PDA

View Full Version : Lets talk OUYA



WILL
05-06-2013, 09:27 AM
Hey everyone! I just wanted to show my face around here so people didn't think I'm hibernating under a rock or something.

No in fact I've been actually developing one of my game projects. (and I have a baby girl now so that takes up a LOT of time too)

Paul and I have been really pushing ourselves on Subject 33 and trying to get our game running and working on Android in preparation to get it running on the OUYA. A brand new gaming machine that I hope will be delivered to my home within the week. I know of only 1 other Pascal developer that has one of these consoles and Paul is going to be getting one soon so we can both test and tweak the game to perform better on it.

At only $99 USD it's highly affordable too! OUYA website (http://www.ouya.tv/)

Has anyone else gotten very interested or is as excited about making a game for this console as I am? I want to see how my games perform on the Tegra 3 (http://en.wikipedia.org/wiki/Tegra_3#Tegra_3) (T33) as well.

My tool of choice is of course Oxygene for Java (http://www.remobjects.com). I can't think of a better tool myself considering the cross-platform capability it provides me and the fact that it builds, installs and debugs right from the IDE onto the device. I'm not a fan of Visual Studio, but it's a pretty cool setup for the most part and I have my game running on an Android phone (Samsung Nexus S) and both Paul and I are in the process of trying to speed things up for better performance on the mobile CPU.

When I finally get my OUYA and get the game running on the device I'll show off a post here in this thread.

laggyluk
05-06-2013, 09:55 AM
interesting platform for indies, would be nice to release a 'console' game someday ;)

Dan
05-06-2013, 10:50 AM
I have ordered it a few days ago as well. I'd like to see if I can make my gamedev framework work there=)
overall it seems like a very friendly platform for both the gamers and developers.

laggyluk
06-06-2013, 12:49 AM
btw what are current 'pascal tools' for android?
building in lazarus seems like painful process, oxygen costs money, any other options?

WILL
06-06-2013, 04:58 AM
btw what are current 'pascal tools' for android?
building in lazarus seems like painful process, oxygen costs money, any other options?

That's basically it; FPC only, Lazarus or Oxygene for Java.

FPC and Lazarus solutions would definitely be a pain, esp. for mobile and OUYA dev. The way to go for Android and OUYA is definately Oxygen for Java, esp. if you seriously want to get a game on these systems and not be struggling with the tools. Also you cannot debug on the hardware it's self so you wouldn't be able to weed out any stability issues to make your game really shine as easily.

Honestly when it comes to the whole costs money part... how worth it to you is it that you can actually do this? Are you serious about making games or really not? Think about this, if you buy console a console game it goes for about $60 for each new game you get. Buy 14 new console games and that's about the money it takes to get the best tools to make your own games on a console.

Cost breakdown for equipment used by us at Red Ant Games to develop and test Subject 33:



1 OUYA + 1 controller: $99 + $30 for shipping (to Canada) *
+1 extra controller: $50
Oxygene (includes Java, .NET and Cocoa platforms): $699 *
Samsung Nexus S Android phone (second-hand used): $75
+1 extra OUYA + 1 controller: $99 + ~$30 for shipping (to Australia)


Our Total Equip cost: $ $1082 (USD) (* All you really need: $99 + $699 = $798 )


Now you don't really need all of that stuff, you'd just need the OUYA and controller for development and testing and I still highly recommend Oxygene as it's a great tool and you aren't just getting for some platforms, but ALL of their compilers' platforms so with a little bit of work you can port your games over to the App Store for iPhone/iPad and Mac as well as all the .NET stuff too. So maybe XBox Live Arcade if they are still supporting .NET on their new XBox thing.

But that's it really, for less than the cost of a decent PC rig you can make your own games on a pretty cool new console.

Content is a whole other issue, but if you have someone that is committed enough and wants to do the artwork then that's another expense it'll save you.

laggyluk
06-06-2013, 06:53 AM
so far making games is a hobby of mine. I like pascal but not enough to spent $$$ if I can make same game in unity, openframeworks or whatever. hopefully developing in fpc for android will be smoother till I get to making next game :P

Dan
06-06-2013, 07:37 AM
I'd really have to disagree with your representation of fpc/lazarus. sure, it's not a walk in the park where you simply install the software, make a new project, click run and it works on the device. fpc requires some advanced knowledge of its tools, but once you have that you'll see that it is very powerful.

one indisputable advantage that fpc has over any java based tool for android, is that it can compile native code for the platform. I speak out of experience here, because before I started using fpc for android development I did try to make a game using java. Half way to completion I ran into some really bad performance issues and that is already using a multithreaded architecture to separate the logic, rendering and input. in the end I decided to abandon the whole project and restart it using fpc. you can now find the finished game here: https://play.google.com/store/apps/details?id=com.pascode.gemmasterdemo and I am planning to release a new game for android very soon.

and yes fpc and lazarus are free, so if in the end you find that it's just not working for you then you won't loose anything.

