Comments on: PolyVox on Android

By: David Williams

David Williams — Sun, 18 Nov 2012 22:33:53 +0000

I don’t have access to a Blackberry PlayBook so I haven’t tested there. I answered you other questions on the gameplay3D forum as it seemed better to keep the conversation in one place: http://www.gameplay3d.org/forums/viewtopic.php?f=5&t=6&p=736

By: Ihor

Ihor — Sun, 18 Nov 2012 20:50:07 +0000

Hello, any updates related to build PolyVox under mobile platforms? Can you provide instructions how to include PolyVox
library to the project based on Gameplay SDK? I like PolyVox and I planning to use it as voxel engine for Blackberry Playbook development. Did you had a chance to launch PolyVox under Playbook OS?

By: Where's the next PolyVox release? - Volumes Of Fun

Where's the next PolyVox release? - Volumes Of Fun — Fri, 19 Oct 2012 19:59:10 +0000

[…] filled the first 6-8 months and is still in the background), but we’ve also been working on gameplay integration and having a busy couple of months at work. Still, there has been a steady stream of […]

By: PolyVox Marching Cubes

PolyVox Marching Cubes — Sat, 13 Oct 2012 07:08:40 +0000

[…] smooth terrain into RIMs gameplay3d engine. This follows on from our previous work on cubic terrain. The smooth terrain is no more difficult than cubic terrain, and the main challenge has been […]

By: David Williams

David Williams — Sun, 09 Sep 2012 20:00:52 +0000

It didn’t compile straight out of the box on Android but it wasn’t too hard to get it working. I had to delete some code that was using std::function as the Android compiler didn’t recognise that, and exceptions were also a problem. There may well be compiler flags I was missing to add support for these features though.

At any rate we’ll keep these issues in mind, and I think we’ll try to make PolyVox more mobile compatible in the future.

By: Robert Stehwien

Robert Stehwien — Sat, 08 Sep 2012 21:30:47 +0000

Got me looking at the gameplay engine now. I’ve been following polyvox for a little while but integrating it into Unity is a pain (not to mention requiring pro). I started working on a voxel engine for Unity and said to a friend “I’d rather use C++ and Lua so I can use some libraries (like PolyVox)… but the problem is getting them all to compile on the different platforms and mobile.” Along comes your article and I’m giving gameplay a try. So far I like it.

Did you come across any snags using it with PolyVox? It would be very interesting to see you port Voxeliens to gameplay.

By: David Williams

David Williams — Tue, 14 Aug 2012 08:54:34 +0000

I agree that the gameplay matches mobile a lot better – it’s the kind of game you’ll play for 10 minutes while sitting on the bus rather than playing on your PC when you get home from work. I think the controls are solvable, for example your right thumb could control the rotation of the world while your left thumb controls the tank… then just poke it with your nose to shoot.

I’m mostly concerned about performance, and in particular drawing the large numbers of enemies. Not sure if mobile phones support much in the way of instancing. I need to do a lot of research here.

We’ve been on Voxeliens for ages and are both looking forward to doing something new… but at the same time we should make the most of all opportunities and we already have the art assets and design for this. At least we’re playing with new tech, and it provides useful test of our next engine. We’ll just have to see where it goes 🙂

By: Freakazo

Freakazo — Tue, 14 Aug 2012 02:58:17 +0000

The gameplay certainly does, but getting the controls right will be incredibly hard.

By: glue

glue — Tue, 14 Aug 2012 01:52:55 +0000

Very cool. I think it would make a lot of sense to try and get Voxeliens on the mobile platforms. Frankly I think it fits better in that market space than the PC market. I could be wrong of course, but that’s my opinion.

By: David Williams

David Williams — Mon, 13 Aug 2012 09:53:23 +0000

Unfortunatly there’s no simple answer to this question. In the video the frame rate is varying between about 40-50 FPS, and this variation seems to depend on which angle the terrain is viewed from (even though the same number of cubes are being drawn). The performance basically depends on where the bottleneck in the pipeline is (vertex shader, fragment shader, interpolators, etc) and this will depend on the application. I expect we’ll do more experiment in the future though.