Idea for a future construct

tumira

That was part of the whole idea of C3.

A way for third parties to add stuff like that.

It's too "minimal" for me....

I don't like the "spaghetti" , and when you see the bar, you can't see everything well at the first time, you have to look carefully, and the icon too, too minimal for a software like construct, you have to see everything well...

construct2 is easy e perfect, maybe something can change.... but not a revolution

Ashley

Can we have timeline animation in the editor ? Basically we can do animation like tween, move,scale and etc by just using timeline or animation feature like

http://booty5.com/booty5-1-8-7b-out-now-flash-style-animation-brought-to-html5-developers/

http://booty5.com/html5-examples/timeline-animation/index.html

Since godot has it and also booty5 just added it into their open source editor.

I hope this will be added in C3

TQ

like flash? I was using flash for a long of time, and I love it, but there is something to do....

It's easy if is possible to set a timeline for every sprite, like, if you insert a sine, a little bar (timeline of the sprite) appear on the top of the layout, and you can set from event, the frame of the single sprite, change the position......this would be awsome...

with this, you can make a sprite with costum movement, and set from the event what "animation" you would like to set.... this can be perfect for arcade game,

make sprite > set animation on frame 120, if the animation go in 400, return in 120. this is like flash

Ashley

in future, would be possible to make a sprite like on flash? Where is possible to mask,insert another sprite, and insert event?

thank you

newt

Yup. But I don't want to wait for third party to do that. Since we need to pay to upgrade to C3, why not we have it officially. Better support and more features.

Ribis

I think Godot game engine already has some of that( If im not mistaken). You can add animation to the node or object and timeline will appear. I did not explore it fully yet. Trying to finish current game first.

Interesting idea, but it is much more limiting than events and most of the behaviors would be ambiguous what they would do imo. The main tricky bit is it becomes harder to reference values you set in some levels of your node tree, since the tree is run in a way depending on what's connected. Events are simpler in that respect since they are run top down.

That said it could be an interesting tool to make complex motion equations, but probably not in C2. It would also need some designing to flesh out what nodes you could have and what they'd do. Behaviors for instance wouldn't work so well being put anywhere because they depend on the object itself.

I do like the concept, so I made a capx to play around with it.

—, awesome as usual

Seems like this would be a bit conflicting with the way the IDE works now. Flow chart/node systems are great for some types of logic, and terrible for others. This brings us back to one of the improvements for C3 that most of us can agree on: a UI/Engine SDK would make systems like this possible for those who want to use them, and optional for those who don't.

That being said: I would love to be able to initialize instance variables and obj/behavior properties with global variables (with autofill and auto-name change in full effect), or even the returned value of a function.

I'm thinking of all plugin properties having a pulldown where we can choose from a list of all available global variables, or define a function name. Mind, the values aren't 'locked' to the global value, they simply use it to initialize the given parameter when the object is created.

Other random idea:

Wouldn't it be awesome if it were possible to build IDE extensions (widgets, windows, what-have-you) inside C2 itself? They could then be exported as self contained files and shared. These extensions would run as NW.JS apps inside C2, where they could be docked or free floating. Each project could have defined in its propeties which widgets it was using, and capx files would include all used widgets with it so people could still trade work sessions easily.

I'll confess I'm a bit short on specific use cases for this. Just seemed like a cool idea, figured I'd share it. Remember, right now, we're all just brainstorming, hoping to come up with some good ideas that might strike Ashley's fancy.

R0J0hound

Love it.

The question is, can that be done where it doesn't bog down the system?

Plus given the past experience with driver issues it kind of takes things that are more graphical, down in the want list.

What would be interesting is a live preview within the editor. Even better, the ability to edit the live preview in real time.

What would be interesting is a live preview within the editor. Even better, the ability to edit the live preview in real time.

Oh yes, especially the latter. We can kind of do this with debug, but holy hell is it clunky.

Here's what I imagine: being able to drag a slider to change values for a litetween object, or the accel/decel of an 8direction or car behavior, while actually playing the game.

I guess we can do this now by adding our own debug overlays, but the whole idea of construct is to make building games faster and easier, and encourage experimentation and discovery. Being able to modify values in realtime with immediate visual feedback would be a great improvement.

Interesting idea, but it is much more limiting than events and most of the behaviors would be ambiguous what they would do imo. The main tricky bit is it becomes harder to reference values you set in some levels of your node tree, since the tree is run in a way depending on what's connected. Events are simpler in that respect since they are run top down.