"Are you serious about making games or really not", I'd say that if you are then you should not be afraid of investing some time in learning your tools. once you have fpc/lazarus and your gamedev framework properly setup to build android games then it becomes as easy as clicking two buttons (once to compile with fpc and once to assemble an apk and run the app). you can see the process here: http://www.youtube.com/watch?v=dr8QSAnSl1E

Darkhog
07-06-2013, 11:40 AM
In any case, how one would compile Pascal game for Ouya e.g. from Lazarus or even directly FPC?

Dan
07-06-2013, 04:33 PM
as far as I can tell it's just the same as compiling for an android phone/tab (arm-linux), since tegra 3 is a quad-core ARM processor. I should have some fun with that as soon as I have the console.

laggyluk
09-06-2013, 08:06 AM
I feel like devils advocate here :P i'm still making a game in pascal you know? hoping that it will be a good game and that people will like it. in the process showing that pascal is not dead.
What got me started on this thread was statement that if I think seriously about making games then spending almost $1000 on a Oxygene is a right way to go. So my point is: I rather switch to c++ than invest that kind of money into pascal development just because I like the language. just not rational imo

Daikrys
09-06-2013, 09:03 AM
as far as I can tell it's just the same as compiling for an android phone/tab (arm-linux), since tegra 3 is a quad-core ARM processor. I should have some fun with that as soon as I have the console.

correct, you just need to implement the conroller handling via ouya sdk and thats all :)

Some post content moved to more appropreate discussion thread here (http://www.pascalgamedevelopment.com/showthread.php?21839-Another-Pascal-vs-C-argument)

WILL
10-06-2013, 12:29 AM
it is still by far not a bad choice. but what is THE best choice then?
btw we're going way off topic here=)

Oxygene. :) You can debug in the IDE right off the device as it runs. Hands down that makes all the difference. To compile for Android you need to make a seperate project from what you would run off your desktop to develop initially. And since the hardware requirements are vastly different for Android devices and of course the OUYA how else could you tell how well it performs or fix any platform/hardware speicifc issues unless you can debug while it's running in the exact environment you will deloy your app on.

Lazarus is great, I love it for desktop development on both Mac and Windows, (and I've had some fairly decent experiences on Linux too) but for mobiles it's not useful enough. You *can* do it, but it doesn't mean tyou *should* do it. :)

I wouldn't use Lazarus to make games for mobiles, I just wouldn't be able to make them stable and fine tuned enough for what quality I would want to produce games, so I can't suggest that others ignore this as well.

The right tool for the right job afterall.

So in saying that, you want to use Oxygene, especially if you plan on selling your games for profit. And the OUYA "Store" will be a great way for indies to make some money off their work. More so than any other platforms out there. Though Google Play and iTunes App Store are good markets, but more saturated than OUYA's offerings.

I don't think that even Ecplipse can debug on device. Please correct me if I'm wrong, I would sincerely like to know if that's not the case. (Paul and I have even considered, but have not looking ointo Eclipse.)

WILL
10-06-2013, 12:35 AM
correct, you just need to implement the conroller handling via ouya sdk and thats all :)

We have found libGDX to be excellent at this! capturing mouse and touchscreen actions has been very easy as they are treated the same. So you can test on desktop for single finger touch using the mouse. Multi-touch obviously can't be tested off-device, so testing on device is necessary. You won't need to worry about about this on OUYA unless you plan on using the touchpad on the controller.

And there is this great post about using libGDX to receive OUYA gamepad input in your existing apps/games: http://www.badlogicgames.com/wordpress/?p=2733

There are others that may work really well, but Paul and I have had only great experiences with libGDX (http://libgdx.badlogicgames.com/) so I would recommend trying it out. If you guys know of any others post them here! :D

WILL
10-06-2013, 12:40 AM
And lets keep the C/C++ vs. Pascal discussions out of this thread please. This is here to talk about the OUYA and making games (or apps) for it. Thanks!

I made a new thread for you guys to continue your conversation here: http://www.pascalgamedevelopment.com/showthread.php?21839-Another-Pascal-vs-C-argument

WILL
10-06-2013, 12:56 AM
In any case, how one would compile Pascal game for Ouya e.g. from Lazarus or even directly FPC?

