at least semiprofessional game development when we reach version 1.0 - you could look at our release date history and make your own guess as to when that might be.

So while I would not advocate serious or professional usage of Construct until 1.0, I would welcome any dabbling and reporting of issues, bugs and those annoyances that send you fuming, so we can fix them for the 1.0 release. That's how 1.0 is going to end up stable and useful - with the input of our testers!

Hi! I used the demo of MMF2 and Construct in comparison is MUCH superior. Construct takes advantage of DirectX which means great visuals, speed and pixel shaders ( I'm not a tech guy I just know some terms lol)

Right now Construct is much more stable than the first incarnations of it, but the fact we are able to use it now is a blessing in it's own. Check out some of the bump mapping or 3D box demos to see what construct can do. It's very impressive.

I suggest you try it out and see if it suits your needs. Obviously Construct is still in development but Ashely has made AMAZING progress in short time!

Plus, this community is nicer

I am an MMF1 convert and, while that may not be as helpful a position to judge as one would like, I am happily working away on a project that simply could never be done in MMF or MMF2. I have my own list of bugs I'm waiting for Ashley to get to, but that's the thing - Ashley is FAST. All of the time I've been here, there has been a major release every week and a half or so. The developers are very active on these forums. The people are friendly and bright. When there is a problem it is fixed. When there is a request it is honored (please see the current feature to-do list).

For Construct's speed, graphical power and customization, I whole-heartedly suggest you give it a try. Make a few demonstrations. Try to push the current limits so that they may be expanded upon. And more importantly, check back with us regularly as we are here to create and help and bugfix just as you could well be amongst this young, fertile, and as-yet unjaded community.

hey !

i am the foreinger in this forum, as u will notice in my english and in the way they drown my treaths to the bottom of time.

But, if you want an honest opinion, i can do that. Althought i know most will not agree.

Construct is Beta as ashley states. I think it is Beta in the shaders, but still alpha in the events.

I also think that it is inevitable that the event sheets very nearby in the future will break with today,

and be incompatible with what is implemented today.

A short pro-contra list.

PRO

Construct seems to be very alive. New releases come quick. Major bugs get fixed fast. And it is all done by Ashley reading those forums. And thats a BIG plus.

A lot of events take "expressions" as "input". And thats the real power of construct. Thats what lifts Construct above all other "game makers" that i used.

Behaviours, the way objects behave, are plug-in based. And that will be in the future a great thing. Especialy because there is a SDK available, and one day there will be Docs for it too.

CONTRA

The bugs are not a contra. They are still opportunitys. Construct is in beta.

Picking and isolating objects can be a real pain in the ****. The events/actions system is a puzzle.

Its not documented yet. Things that u expect to work as triggers, dont work as triggers. The events can be done with the power of an expression, but the action dont carry that power on, because it will force you to "pick" an object. Where it is more logical to let the events pick/isolate, and keep the acions general. This way the action will work on every object picked by an event. And thats not the case now. And although Ashley is not open towards those questions, it will be necensarely to change

the events-system to bring construct on again a higher level.

I asked Ahley if he planned individual "event sheets". He did't answer me (as usual). So i guess not.

No individual event sheets are a major draw back.

I guess, you must complete the ghost tutorial to understand what i am mumbling about.

What do you mean by individual event sheets?

i got to go home !! .. lol .. work is done.

Private event sheets. As we have private variables.

If i want an object to multiply itself every time it hits a specific other object, and the multiplyd object wil do the same.

Then that is so much easyer done by givng the object a private event sheet.

Moment it gets spawned/created it starts its events.

Another example ?

An object that fals over the screen, when its outside screen, it jumps back to top of screen.

So much easyer with private event sheet, and very flexible. Easy to use, make and change.

See it as behaviours that u can write, with the powerfull tools given by construct.

The ways to use this are endless.

TheInstance brings up good points, but please understand where he is coming from. He has been an invaluable source of constructive criticism (no pun intended) when it comes to the deeper structures of the program. His scrutiny has led to many of the major feature additions to version 0.95, as well as many future additions to be made in the near future.

His portrayal of Ashley, Rich and the community in general is somewhat skewed, however. There has been friction between TheInstance (and his other usernames) caused by a language barrier that makes him sound more contentious than he means to be. This being said, he has never been left out of discussion or outright ignored. Instance's topics die most commonly in the throes of debate over semantics... an issue I believe is directly tied to said language barrier. What makes perfect sense to most Construct users sometimes goes against the language structures Instance grew up with as a non-English speaker. He is not wrong, really. His requests are valid, but ultimately it is Ashley's decision as to how events are laid out.

Now if we may please get back to the original topic.

He did't answer me (as usual)

If I don't answer a thread, it's usually because I don't have time

[quote:32kn1z8n]No individual event sheets are a major draw back

No they're not - I've read and I think I understand your suggestion for them, and they wouldn't make eventing more flexible. They'd be purely for organising your events in to different places. None of the things you have mentioned sound difficult, let alone impossible, with ordinary events!

Still (back to topic), another interesting point about being pre-1.0, is if you have your major frustrations or bugs or ideas, they have a decent chance of being implemented.

I'm very impressed by all this honest feedback and I thank all of you. The only thing that sounds like a real headache is the the events not narrowing down the scope of objects -- but I don't know how bad this actually is, so that'll be the first thing I test.

Anyways I'll be sticking around then, because even if I don't use it now, I'll hardly be able to wait to use it later. Thank you, everyone!

I'm honored to have you here oh lord Kayin, great creator of IWBTG.

Construct is becomming more and more stable, but it's still not stable enough to make a real game. I've been following construct since it first started to grow after the first break, and i've been trying to create games from time to time. I allways run in to some bug that makes it impossible, and one time construct crashed and ruined the .cap i was working on INCLUDING all the backups i had! (have you given up on fixing that .cap yet ash?). You should absolutely not start a big project with this program in it's current state. The best thing to do is to create small games that make use of certain things you want in your big game, and report all bugs you run into until everything works as you want it to.

Construct didn't ruin all your backups, object groups caused any file saved to not load. It's now fixed.

one time construct crashed and ruined the .cap i was working on INCLUDING all the backups i had!

Just to shed some more light on this, along the lines of what Rich was saying: we figured out a ribbon bar button corrupted the application in memory, which meant every time you hit save it saved a corrupt .cap file which would never load. I'm sorry this happened, but it's fixed now (thanks in part to your reports), and unfortunately these kinds of things happen with pre-release software

I just started using Construct a few months ago and was impressed by its general usability, the speed of new releases, and by how active the forums are. I have never used MMF2, but I have tried gamemaker and Construct is a LOT better. I kind of agree with the suggestion not to start a big project. I would suggest to make small programs that test different features you want in your program. If you can produce all features individually, then try implementing them in a larger program. Over all, I would say to try it out and if you find a bug, report it. Then wait for Ashley to work his magic and continue.