At 10/30/11 03:13 PM, funkycaveman wrote: I dont want to post it on the forum for limited exposure reasons but if anyone wishes to help out i can PM you a link.

pm me.

Err guys, it might be worth changing your password on FlashGameLicense and sites using the same password.
Relevant blog post

Air 3.0 for iOS

Fuck. Me. It. Is. Good.

And by good I mean "it actually works". I did some test with iOS and Air 2.6 and I struggled to move to vectors on the screen. With Air 3.0 I've got 20+ layered objects, all alpha'd on top of each other and it's still running at 50-55 fps. If I was to start blitting stuff I could do what I want.

Air3.0 + Unreal Engine = thinking twice about Unity.

Also in the last month I've been promoted and my first iOS app went live. I am the king of all things.

At 11/4/11 01:12 PM, PSvils wrote: I'm always skeptical of Adobe, but if Air3.0 really works that well, I will definitely give it a shot! (You don't need a developers key to just start testing, no?)

For iOS you need to be a registered Apple Developer ($99 per year) but you don't need any registration to publish for Android.

You can test in the simulator for free, but it doesn't simulate performance. It basically just makes your swf the size of an iphone screen and response to touch events from clicks.

I'm still working on my silly A* demo thing. This is just a branch of my EasyStar demo. I am leaving EasyStar untouched. I am trying to turn this into an actual game now. I put physics in... and then took them back out... Well. I decided I like it better without them.

I have been fooling around with GAE and I had two clients being able to see each others movement which was fine. I was doing some polling and it was pretty inefficient. I might try to mess more with that but I want to try to flesh out the game some more. Maybe put in some little monsters that you can fight or something.

At 11/4/11 08:35 PM, PrettyMuchBryce wrote: You can test in the simulator for free, but it doesn't simulate performance. It basically just makes your swf the size of an iphone screen and response to touch events from clicks.

Even on the devices Mouse Events work. Tapping things will fire Mouse Down and Mouse Up events etc.

At 11/4/11 09:22 PM, Kirk-Cocaine wrote: Even on the devices Mouse Events work. Tapping things will fire Mouse Down and Mouse Up events etc.

Yes but I believe if you click in the simulator it will fire a TouchEvent.

At 11/4/11 09:27 PM, PrettyMuchBryce wrote:

At 11/4/11 09:22 PM, Kirk-Cocaine wrote: Even on the devices Mouse Events work. Tapping things will fire Mouse Down and Mouse Up events etc.

Yes but I believe if you click in the simulator it will fire a TouchEvent.

Ew.
Hey does flash support multitouch yet?

At 11/5/11 02:16 AM, PSvils wrote:

Plus it's my birthday! :O

You better be telling me the truth, because I created a thread for that.

At 11/5/11 08:05 AM, PSvils wrote: What's everyone up to currently?

I can't say for others, but I'm pretty much bored right now. I'll try to work something in flash, yadda yadda, and stuff like that.

I now have 1,000 posts! :D

At 11/5/11 02:16 AM, PSvils wrote:

At 11/4/11 09:20 PM, PrettyMuchBryce wrote: I was doing some polling and it was pretty inefficient.

Yuck! For realtime multiplayer?
You probably already know, but you gotta go socket server with that. Oh, and by the way, there are a few optimizations you can still do for your A* thing, like not using Array.sort(), Array.shift(), and Array.splice() (Not sure if you're doing that even though...), and also treating the Open list as more like an Object pool rather than an Array.
So instead of adding to the array, you do openList[ len] = newObject, where "len" is a separate variable denoting the arrays length (You manually increment it).
To delete an object from the array, you swap the object that's being deleted with the last element in the array, and decrement len by one.
Also, there's no point in sorting the entire array, since you only need the Node with the lowest F value. It's faster to for loop through it and find it that way than using Array.sort().