Here is a link to the FPC Wiki, but I don't know much about the specifics. http://wiki.freepascal.org/FPC_JVM_Android_Development Again, I wouldn't use this tool as I can't debug while it runs on the device using FPC. (Unless there is another tool or someone made a tool.)

WILL
10-06-2013, 03:09 AM
1165

Testing Subject 33 on an Android device from Oxygene's IDE. :)

It's a Samsung Nexus S phone in case anyone is curious.

WILL
10-06-2013, 03:56 AM
The developers of libGDX and Spine showing off some details on using libGDX with the OUYA.


http://www.youtube.com/watch?feature=player_embedded&v=WHdyF8xcY6g

It's a Dev Kit version that was sent out to backers who were willing to pay $1337 USD for one. So it doesn't look as nice as the final one. (More bare-bones.)

Dan
10-06-2013, 04:51 AM
Here is a link to the FPC Wiki, but I don't know much about the specifics. http://wiki.freepascal.org/FPC_JVM_Android_Development Again, I wouldn't use this tool as I can't debug while it runs on the device using FPC. (Unless there is another tool or someone made a tool.)
to take advantage of fpc on android you should really forget about JVM or DVM and instead go with pure native code. here is the easiest way that it can be done (this is a fairly new article): http://wiki.freepascal.org/Android


I don't think that even Ecplipse can debug on device. Please correct me if I'm wrong, I would sincerely like to know if that's not the case.
you are wrong. Eclipse can debug the java code on the device out of the box. Put a break point and wait for the program to hit it.

native code debugging is a bit more complicated but definitely possible. here is an article on how to do the setup for native debugging:
http://mhandroid.wordpress.com/2011/01/23/using-eclipse-for-android-cc-debugging/

so what exactly is the advantage of using Oxigene?

WILL
10-06-2013, 10:53 AM
to take advantage of fpc on android you should really forget about JVM or DVM and instead go with pure native code. here is the easiest way that it can be done (this is a fairly new article): http://wiki.freepascal.org/Android

I couldn't disagree more. Java bytecode IS native on Android. Which CPU architecture would you support? And what others will you not let run your game? Android was designed to run Java not CPU specific code. If you want to be 100% native you need to generate Java bytecode.

Your suggested method of creating Android apps would leave other speific hardware out in the dark and that's not a good idea considering all the different hardware manufacturers that make Android devices out there. And who is to say that there won't be a change from ARM to x86 or something else?

The idea of only "native code" is the way to go should have went out the window when you decided "hey I'd like to make an Android game/app." Google made this a design choice when they designed the OS.

I can't subscribe to this idea as it's just making extra work to get the "sense" that you are making a better app than taking full advantage of what the Android OS provides as a platform. You shouldn't have to work so hard just to make it fit on a select few devices.


you are wrong. Eclipse can debug the java code on the device out of the box. Put a break point and wait for the program to hit it.

Ok thanks! I was going to look into it eventually, but I hadn't gotten around to it.

Dan
10-06-2013, 12:34 PM
Google made this a design choice when they designed the OS.
Yep and then they had to redo it all with the NDK=) because the "native" java code was way too slow for any performance intensive app. as for different cpu architectures, so far ARM dominates the market of the android devices and even if there will be another cpu in the future recompiling your app for that architecture will just be a minor task. overall considering that you get significantly increased performance by using the true native code, I'd say it is worth the trouble.

Daikrys
10-06-2013, 09:23 PM
have to agree that true native code will run alot faster, for example my games made with zengl run really good on my superpad 2 (which is a cheap chinese made pad) and the java games run very poor even if simple made

Dan's game also runs smooth enough to play, which is true native code?

WILL
11-06-2013, 12:35 AM
I would tend to agree with you guys on the principle, however that's just the way the OS was designed. :)

On the note of Java being slow, I should point out that originally I was dead against it for making games having experienced it long ago in my own attempts to learn it's language. Times since then have certainly changed however and taking full advantage of OpenGL/OpenGL ES has seemed to help it a lot. All apps on all your Android phones are a testament to how improved performance has been in Java bytecode. Minecraft is another example of efficient and fast Java. (considering all the data and chunks of network data it has to deal with)

I think that the idea that Java is too slow for games is an old idea and if the VM that runs it is fast, so too will be the game/app that it runs on. Case in point Another World is written in interpreted code and it ran extremely well on old DOS and Amiga computers.

A lot of technology that we used to think just stunk has seemed to have gotten a lot better over that last 10 years.