Well, out to in - if we as users can figure out multiple behaviours multiple controllers wouldn't be too complex either.

That said it could be an interesting tool to make complex motion equations, but probably not in C2. It would also need some designing to flesh out what nodes you could have and what they'd do. Behaviors for instance wouldn't work so well being put anywhere because they depend on the object itself.

I do like the concept, so I made a capx to play around with it.

Wow, I had an intention to make a visual demo to show the SPEED this gives you but you have gone one further - a working example! You also included right side inputs which I had omitted to keep things simple.

Now how anyone can say this isn't cool and easy is beyond me - a couple of clicks and you have complex movements that would take minutes to type in and that's if you have a perfect idea of what you already want.

Also - when you have behaviours if you remove one it'll wreck the entire system, all the events will disappear and good luck reworking that. In a system like R0J0 has shown here you just relink a couple of nodes and are good to go.

Input controllers could be introduced the only function of which is to provide values from events.

--

It is great that this thread has at least spawned some discussion - one real hope for C3 would be an open system that lets you make plugins that make something like this possible and not just have a couple of values to awkwardly affect using events. C2 has many things that have been done to a little bit over the minimum - image editor, tilemap editor, even the UI of the event system (drag a subevent under the bottom event on the first try, I dare you). If nearly anything could be modified by plugins those problems would be a thing of the past.

So Ashley's going to have to figure out a way to do scripting for the edit time.

Will this be using event logic as well?

I like Somebody 's idea of having the object parameters exposed and manipulated visually with nodes. It's like having all attributes available from withing the editor for an initial setup state. Afterwards one could manipulate the hell out of them with events. I see no conflict here, other than more development time for the developers.

I am afraid that Lately I see Ashley being fast to reject ideas coming from forum members. Perhaps it's just me, having no experience from other game developing forum and not being a very old member my self. I know that answering to all us is time consuming and perhaps most of the time a mundane task, and I really appreciate and honor his effort. It's just that I'm noticing either a "it can't happen" or not commenting at all policy lately.

There are many-many things that float around and probably inhabit the todo list, but here are (again) my top 5!

1) Timeline: A way to animate-timing up stuff within the layout and be able to trigger those animations with events. Perhaps this can replace the path system feature request. If not,

1.5) Path System: A way to make stuff move along a path over time, with the ability to control the position/path percentage, the curvature of the path and the easing in/out between each path point.

2) Masking: A proper masking system that allows any Sprite or Layer to mask in/out any other sprite or layer, taking into account either the opacity/alpha or the Gray-scale values.

3) 2.5D work space: Parallax offset/scale is just annoying. It happens from 0,0 coordinates and it's just unintuitive to frame/compose a complex level this way. A workspace that has z-depth and a "real" camera will allow for a natural compositing workflow.

4) Lighting system: A proper and Global lighting system that has a Diffuse-Normal-Specular setup within the Sprite Object and illuminates-casts shadows accordingly. I feel graditute for the system that we have, but it's just not there yet.

5) Audio enhancements: This is very well covered on other topics, the thing is that the audio workflow needs to be greatly improved. Synching multiple audio tracks is impossible, there are no mixing/bus tools and many-many other things that could vastly improve the outcome audio-wise

+1 for real time preview-editing!

Lastly, since this is a recurring subject, I haven't voiced my opinion about native exporters for a long time, but after seeing this on this topic, I must say WOW! This is a huge difference! I don't know about fps or any technical mumbo-jumbo, but it just feels so damn fluid and good! No jitters, no jerkiness, no wait a few seconds for the code to settle (in layman's terms), no nothing!

Anyone noticed that the mouse doesn't lag behind? In C2 can anyone manage to make "every tick---> Sprite set position | mouse.X, mouse.Y" and not having the Sprite lag behind the mouse cursor?

Anyway, I don't know which of the above fall to the "New Editor" or "New Engine" category, so I'll probably be flamed for being naive and out-of-topic. As an artist though, these are the things that stand out for me and I would like to have them in this game creation tool.

As an artist though

I totally agree with everything else you said, including the quickdraw "meh", but the part I quoted might be the most important one - there's plenty of artists/designers who come to C2 for the ease of use - they are told "Make games without coding" and while that isn't really 100% true the most annoying aspects of real programming are taken care of which is great.