Hey those are some really good tips. thanks. Yeah i know tcp is the way to go, but I haven't played around with that much. I've never written my own socket server before. GAE has this cool thing called the channel API that can push messages to javascript, which can then call a method in your swf. This way you don't have to poll the server. I was thinking I could do some sort of fake realtime wherein it just gives you everything on screens current position every 250ms or so and then pathfinds them there. So it may not be completely synced, but as far as the player is concerned it looks like it is.

Plus it's my birthday! :O

Happy Birthday! :D

Hey the re-design looks pretty rad!

Sarcasm.

At 11/4/11 01:56 PM, Kirk-Cocaine wrote: For iOS you need to be a registered Apple Developer ($99 per year) but you don't need any registration to publish for Android.

Isn't it $25 to publish on the market? I think thats a perpetual unlimited apps?

Who cares about money anyway?
There is enough money for everybody on the world ok? Its about the things you do.

At 11/5/11 03:38 AM, Cryoma wrote:

At 11/4/11 09:27 PM, PrettyMuchBryce wrote:

At 11/4/11 09:22 PM, Kirk-Cocaine wrote: Even on the devices Mouse Events work. Tapping things will fire Mouse Down and Mouse Up events etc.

Yes but I believe if you click in the simulator it will fire a TouchEvent.

Ew.
Hey does flash support multitouch yet?

Yeah since 2 years ago. ;)

Just tried out FlashDevelop. Why didn't you guys tell me it was 394872 times better than Flash Pro's coding environment?! Flash Pro will only be used for vector image editing now.

At 11/5/11 10:43 PM, MSGhero wrote: Just tried out FlashDevelop. Why didn't you guys tell me it was 394872 times better than Flash Pro's coding environment?! Flash Pro will only be used for vector image editing now.

You were using Flash Pro for your IDE? oh jeeze X_X. Have you tried flash builder?

At 11/5/11 10:55 PM, PrettyMuchBryce wrote: You were using Flash Pro for your IDE? oh jeeze X_X. Have you tried flash builder?

Nope, what does that do? And yeah...no one told me about all these fancy things

I like FD,
But it misses some key features like:
- auto-formatting code (duh its kind of basic)
- static code analysis (you can make obvious syntax errors now) and you need to compile the whole thing only to find out you got a semicolon missing or spelled some name wrong.
- search options, you cant find MouseEvent for example through normal Ctrl + Space (you need to do Ctrl + Space + Alt to find classes throughout the whole class tree)

But then again its being made by volunteers so you cant really cry about these things.
Its a great product and its free.

While everyone is talking about FD et al.
I still use the Flash IDE to code in I guess 'cause I'm a bit lazy.
Though the only real quarrel I have with it is when the project gets big it takes ages to build since it rebuilds the entire thing each time. Does FD or flex have to do this as well or does it just need to compile the edited files (and all of its dependancies)?
Would be a big time save at the end of projects.

It looks like I'm going to have to remake my half-finished game in something other than Flash, which has forced me to cut out feature after feature to solve performance issues.

I'm not really sure what else I can do--I want to include all of my ideas but Flash simply doesn't have the power for it. Would it be a good idea to cut everything troublesome out and release a "bare" version of it, then later release the full game, all features included, but programmed in Java/C++/etc?

What a sad day indeed :(

At 11/6/11 11:24 AM, Archawn wrote: It looks like I'm going to have to remake my half-finished game in something other than Flash, which has forced me to cut out feature after feature to solve performance issues.

I'm not really sure what else I can do--I want to include all of my ideas but Flash simply doesn't have the power for it. Would it be a good idea to cut everything troublesome out and release a "bare" version of it, then later release the full game, all features included, but programmed in Java/C++/etc?

What a sad day indeed :(

Do you *really* need to do whatever it is you're doing that's slowing down the game so much?

I think it would be a good idea to indeed try and finish the game within flash's limitations, because y'know, having limitations always spawns creativity. You could perhaps try to replace some of the effects with pseudo versions of them that, while would hardly be as impressive, would give the same idea. Because really, when it comes to games I don't think it matters how the engine was made as much as that the player experience is a good one.