Darkhog
11-06-2013, 02:15 AM
Yeah, but still native code will be faster than bytecode.

Also Java's slow. It is faster than it was earlier, but still is slow.

Case in point: Blockscape, which is minecraft-inspired game with MUCH higher level of detail and written in C# is much faster than Minecraft for most players (Minecraft's FPS:17-25 at max, Blockscape's: 60-120). Both are bytecode, and yet C# is faster there.

Anyway, I believe native version would be even faster than that. I think we should lobby Ouya makers to allow native apps (if it doesn't already).

SilverWarior
11-06-2013, 02:45 AM
Case in point: Blockscape, which is minecraft-inspired game with MUCH higher level of detail and written in C# is much faster than Minecraft for most players (Minecraft's FPS:17-25 at max, Blockscape's: 60-120). Both are bytecode, and yet C# is faster there.

You do know that Minecraft graphical engine is not the best one. Yes it has improved quite a bit in past years but still not enough. But there was a HD graphics mod for Minecraft which actually raised Minecraft FPS way abouve 60 and this on almost every computer with decent graphics card.

Another problem in Minecraft is that its core is quite porly written and can lead to lower FPS. You can clearly see this by the fact that you are getting way higher FPS and much lower CPU utilization when playing Multiplayr in comparison of playing Singleplayer game. And yes Minecraft performance went down again quite a bit in newer versions.
Now several mod packs like Technic Pack and such actually change part of default Minecraft code to make it much faster. Notch doesn't put much on optimization.

So using comparison of Minecraft with Blockscape to determine wheter JAVA is so much slower than native code is not best idea.
Best idea would be to write some benchmarking tests for true comparison.

WILL
11-06-2013, 04:58 AM
So using comparison of Minecraft with Blockscape to determine wheter JAVA is so much slower than native code is not best idea.
Best idea would be to write some benchmarking tests for true comparison.

Ok maybe not the best example, but considering how much the guy made off of that game sold at a reduced price with a free and/or demo version to try first, it does prove that Java isn't so unusable for games.


Anyway, I believe native version would be even faster than that. I think we should lobby Ouya makers to allow native apps (if it doesn't already).

I've seen some OUYA games so far and they play beautifully. So I seriously doubt that "the slowness of java" is such an issue as you guys make it out to be. It's not like the console is running a super low-end CPU either.

I believe that your slow issues will be in your graphics implementation and processing.

Dan
11-06-2013, 07:37 AM
the "slowness" of java comes from its native interface intermediate layer. basically speaking even if the java code itself is fairly fast (especially if it uses multiple threads) the bottle neck will be the interactions with the native libraries. for example every call from java to opengl library has to go through this native interface layer which slows down every individual call. it's not noticeable if you make 5-10 calls, however, 100-1000 calls already make a significant slowdown. now if you are making a game, how many opengl function calls are you going to make to do your rendering?

I agree that for OUYA this may not be a major problem especially if you are making a small game, because the hardware is fairly powerful and is in fact designed specifically for games. however, for the mobile devices and performance intensive games this is a very important issue. many android games are now made using the NDK instead of java.

Daikrys
11-06-2013, 03:02 PM
as Dan stated its not like Java is totally unusefull for games but its still has its performance issues if you build something bigger
indeed the ouya is far more powerful then any tablet so this might not be a problem for any well optimized game, but if you plan to publish your game on some other android plattforms native code would provide you a bigger, or better more satisfied, audience

im very interested in what we will see in the ouya store after the offical release, as for now i heared there are alot of quick ports from smaller android games. some gameloft style HD games would be awesome and a it would be way cooler if they come from indies and not from the big developers (like gameloft)

WILL
13-06-2013, 08:59 AM
Well most of my bias for Java had stemmed from (and rightfully so!) how slow and crappy the original sluggish and buggy JVM or JRTE stuff ran under Sun Microsystems' development. I'm not saying that it's worlds better than what Oracle has done with it (though it may be), but from what I can tell Google's implementation of Java RT into Android seems to perform much better than I remembered from the Java of years ago. Then again the first time I looked into Java seriously it was on a desktop computer that had a far less powerful CPU than the mobile devices I'm using now. That was a long time ago though. We have to take all this into account if we are to have a meaningful conversation about the end-result performance of the Java byte-code.

Graphics is a whole other bag of worms. You can make a poor-to-fair graphics implementation on a game for desktop and have it look alright on those systems, but take that to mobile and you see that it doesn't run so great. To get your games running great on that hardware you have to make sure that your games are optimized and tuned for that level of hardware. You can't be lazy on graphics or this is where your game will suffer most in performance. That also means using a good game library that does this for you or getting to really know your OpenGL!

