Posts by Ethan

I'm interested in a fix for this as well. I'm also having the problem on a project I finished, and unfortunately I don't have MMF2 installed anymore, so I can't try that route.

Here's a fairly simple and fast 1-reflection raycast example I made a while back. Please login to see this link. It can be adjusted to suit however many reflections are needed, but it's limited in that it requires more events in order to add more reflections. If you ever need a simple variation of raycasting, this could be it. I used the advanced direction object, but you don't really have to, and the surface object which is simply for visual purposes. Everything else is done with active objects.

There are some parts of that, I didn't check beforehand, so just ignore the last 4 lines. I don't know how to edit or delete the post either though, so I'm in a bit of a bind.

This seems like an extremely old topic, but I thought I'd provide an answer anyways. Even if TheSynapse solved his problem one way or another the thread didn't end like that.
One way to create an easy to use maze game exploits a feature of MMF2 that might have even been there before MMF2! The ability to add a backdrop of an active object, and for when that doesn't suffice, the overlay and surface objects. If you have an active object which stores multiple tiles either as separate animations or as separate frames, then you can cycle to those frames or animations during a loop. You start with an array object or the dungeon object, using the array object you can either load pre-built maps which use numbers within a 2D or 3D array to correspond to different tiles. A 3D array can be used to load tiles on multiple layers, so as to provide a parallax effect with different speeds or simply to provide a background layer, colliding layer, and an overlay layer. If you have sufficient math skills you should be able to provide algorithms which can generate a map on the array based on what you want. After you have your array setup (assuming you go this route) you can run a fastloop (it won't be that fast) with the number of points on the array, from here you can cycle through each of the points and start drawing. The other method would be to use the dungeon object (which I have not yet figured out how to use) it should allow you to generate dungeons with various sizes and grid sizes, and although I'm not sure why you can't simply run the "paint" event, but there is a separate even for pasting to an overlay. And that brings me to the main method of drawing which is pasting to backdrop and blitting to overlay. In the case of pasting to backdrop or adding a backdrop, the method is to set the position of the active object with the tiles in it, and then setting either its frame or animation to that tile and then pasting to backdrop or adding a backdrop. Neither of the backdrop options create a new object but simply add the object as if part of the background, from here we can say that it will be pasted as an obstacle, which will allow us to tell when an object collides with any of the places that were pasted to. The problem with pasting the active is it will disappear if you have a scrolling frame, and the backdrop leaves the window. Thankfully if you use "Add backdrop" it will remain even if it leaves the window. The problem with both of these is if the backdrop is not within the window it will not be taken account of for collisions or overlaps (there is an option for these collisions in the frame properties, but it doesn't work as well as it says it does). This is why I prefer to use the overlay object or the surface object. The surface object has all of the features of the overlay object and much much more, but while it is in view, especially at large sizes, the surface object will often slow down your application. This is why in this case I will suggest above all, the overlay object, with it you can use the dungeon object or blit an active object to it, and it will collide while parts are totally out of the window. All of this can be said of the surface object, but because of the surface object's additional features it isn't as fast. At some point someone apparently created a "Tilemap object" and a "Tilemap viewport object" but I haven't looked into these enough to see how they work in comparison, so I would suggest that you compare for yourself. These are my solutions to the problem of creating a tile-based maze game with optimal use of objects.

And Looki has replied saying that the Flash SDK makes it impossible to support Surface collisions which puts that out of the picture. Also with my only possible substitute being pasting actives into the background, I may find myself just not porting to flash as I have heard that could be a bit buggy.

As I can no longer edit this post I thought I'd reply after having received a response about this from Fvivolo. Flash Files are simply saved as cookies which cannot be deleted by anyone but the user of the browser. I didn't quite understand his explanation of Lacewing files, "lacewing will work with the server and notice that you need to define a cross domain". However, the array object simply saves its files as cookies. Flash applications do not have as much files access as I thought according to his explanation, which would show as to why the file object was never ported, which I no longer need regardless as I will either no be porting to Flash or I will be porting without the the need to delete files. As for the Surface object, I'm still looking for a substitution, most likely from Looki.

Normally, I would just bring this up with those who are clued into the topic in the Clickteam chat, but since the specific group of people have yet to join the chat in the longest time I have decided to post a topic about it in the SWF Exporter Section. The first subject I would like to bring up deals with how you can send and receive files through a flash application using the lacewing object, as well as saving and loading files with the array object in flash. This made me think it probably worked the same way as it did in an executable where it can just save the files right outside of the file location and then load them the same way, and although the actions being the same also makes it look that way, another clicker told me over Facebook that it probably uses cookies otherwise that would be horribly unsafe. If this is the case, then I will have trouble getting my game to work the same way in flash, but I'm sure someone can point me to a workaround for loading the cookies. However, the way my game works it also deletes the file so nothing is left to clutter up the users computer aside from the executable, which turns out to be a problem because I don't know how to delete a file in flash the same way, or a cookie as it may indeed be. The file object hasn't been ported probably because of this dilemma with cookies, but that leaves me unaware as to how the array and lacewing object indeed have been ported.