But oh so often I get the feeling that a great many things are done "for the checkmark" - as previously mentioned - we got this and this and this. Neither of which is truly fully featured. Which makes sense since a single person is making the thing - a single person cannot really be fully invested into all possible elements of making a game - be it art, sound, animation, etc.

Which is another reason to hope for some sort of a modular system that would let those who can and want to remix the systems, at least those that supplement the event system - like the image editor, animation editor, tilemaps, etc.

I am afraid that Lately I see Ashley being fast to reject ideas coming from forum members. Perhaps it's just me, having no experience from other game developing forum and not being a very old member my self. I know that answering to all us is time consuming and perhaps most of the time a mundane task, and I really appreciate and honor his effort. It's just that I'm noticing either a "it can't happen" or not commenting at all policy lately.

I don't understand why Ashely have to change the design of construct2, and take the idea to have the "spaghetti"... for me is unclear when I see the line with the little windows...You can't see how the "script" works... When you make a real script, It's like writing, in construct2 you see your script very clear, you start from the top, and just write, write, write.... with "spaghetti"? (same for behaviors)

1) Timeline: A way to animate-timing up stuff within the layout and be able to trigger those animations with events. Perhaps this can replace the path system feature request. If not,

2) Masking: A proper masking system that allows any Sprite or Layer to mask in/out any other sprite or layer, taking into account either the opacity/alpha or the Gray-scale values.

+1 for real time preview-editing!

Agree with you, timeline can make simple a lot of things, but I suggest to insert a special behaviors for every obj, when you add in your object, just make a timeline on the top of the layout screen, where you can make the action "stop, play,loop and insert a specific event on it)

Masking system, it's the first thing I though when I Use construct2 at the first time, I was using flash so, you can't make a special things like flash...the best method for the sprite, where you can place sprite inside a sprite with your code, and masking everything... but I don't know how can be difficult make in construct2 (sprite in sprite)

Lastly, since this is a recurring subject, I haven't voiced my opinion about native exporters for a long time, but after seeing this on this topic, I must say WOW! This is a huge difference! I don't know about fps or any technical mumbo-jumbo, but it just feels so damn fluid and good! No jitters, no jerkiness, no wait a few seconds for the code to settle (in layman's terms), no nothing!

Anyone noticed that the mouse doesn't lag behind? In C2 can anyone manage to make "every tick---> Sprite set position | mouse.X, mouse.Y" and not having the Sprite lag behind the mouse cursor?

native export is the best solution for run a software... because when you export your project the script is "light", and java is not very good for the performance....(like flash, in my opinion), if you make a simple test, one in construct2 and another one in lua,c, or other script, the app will run better...this is one of the most reason why who want to make a BIG game, have to use others engine (like unity), where you can make your script in different way and export in native...

obviously, when you use construct2 you have to know how make everything good without waste memory or something like that...

> As an artist though

>

I totally agree with everything else you said, including the quickdraw "meh", but the part I quoted might be the most important one - there's plenty of artists/designers who come to C2 for the ease of use - they are told "Make games without coding" and while that isn't really 100% true the most annoying aspects of real programming are taken care of which is great.

Construct2 is the best software without coding, and the performance is very good.... In construct2 you can make a fantastic game/app/... without study a real programming, but when you use it, if you had the logic of the programmer is like a piece of cake... ask someone how is annoying just for make a sprite, animation with collision in real script ...in construct2 you can do it in two click, and I try a lot of software (some for fun, and some for work)...

the artist is advantaged in this case, because it's easy to make whatever someone want in construct2, and the most difficult part when you are using a visual script is the arts/music (I spend more time to think about arts and music....)

I am afraid that Lately I see Ashley being fast to reject ideas coming from forum members.

If I see problems with suggestions, I will point them out. If people argue convincingly against me, I will consider what they say. Also as ever we get far more feature requests than we can possibly deal with, some of which are multi-month projects, so it is important to set a high bar and not get sidetracked with loads of experimental ideas that may or may not work out.

In particular new UI features are very time consuming to develop, and what is outlined here is probably several months of work. There is a great opportunity cost there, since if it doesn't work out or few people end up using it (like with multiplayer), then we've lost several months work that could have gone on something more useful.

So I can see how it might look like I'm trying to shoot everything down, but really it's how I test ideas to see if they're worthwhile.

So I can see how it might look like I'm trying to shoot everything down, but really it's how I test ideas to see if they're worthwhile.

So we can assume you see no merit in the often-mentioned Boolean global variables idea? (As one simple example of decent ideas from users that haven't seen any response). If so, why?