Now I believe that our first conversion of Subject 33 ran graphics ALL in immediate mode. That's about as slow as it gets. Don't do this new people! It was our first rendition of the code and we are working at speeding things up. We are currently working on a much smarter way of dealing with our graphics by using VAs and/or VBOs through libGDX.

You can check out whats available on the OUYA right now on the website: http://www.ouya.tv/games/

There is quite a list so far (Esp. compared to Wii U's initial line-up!) and it's growing all the time. One that I'm excited about and hope to see more of is like the one below, Elliot Quest (http://ansimuz.com/elliotquest/).


http://vimeo.com/68026635

But then there are games like FF3, ChronoBlade and Flashout3D. I don't know if any use the NDK (http://developer.android.com/tools/sdk/ndk/index.html) or not.

WILL
24-06-2013, 06:53 AM
Wow it is so hard to get one of these consoles right now. I'm an original backer and I'm still waiting. It has been shipped and I have a tracking number, but the official launch is in 3 days (27 June) and pre-ordering on the site doesn't seem to help speed things up either. Apparently you can buy one at BestBuy in the US.

http://ouyaforum.com/showthread.php?3344-OUYA-at-Best-Buy-Not-Gamestop

I'm a little chuffed at the prospect that some people may have already bought one before the street date (due to some poor employees selling them early) as the promise was that ALL backers would get their OUYAs before the "official launch". Maybe I am mistaken and read that wrong somehow? ???

Oh well it's supposed to be coming so when it does I can say more about it. We're substituting the actual console with our Android devices and some gamepads. Thanks to libGDX it's pretty seamless to use the controllers API no matter the actual platform. Makes for less headaches this way.

This is what I'll be using to test gamepads on PC: http://www.bestbuy.ca/en-CA/product/mad-catz-pc-con-usb-controller-pc-mac/10051357.aspx

Eventually we'll port the game to all platforms, but our focus is all leaning towards a good OUYA version before we spread out onto mobiles and then the beastly PC markets.

Dan
05-07-2013, 06:21 AM
I have just got my OUYA! Will start experimenting right away=)

Daikrys
05-07-2013, 10:21 AM
hey congrats, hope you can tackle the ouya without any problems :)

pstudio
05-07-2013, 04:57 PM
I have just got my OUYA! Will start experimenting right away=)
Are you a kickstarter or have you just bought it now? I'm asking since I'm still waiting for mine.

WILL
11-07-2013, 07:05 PM
Congrats, I got both my OUYAs myself. My backer OUYA finally arrived on the 8th and the one I ordered through Amazon.com (retail version) arrived yesterday.

It's a pain to setup though. Make sure you hook it up to Ethernet instead of playing with wifi because it's a pain. Even with the firmware update.

I'm not a huge fan of the d-pad's directional detection or the responsiveness of the O-U-Y-A buttons. But so far the few games I tried seem to play ok.

Apparently you can use a PS3 or XBox controller as well if you want, but you'd need to design your game to take advantage of these.

Dan
12-07-2013, 06:32 AM
I am not a kickstarter backer but I did pre-order it. so far I had absolutely no problems, wifi works fine too. form a game developer's perspective I am a bit disappointed because while OUYA takes 30% of all the sales they still make you take care of all the sales taxes etc. so it's a lot more unpleasant than say publishing a game on google play. in fact it is already possible to install games from google play on OUYA.

Daikrys
12-07-2013, 12:14 PM
the Ouya team did really bad thing to backers cause they could not produce enough to ship to backers and their sales partners, so sale partners are more important to them and alot of backers still wait

also what amazon did here in germany is a pain, its only the pre-order price but its about 120€ (~156$). as we calculated with the shipping and duty it would be around 92€ at max

Dan
15-07-2013, 09:08 AM
after a few tries to connect to the ouya device I am giving up, at least for a while. it is very frustrating. first of all they do not supply a micro usb cable which is essential for development, so I had to use my android phone cable. then the thing just doesn't show up in the list of devices no matter what I try. I have read their very scarce documentation, followed the steps and no luck. all the other android devices that I have connect without any problems. if anyone else has any luck connecting to ouya let me know.

Cybermonkey
15-07-2013, 01:43 PM
I think the overall impression is not too good. I definitely won't buy one. Here's a review: http://www.digitalspy.co.uk/gaming/review/a495053/ouya-review-open-android-console-lacks-in-both-games-and-hardware.html
Also the German magazine c't titles "Squandered Trust".

wagenheimer
15-07-2013, 04:15 PM
Is Ouya rooted?