Also, the only problem with making an exe version of the game is distribution. You can easily get millions of players on your flash game, and monetize those as well. But for a downloadable game, things are much harder and you'll most likely always find a smaller audience.

Those are just my thoughts.

Also, I watched goats and cows getting decapitated today, what have you guys done?

There was blood, blood EVERYWHERE

I watched Life in a Day on YouTube, great documentary. Coincidentally, it showed a cow getting slaughtered.

Any of you guys have experience with Q learning or SARSA or some other reinforcement learning algorithm? I'm looking into it for a new project, but I don't quite understand "states." Like how general does the state have to be or what should I include in it. It's gonna be a fighting game, maybe with combos and stuff.

Q learning is somewhere in this article...
http://www2.econ.iastate.edu/classes/eco n308/tesfatsion/LearnAlgorithms.LT.pdf

At 11/6/11 11:59 AM, 4urentertainment wrote:

At 11/6/11 11:24 AM, Archawn wrote: It looks like I'm going to have to remake my half-finished game in something other than Flash, which has forced me to cut out feature after feature to solve performance issues.

I'm not really sure what else I can do--I want to include all of my ideas but Flash simply doesn't have the power for it. Would it be a good idea to cut everything troublesome out and release a "bare" version of it, then later release the full game, all features included, but programmed in Java/C++/etc?

What a sad day indeed :(

Do you *really* need to do whatever it is you're doing that's slowing down the game so much?

I think it would be a good idea to indeed try and finish the game within flash's limitations, because y'know, having limitations always spawns creativity. You could perhaps try to replace some of the effects with pseudo versions of them that, while would hardly be as impressive, would give the same idea. Because really, when it comes to games I don't think it matters how the engine was made as much as that the player experience is a good one.

Also, the only problem with making an exe version of the game is distribution. You can easily get millions of players on your flash game, and monetize those as well. But for a downloadable game, things are much harder and you'll most likely always find a smaller audience.

Those are just my thoughts.

The game is completely about the experience, which is why I'm hesitant to keep removing features/effects. Right now it's very BitmapData-intensive because I have a lot of special color transformations and various other visual effects. I feel like without these, the game loses its "atmosphere", which to me is very important. It needs to feel real and alive.

Potentially I could rework the entire engine one last time, but I've done that twice already and although it has improved performance quite a bit, the game still runs at 15fps on a majority of the computers I've tested it on. The target framerate was originally 30fps, but I lowered it to 24fps and if I go any lower I'm afraid I'm sacrificing too much of the game's beauty.

(one reason this happens is because my computer runs it at a solid 40fps with all the effects in place, so I've had a lot of trouble trying to improve performance without a shitty computer to test on)

At 11/7/11 02:53 PM, PSvils wrote:

At 11/7/11 02:26 PM, Archawn wrote: The game is completely about the experience, which is why I'm hesitant to keep removing features/effects. Right now it's very BitmapData-intensive because I have a lot of special color transformations and various other visual effects. I feel like without these, the game loses its "atmosphere", which to me is very important. It needs to feel real and alive.

Now I'm getting really interested about what you have going on...There are many ways to get color transformation effects in Flash, and some of course are more costly than others. You seriously don't want to enlighten us as to what you're trying to do? I'm pretty curious :)

I'm trying to tell a story and convey the emotions of the characters in the game using only sound and color--and to do that seamlessly has proven to be very hard for Flash to pull off. Hopefully that's not too vague ;)

At 11/7/11 03:48 PM, PSvils wrote: Alright, well I was thinking more in terms of filters you are using etc. :) But all is good! I'll just wait for the final release version to play :)

ColorMatrixFilter, BlurFilter, ColorTransform, BitmapData translations/scaling, blood splatter/smear using copyPixels and colorTransform, etc..

One reason I'd like to use C++ or something more powerful is the fact that Flash slows down quite a bit when constantly updating BitmapData multiple times per frame.