LÖVE

Hey guys,
I was wondering if we should continue the port on version 0.7.2 or if we should switch to the brand new flavour of 0.8.0? 0.7.2 feels like riding a dead horse...
Cheers

0.8.0 sounds cooler

I am not a love-native-android developer, so I do not think I should be given a vote in this matter. However, I still want to share my opinion:

Stay with 0.7.2, unless a move to 0.8.0 solves some problems love-native-android currently has. I believe the top priority should be to get a working version out as soon as possible. This would allow developers like me to get their games out there, which is the whole point.

And, as an added benefit, if you get it out there as soon as possible, games will LÖVE will start to appear in public markets like Google Play, which will attract more attention to the LÖVE project as a whole and specifically love-native-android in this case.

Also, once 0.7.2 love-native-android is finished, there will be plenty of time to port 0.8.0 afterwards. Nobody will blame you for finishing what you started. I am just afraid that moving to 0.8.0 now will only introduce a lot of new bugs, while not giving Android developers any important benefits.

EDIT: I've been thinking a bit about this during this morning, and I have kind of changed my mind a bit. If working with 0.7.2 really feels like "riding a dead horse", you should probably make the jump to 0.8.0. It is always more fun to work on a project when you know that the end result is going to be a little bit more awesome. So I change my opinion to "make the jump to 0.8.0 if it won't slow you down".

The main difference will be the OpenGL part, the 0.7.2 port is based on ES 1.1, but 0.8.0 will use ES 2.0. This is because of shaders which work with ES 2.0 onwards. I have the bad feeling that I would have to do a lot of adjustments two times since both differ a lot, e.g. font rendering...

Moe wrote:The main difference will be the OpenGL part, the 0.7.2 port is based on ES 1.1, but 0.8.0 will use ES 2.0. This is because of shaders which work with ES 2.0 onwards. I have the bad feeling that I would have to do a lot of adjustments two times since both differ a lot, e.g. font rendering...

Don't do it if you think it will slow development down significantly.

You can tell that all I want is a working version as soon as possible, can't you?

actually a switch to OpenGL ES 2.0 would speed up development, because I haven't done a lot of stoneage like OpenGL like in ES 1.1 for years. I do a lot of 2.0 (or equivalent desktop versions), so I can much easier resolve issues. 1.1 feels always clumsy...

PS: Two words to your signature: Challenge accepted.

Moe wrote:actually a switch to OpenGL ES 2.0 would speed up development, because I haven't done a lot of stoneage like OpenGL like in ES 1.1 for years. I do a lot of 2.0 (or equivalent desktop versions), so I can much easier resolve issues. 1.1 feels always clumsy...

PS: Two words to your signature: Challenge accepted.

Well then, go for it!

Just downloaded the new source code. Next step: start the "you will hate me"-solution, so that we can easily switch between 0.8.0 and 0.7.2

Moe wrote:Just downloaded the new source code. Next step: start the "you will hate me"-solution, so that we can easily switch between 0.8.0 and 0.7.2

Wait, you're going to have both?! Won't that be too much to do? And besides, 0.8.0 is very backwards compatible with 0.7.2. Pretty much the only thing that isn't backwards compatible is framebuffers (that only has their name changed), and they don't work anyway (which is fine for me by the way).

Just wondering, what sort of benefit would there be to keeping 0.7.2 support? Because I can't think of any.

Nearly always when a new Love version was released, all of our games were broken. But if this is not the case with 0.8.0, we can drop support for 0.7.2. I simply didn't expect this.