You can try this! http://slomin.blogspot.com.br/2012/06/easy-way-to-debug-your-android-apps.html

I can only debug on my Razor I Cell Phone using this mode (Wifi). (Eclipse does not seems to recognize it).

Dan
15-07-2013, 04:35 PM
I have also looked at the possibility of connecting over wifi but the problem is you need to set up both the development system and the device for that. and the difficult point is the device here. I am still back at it though. dont give up that easily=)

Dan
15-07-2013, 07:36 PM
alright, with enough persistence you can achieve anything! http://youtu.be/kaRclecNRiE
g2mp on OUYA, all that's left is to add the controller support.


http://www.youtube.com/watch?v=kaRclecNRiE

WILL
18-07-2013, 11:16 PM
Configuring my OUYA to compile to wasn't overly easy, I'll admit. But thankfully they did eventually provide instructions on the developers section of their website. (link (https://devs.ouya.tv/developers/docs/setup))


www.youtube.com/watch?v=454JFvFTxww

They aren't 100% glitch free. Some places they use the wrong names and your configuration might not match up exactly with what it details, but with a little trial and error you get there.

Well *I* managed to at least so what's your problem? ;) (Those that have worked with me in the past know I hate dev tool setups.)

Both Paul and I have Oxygene for Java building and deploying Subject 33 directly onto the console for testing now.

The console isn't 100% perfect either. Far from it. It's clunky and the controllers are probably the worst part of the system's problems. But it's not a total write-off and it will get better with time.

There is still lots of potential there and it is still very open in a way that no other console has been since before OUYA popped it's head out. Apple and Google have played a part in this too I think. You'll notice that Sony and Nintendo are noticing the trend and have been trying to approach their own developer strategies in a similar way.


Now if I had to make a list of the console's problems:

WiFi antenna sucks causing poor wireless networking connectivity.
Probably due to some oversight, you are required to enter credit card information before you are allowed to get access to the console upon start-up.
Controllers have really bad lag depending on where you sit from it and often "run away" on you, probably due to poor bluetooth reception as well.
UI is really vanilla and needs a lot of features, I mean a lot! Not a huge problem yet, but it will be once more games show up.


The really good points about it:

PS3, XBox 360 and even Wiimote controllers supported!
Great emulation machine. I was able to emulate all the NES, SNES, Genesis, GBA, PlayStation 1 and GBC games in my collection.
Followup from the first, the console acts like a USB drive when hooked up to your computer. You do NOT need to install the drivers for this to work.
Some really smart indie games are starting to arrive. Deep Dungeons of Doom, Hidden In Plain Sight, ittle Dew, Tower Fall, League of Evil, etc...
It's hackable so you can fix some of these issues by reading threads in the unofficial OUYA forums (link (http://ouyaforum.com/forumdisplay.php?13-OUYA-Hacks-amp-Mods))
Dude you can self publish your own games on a friggin' popular game console! :) And you aren't restricted by a teir system like on iTunes App Store.
...and most importantly you'll be able to get and play Subject 33 by Red Ant Games. ;)




http://www.youtube.com/watch?v=jSzLAs7ONZA


http://www.youtube.com/watch?v=AgV3Itdua3c


http://www.youtube.com/watch?v=sEbqAjfKE1Q


http://www.youtube.com/watch?v=nn8gAiSuWcQ


http://www.youtube.com/watch?v=oR4lDCbldb4

paul_nicholls
19-07-2013, 06:04 AM
Stalagflight on Ouya is rather neat...hard, but retro with atari sounds it seems :)

http://www.youtube.com/watch?v=zsCMlgavHSA

Darkhog
19-07-2013, 04:13 PM
Atari sounds? Not so much. More like NES/Famicom sounds since Atari games didn't have any music and (I think, may not be right) Atari didn't have noise generation support, only square wave.

Anyway, great game, I hope it'll find its way to PC.

WILL
20-07-2013, 01:40 AM
Depends on which Atari you are talking about no? I never had one of the old Atari machines, I was an Intellivision kid. But I know that there were several models so I'm sure that there had to be some sound improvements from system to system. Actually there is an Atari emulator for the OUYA in the store.

I saw that one and it looked neat, but to be honest, I just didn't have the heart to play anymore endless run games. Having both an iPhone and 2 iPads, I've had my fill of 'em. :)

I wonder if many people would be interested in a competition where they would get their game ported to the OUYA as the top prize? Such an idea would be an easier sell if I could get RemObjects in on the event too. See if they will do another promotion.

