In a nutshell, what's not working actually ?

[Roland_Yonaba]

Hi,

I recently bought a new Androphone, HTC Wildfire S, running under Android (2.3.7), actually. I spent some time sneaking on the net, trying to understand how android works in general, cause I wasn't that familiar with the OS itself.
Then I started googling for ways to start developing on Android...I discovered Android SDK...But I left it quickly because I hated the way the SDK itself have to be installed before starting to work (many things have to downloaded online, and I didn't want to work with Java...).
I also discovered Titanium AppAccelator but it relies on the Android SDK, and features JavaScript style programming...Then I left it too.
Later, I found Corona SDK...That was sweet, cause Corona features Lua...I thought I finally reached my heaven, but I was disappointed (again) when I found that Apps produced with Corona couldn't work on my phone (hardware related, corona's apps works with Arm7 processors, mine is Armv6)...

I looked around again, and I dropped here...I tested the application itself, I mean, Love-Android, it worked fine on my phone. Plus, i'm already familiar with love API, the way it works, callbacks, the whole stuff...Then i am definitely interested in this...Then before starting making games, what do I need to know about LoveAndroid ? What's not working so far ? What are the plans of the support team for the next days ? Is Love Android actually ready to be use in large projects...? Are there some limitations ?

Thanks...And congratulations to all those who contributed to porting Love to Android! Keep up the good work!

[T-Bone]

The way love-native-android works right now, you don't need the Android SDK to start working. However, if you want to distribute a stand-alone application (an individual app that simply runs your game) you will both need and want the Android SDK. Love-native-android has no current support for native applications. This functionality is planned, but seems to be kind of low priority for now.

I am personally developing a game specifically for Android, and I intend to see it released as a stand-alone app as soon as possible. So you can trust that even if the developers of love-native-android do not give it that functionality, I will do everything I can to tweak the source code to get it to run .love files from within the app, instead of the way it does things now, which is running .love files from the SD-card partition. It should really just be a matter of chaning a few lines in the code. I hope.

As for how far love-native-android has come, I'd say it's quite a bit away from fully functional. Graphics works fine mostly, with some limitations (like no framebuffers, but spritebatches totally works, and some specific stuff don't do anything like love.graphics.setLineWidth) and sound causes crashes. Also, it crashes if you close your phone while it's running. Also, no real extra features outside of ordinary LÖVE are added yet, like multitouch. But it's coming.

I'd say love-native-android is mature enough to start developing stuff for it, but it's not mature enough to actually release anything.

Also, I am not a developer of love-native-android, I just REALLY love this project (and desperately need it for my game) which is why I'm so obsessed about it.

[Roland_Yonaba]

Thanks for all that informations.
Well, you're totally right. It appears to me that tool is not mature enough for game release.
IMO, Lua is perfect for game development...
And because I intend making games (even little ones), seeing a project like LoveAndroid makes me enthusiast...

Well, can i know a bit more about your personal project ? In terms of coding, how far have you been able to go ?

[off-topic] : Maybe you'll have to update your signature...

[T-Bone]

Thanks for all that informations.
Well, you're totally right. It appears to me that tool is not mature enough for game release.
IMO, Lua is perfect for game development...
And because I intend making games (even little ones), seeing a project like LoveAndroid makes me enthusiast...

Well, can i know a bit more about your personal project ? In terms of coding, how far have you been able to go ?

[off-topic] : Maybe you'll have to update your signature...

I've gotten quite far, the game engine totally works and runs on Android as well. I will implement sound and multitouch when it gets implemented in love-native-android. Other than that it's just a matter of designing levels (I've already made about 14 so far) and getting my designer to draw more graphics From a performance standpoint, my game runs at about 40-50 FPS on my two year old HTC Desire. I consider that pretty impressive.

[Roland_Yonaba]

Quite impressive...
Is there any downloadable version right now...?

EDIT: MILF ?

[T-Bone]

Quite impressive...
Is there any downloadable version right now...?

EDIT: MILF ?

Haha, no, MILF was just a fun little project.

There is no downloadable version. I intend to keep this closed source and sell it once it's done, so I'm not posting it here. I believe the concept is good enough to be sellable, so I want to keep it to myself for business reasons

Also, a whole other thing. You should note the difference between the two, almost entirely unrelated, projects love-android and love-native-android. Love-android uses LuaJ and is therefore quite slow. It had several cool things implemented like accelerometer support and compass support. However, it has some pretty serious incompatabilites with normal LÖVE, like love.draw running at the same time as love.update, but in different threads (which caused wierd crashes unless you were careful), and it is now abandoned in favor of love-native-android which uses real Lua/LÖVE built with C on Android. This gives much better performance, but it's relatively new and unfinished.

[Moe]

I didn't want to work with Java...

Android Java is much nicer than Sun/Oracle Java. The API is much nicer to use, it is actually the first time that I do not get mad when working with Java. Unfortunately, you have to do a lot of message passing because of Android's threading (you shouldn't do this in thread a and that not in thread b), and you still have this verbose writing I cannot get comfortable with.

I looked around again, and I dropped here...I tested the application itself, I mean, Love-Android, it worked fine on my phone. Plus, i'm already familiar with love API, the way it works, callbacks, the whole stuff...Then i am definitely interested in this...Then before starting making games, what do I need to know about LoveAndroid ? What's not working so far ? What are the plans of the support team for the next days ? Is Love Android actually ready to be use in large projects...? Are there some limitations ?

At the moment, we support version 0.7.2 only. I just started to look into 0.8.0 and how much effort it would be to switch or to support both. Most problems are currently caused by unclean shut down of Love. We also need to add a state saving mechanism if the player leaves android-love-native, because Android does not give any guarantees about the memory state when the user returns. We had some fixes for sound and I it seemed to work for me since then. Well, volume control does not work, but this is known and I would have been surprised if that would be different
Feel free to develop, use the native version only. If you find bugs (and I am pretty sure you will), please report them with as much information as you can provide. I am not happy with bug reports that say "hangs up" without any details.
Oh, there is one more thing I am working on: Get hagish to contribute again

Thanks...And congratulations to all those who contributed to porting Love to Android! Keep up the good work!

Thank you.

[T-Bone]

I've realised that all I need for my game that currently isn't implemented is

1. Fix the freeze when focus is lost (being worked on)
2. Multitouch
3. love.event.quit/love.event.push('q') (some way to quit)

I think that's all. Graphics and audio work just fine, performance is pretty good. I am impressed and happy Hopefully, these things will be fixed by the time my game is finished

[frankmiller]

I agree with you.
Multitouch has to be fixed yet

[Moe]

3. love.event.quit/love.event.push('q') (some way to quit)

Just wrote the code for it. Pray for bad weather tomorrow so that I will test and commit it. If not ... nah, you don't want me to do "if it compiles, ship it".