Additionally, I'm having trouble understanding from Looki as to how much of the Surface object has been ported. It might not be problematic, but I need to know whether blitting actives to positions, resizing the surface, and objects colliding with the Surface object actually work in flash. And if not is there a way to do these things with another object that has been ported to flash? Hopefully I can be filled in fairly quickly and easily on these things, but it probably won't be happening in the chat.

Something else I'd like to point out is that the ForEach loop is using the same object each time, but since it's a different instance of the object it has a different fixed value. So it loops through the first object for each event and then loops through the second object for each event. And therefore the ForEach Loop does indeed work the same as the fastloop, even though the picture might not show that very well. The ForEach loop did this: Event 1 object 1, event 2 object 1, event 1 object 2, event 2 object 2, and the Fastloop did this: Event 1, event 2, event 1, event 2, as you can see the event order is the same, the only thing that changes is which objects are being selected.

I asked about the order in which events will occur for fastloops and ForEach loops for MMF2, and nobody would give me an answer besides, "Test it". So I'm making this post, after a long time without posting, to show you the answer to my question, so you don't have to test it yourself. To start with, here's the .mfa file I used to test it with: Please login to see this attachment. and for those of you that can't open it for whatever reason, here are some screenshots of the setup for the test.
Please login to see this attachment.

Please login to see this attachment.

Here are the results of this test. Where the longer numbers are the fixed values of each of the objects being listed on each event, and the numbers are the number of which event is being run in the fastloop.
Please login to see this attachment.
Hopefully this has helped some of you in some way, if you have had a problem dealing with this but were unable to understand my explanation, feel free to PM me.

I agree with you 100% Nifflas, everything you said there sounds awesome, not sure how sensible a project of that size would be for anyone in the community, most of the big workers already have enough on their plates, been a long time since I posted in a thread like this, but yeah, great ideas, if someone could bring in some strategies to back them up, possibly StephenL or Phi, then we could be on our way to breaking through to games as exciting as Audiosurf and beyond!

Y i no get bday thread, i am on chat almost every day! ;D

Awww, now I have to get the iOS exporter and get into the android exporter beta just so I can have all the awesome icons... :cool:

Looks like time is up, I'll see what I can do with the information I'm still waiting on, and that which was suggested, with as little interest has been shown for such an article, I might just postpone this until a time when I'm not so busy, and so more people can take notice.

Thanks for the data, if I can't get a third person to volunteer by the 10th I'll definitely use this along with any information I can get by then.

I didn't really think about that, but I am controlling some major variables by comparing ratings to overall money making effects in each specific ad category. It will make a lot of things more clear, of that I am sure.

I've decided to balance out my expectations for this somewhat. First of all, I'm only going to require one more developer to help out, and after that I'll start compiling the data and putting together the article, provided of course that both the participants send in there data. As with all changes, however there is a reaction to this change, starting now, anyone who would like to participate has until the 10th of May to send in their PM.

This is not saying that I won't be taking data after the third person has PM'd me, and I am only adding a time frame because I want to help people get their time's worth ASAP.

Still waiting on three more participants, thanks by the way Eliyahu, I can't wait to see all the data you can provide, If any of you are willing to put forth the info, I've decided there's no reason to post it here, you can just PM me with the data like Eliyahu will hopefully do, and I'll be sure to keep your name private if you so wish. And there's no problem if you only use one flash ads provider, all participants are welcome!

I would love a comparison of five different flash developers. I'd like to see the revenue differences between Mochi ads, Newgrounds ads, Kongregate ads, Sponsorship through Flash Game License, and other ad types. This would let me pick the optimal choice for future games. And it doesn't just help me. A chart like this could help all the clickers hard at work on flash games. If five of you see this and are willing to put forth your revenue, sources, and overall player rating. I would be more than willing to write an article on this that I can post on in the click forums, as well as on the daily click and game builder of course.

This would help a lot of people see who to pick to make the most out of their time or if they made a great enough game based on their developer friends' reactions to go ahead and try for a sponsorship.

Ok, it finally started working for me again, after I noticed it was a problem with Box2D, and after updating to the newest version of Box2D. So there's no more need to fuss.

Well, I at leas thought this was the problem, but eventually I did another test, and figured out that it can't be a problem with the frame switching, because I copied the login frame from my application and made it just jump to the next frame, and it worked just fine, so what I'm wondering now is what could possibly be the cause of this, if it looks like it just stays on the first frame in my actual application, is it a problem with one of my objects? (Lacewing build 18 Alpha 3, the newest build of Box2D, the Surface object, the string tokenizer) Or is it a problem with my computer because the problem doesn't happen on my friends computer, I don't get it, and I know everyone wants me to just send my code, but I consider it very private, and I can't exactly just strip it down, because each of the objects are very integrated with the code, so I don't know what I can put on here to help you guys, maybe if there was just one of you, I might feel better sending them my entire code, so I think I'll just send it to Jamie, and if any of you guys have suggestions, then I'll try them.