Making a game for the console is pretty exciting. I would love to see more people joinning us in putting their games on here. :)

I'm also anxious to see what Dan has up his sleeve.

WILL
20-07-2013, 02:00 AM
Actually you know what game might be great on the OUYA. Mini Racing Online (http://www.miniracingonline.com/) by kotai. :)

I wonder how well online games would work on the console. So far all the games I've been seeing are a lot of "offline" single and multi-player games.

WILL
21-07-2013, 10:35 AM
So far I know that Paul and myself are working on a project. Dan seems to be working on something--looks pretty cool too. Jason Farmer is also playing with it, a past community member and past PGD Annual competition 1st place winner if anyone recalls.

Anyone else working on a project or just tinkering?

Darkhog
21-07-2013, 12:25 PM
So how about making article about setting FPC so it can compile to Ouya and what libraries to use?

Dan
21-07-2013, 12:44 PM
it's the same as compiling for ARM android.

WILL
21-07-2013, 10:02 PM
So how about making article about setting FPC so it can compile to Ouya and what libraries to use?

Actually that's not a bad idea. You can set up Oxygene for Java to build and deploy right to the device so learning how to set that all up is just a matter of knowing the tools and figuring out how to install OUYA's "drivers" for your platform.

I'm sure that FPC doesn't have this ability on it's own, but maybe with the help of other software...?

Since we want to be a helpful community, WE could do a set (or short series?) of articles for Object Pascal OUYA development environment setup here on PGD. Would you be willing to write a short guide or list of steps for the Free Pascal portion Dan?

Dan
22-07-2013, 06:10 AM
arm linux fpc+lazarus package (for linux32 only): http://gen2gdk.com/files/ArmLinuxUtils.tar
download, read the instructions and run the installer after that you'll have fpc and lazarus ready for arm-linux compilation.
then you'll still need some tool to assemble the apk (eclipse or android studio) so I suggest that you do some reading
about the android NDK and how to link native code with the android application.

you can also find some decent instructions on how to create an arm-linux fpc compiler here: http://zengl.org/wiki/doku.php?id=compilation:android
another more recent article (looks like things are getting more simple): http://wiki.freepascal.org/Android

I would love to write an article about the whole process up to deploying an app to the device but I simply have no time for it.
if someone else decides to do that I would be happy to answer any questions here on the forum.

Ñuño Martínez
22-07-2013, 09:44 AM
Anyone else working on a project or just tinkering? My new job will include game development (yahoooooo...!!!!) and I was wondering which platforms we'll use. OUYA is on the list.

WILL
22-07-2013, 06:55 PM
Been working on the Game Controls screen for Subject 33. Thought you guys might think this is neat. :)

1182

Now I chickened out on the touchpad for now, nor did I use the Right analog stick for anything yet, but how would you guys use the controller layout to your advantage?

WILL
22-07-2013, 07:15 PM
My new job will include game development (yahoooooo...!!!!) and I was wondering which platforms we'll use. OUYA is on the list.

Sweet! May I ask what company you are working for now?


You know that's something that is sorely missing as well. A way to see all games from a specific company on the OUYA. I'm really hoping for some big UI changes and feature additions over the next couple of months.

It has also been mentioned that adding the ability to give out keys, like Steam, for those you want to have a free copy of your game or access to a beta is something that they plan to put into the OUYA as well.

Darkhog
22-07-2013, 09:30 PM
WILL, I know you are experienced dev and all, but that control scheme is AWFUL!

IMO, it should be as follow:

L2/R2 - prev/next weapon
L1 - shoot
R1 - Mine

Y - Throw clip
O - Toggle dive

Reasons:

Shooting is in games pretty common action, as dropping mines is. Buttons used for this should be available for players without getting away to much. Also using index fingers for changing weapons and shooting/placing mines is pretty straightforward and easy to get.

Putting away clips isn't good thing in most cases, so it should be assigned to button where player is actually 100% sure that he want to do this, so he won't throw clip accidentally, therefore Y.

Dive is not as rare action as putting away clip is, but isn't as common as shooting/placing mines, therefore it should be assigned to accessible key like O.

For right stick I'd propose aiming (left would be for movement).

WILL
23-07-2013, 08:01 AM
Ah... well I guess it might help to know what perspective the game will be seen and played.

1184

Hopefully that sheds some light on why I have 4 fire buttons. The game is designed so that movement and firing in all 4 directions is possible in parallel.

