I'm 24.

Like R0J0 the first game creator I started using was Klick n' Play.

My brother had a game magazine from around 1996 that had a small ad about it, but I only discovered it in 2002 when I found the magazine in a box and decided to revisit it for some nostalgic feelings. I then found some websites and communities about game creation and started using it.

Not so long after I discovered its successor The Games Factory, then Click n' Create, and some time after Multimedia Fusion. I used a lot the MMF 1.5, never finished a game but did some small useful applications.

Those were good times. I remember being lurking around the Clickteam's forums. At that time Ashley (AKA Tigerworks) was one of the best independent plugin makers for MMF. I was not very active at that community but I made some small contributions. I think I was one of the first to introduce the idea of multiple action points by then. Unfortunately they never made it into MMF2, but luckily Construct did!

So after trying briefly MMF2 I discovered Construct. At first I was not very attracted by it since it looked almost the same as MMF2 and yet very buggy. Influenced by Blender success, what later brought me here was it's open-sourceness. And with it the idea that I could use my years of experience with similar softwares, added to my design knowledge, to collaborate in a community effort to make the perfect game maker.

I was already working in a new interface proposal for the yet to come new Construct, when they suddenly announced C2. Unfortunately, soon after Scirra decided to go commercial and closed-source. Still, I liked the community and the improvements Construct had over Clickteam's system. Since then Scirra proved to know how to have a very nice company-user relationship, without being too greedy like most companies do, so I sticked with it.

Don't know if you have access to control the alpha channel and the color texture separately, but if you do I have some features to suggest.

I thought about a "Clear to Transparent" action with transparency control. This would fill the alpha with black with the respective opacity making the entire canvas a little more transparent. If this name doesn't sound so intuitive it could maybe be called something like "fill alpha" or "clear alpha".

This would allow to easily create trail effects without the need to spawn multiple canvases or objects, also saving some resources. With an every tick "Paste object to canvas" followed by a "Clear Canvas to 10% transparent", you could simply create trail effects.

Another idea is to allow drawing to the alpha directly. This would be a powerful feature opening a lot of possibilities. Instead of creating new actions for this, a dropdown parameter "Destination" could be added to all the "Draw" actions with the options of "Texture", "Alpha" or "Both" (being "Both" the default). This respectively would allow to choose to draw to the color texture only, the alpha only, or to both (like it currently is). The color parameters could be automatically converted to grayscale when "Alpha" is selected.

And lastly, just for consistency, I would suggest either making the "Draw Circle" action to draw it filled like the "Draw Box", or make the "Draw Box" draw it outlined like the "Draw Circle" (although I prefer the first option).

I think that's all I have to contribute for now. Keep up the great work!

Just to add to the list I'm really missing it too. It is essential to most of my ideas involving physics.

I disagree about bounding box being the default. It should only be default when a frame doesn't have any collision polygon. Every other case the default should be to preserve the current collision polygon for that frame.

It's annoying when you have made a complex edit on the collision polygon by aligning it perfectly, then you need to update some graphics by reloading them and loose all the work you have put on it. It discourages me from updating my game's graphics.

And I must agree about the other suggestions like copy pasting. I also must add that I miss an option for snapping nodes to pixels when editing.

Since the latest release r72 C2 has a "Stop loop" action in the General category of System object.

*Ninjaed by Kyatric

Here's an example that picks the topmost object under the mouse and move it on top of the others when dragged. I thinks this makes it a more intuitive behavior.

Example

Make every segment of the belt as an object and use distant joints from physics behavior to attach them together. The rollers need to have physics as well with circle collision. Then apply a torque to the rollers through events to make them spin.

As an alternative you can make an object with an animation of a conveyor belt and make the other objects be pushed when overlapping it. This solution would be simpler, but less "interactive" though.

Construct Classic also has a "Overlap by Offset" condition that can easily serve this purpose. It's something I've kinda been wanting to see in C2 as well.

Yeah, I miss it too. I think this is more versatile than the Overlapping Left or Right.

Or use a conditional expression like: "loopindex%2=0 ? 40 : 180"

Example

All that's being asked for (if I understand) is convert to JPEG on export to reduce the download size for certain images... ...it just could be an option to further compress JPEG suitable images when exporting.

What I think would be really useful, and is what HotGod is probably suggesting, is not just a JPEG compression option on export, but a way to link an manage different file types as textures.

Like HotGod said:

All I asked was to keep the source file intact. If the game designer uploaded a jpeg it must have been in purpose. The same goes with Png.

I simply use both formats cause I know what format gives me the best possible outcome.

I share this very same desire.

What comes into my mind is the way 3D softwares manage textures. They normally have a texture slot that you can set to any image you want, using any of the supported formats. They normally also include some options like ignore or invert alpha, use mipmap, interpolation mode, etc. Then the image gets linked to the texture slot.

C2 does almost the same thing, the main difference is that it automatically converts any image you load to PNG. And that is what causes conflict with the designers intentions when using different image formats.

Here's a workaround

You'll probably have to recreate it for every possible occasion.

Thanks Noga, but this is not an ideal solution. If you look closely you'll see that the blue square overlap the background by a few pixels. To get it working just add another "set Sprite2 position to Sprite" inside the loop in my example, just like Rojo pointed. I totally overlooked the fact that the position was not being updated inside the loop.

*I updated my example if you want to see how to make it work.

Silly me. Thanks for pointing that out Rojo, haven't noticed my fail.

Edit: Taking the opportunity, I'm really missing a "stop loop" action like CC for this kind of situations. Any chance of implementing it on a next release Ashley?