I had thought of having mines a separate drop-able without the need to select it as you would any of the other weapons, but was unsure to what button to map it. Or if I just wanted to continue using them as the weapon you have to select to drop them. (These are of course on a very limited supply at the beginning of the game.)

As for the L2 and R2 triggers, these are meant to be toggles. They will allow you to throw ammo clips of your currently selected weapon and activate a combat dive on the run.

L2 + any O, U, Y or A button will allow you to throw your buddy in co-op mode a clip of ammo. Even on the run!

R2 + any cardinal direction on the LS or D-Pad will make the player speed boost into a combat dive. You can do this while already running or from stand-still.

I wasn't planning on getting into the gameplay mechanics so much here, but since you made the effort to help I felt obliged to do so. Your suggestions do make sense for a FPS of course and given the parallels to Subject 33's gameplay some elements are similar so some things will need to be used in a similar way.

Our plans are to create a "good" default and later allow players to customize the controls to their own personal tastes to some degree. LS, D-Pad O, U, Y and A are pretty much set in stone unless we have some kind of great idea that we can pull off. ie. Using RS to fire. But we've not tested that out yet so I'm not sure if it's even going to work mechanically.

What do you think about the shoulder buttons now? Would you prefer the weapon selection and the Dive/Throw toggles flipped? Or do you think that this may be something that is more "taster's choice"? :)

Darkhog
23-07-2013, 04:26 PM
Actually when I thought about your game, I've imagined overhead view. And still I sustain my suggestion, with some additions.

1. Right stick can be used for aim (rstick to the left + shoot would shoot to the left, etc.)
2. Y+Dpad to shoot clip in specific direction. No issue there, you shouldn't break hands here.

WILL
23-07-2013, 07:34 PM
I think your ideas would remove the simplicity of the core player control mechanics. You are now holding or pressing 3 things at the same time to do 2 of the most common tasks in the game. And your scheme doesn't feel natural at all. I wouldn't be able to throw a clip on the move either at least in a different direction from that which I'm already moving.

When designing controls you have to look at it from the player's perspective. What does the player have to do the most and how do you make that simple for him to do over and over as needed in the game. And by no means do you want to try to cram features into your control scheme that force players just to conform to what other games have done.

Darkhog
23-07-2013, 10:18 PM
Humour me and implement it, then try it. If it won't work, you can always change it back.

WILL
24-07-2013, 03:27 AM
Humour me and implement it, then try it. If it won't work, you can always change it back.

Would that someone's whim didn't cost time and/or money. Both I don't have in abundance. Sorry, but I'm not convinced and I doubt anyone else who is putting serious resources into a commercial product would follow you on your idea with very little validation other than you say so.

Now if you would like to make a proof of concept to validate your ideas, by all means, but don't expect people to make sweeping changes to their projects just because you want them to. That's just silly.

I'm all for good debate every now and then, but I have debunked your original suggestions and you have not come up with a decent counter argument. I guess "just do it" seemed innocent enough to you, but it's making you come off a little pushy. I disagree with you. If you are unable or unwilling to validate your ideas with a counter then lets just agree to disagree. It won't hurt my feelings.

Ñuño Martínez
31-07-2013, 10:09 AM
Sweet! May I ask what company you are working for now?
Not sure if we can call it "company". I'll work for a new web-service company (they'll start to work on September) and for the SerieB studio (comics (http://www.chicadeserieb.info/) and design). They'll give me work (web programming, in most cases) and will allow me to work in my own projects including games with graphics done by SerieB studio (and also games based in some SerieB comics).

WILL
01-08-2013, 06:00 AM
Wow sounds like it'll be a fun job. And flexible too if they are willing to support you on your own projects. How much control would they give you of these projects that you get to do yourself?

And hooking into the topic, what plans for OUYA are there? Games? Comic Book frontend on OUYA? I could see that being popular.

Ñuño Martínez
04-08-2013, 01:51 PM
Wow sounds like it'll be a fun job. And flexible too if they are willing to support you on your own projects. How much control would they give you of these projects that you get to do yourself?

And hooking into the topic, what plans for OUYA are there? Games? Comic Book frontend on OUYA? I could see that being popular. We didn't think about this...

Anyway, we're starting so initially it will be something like a test. But they'll give me a lot of control, as they don't have any idea about how to make a game.

WILL
05-08-2013, 07:07 AM
Cool. Can anyone smell collaborations? :)

Hey how crazy an idea would an OUYA game dev jam/competition be? Or maybe just a porting over to a non-Android/JVM game over to it as a prize?