Comments for Miles Macklin

Comment on Atomic float+ by codah

codah — Sat, 22 Aug 2009 09:23:39 +0000

Good work on the new Batman game, graphics look amazing and reviews seem really positive.

Hope everything goes well for you at Lucas Arts. You sure have come a long way dude

Comment on GPU Metaballs by mmack

mmack — Mon, 11 May 2009 20:55:19 +0000

OK the technique is really really simple. Here are the basic steps:

1. Set up a particle texture like the one in the demo, so that it’s alpha channel represents density falling off from the center (any shape will do, experiment with different falloffs)

2. Render a whole lot of these to a render target with additive blending, this create’s a ‘density texture’.

3. Enable alpha-testing, and blit the density texture to the back buffer. The alpha threshold will affect the ‘viscosity’ of the fluid.

Hope that helps
Miles

Comment on GPU Metaballs by kiplantt

kiplantt — Mon, 11 May 2009 20:10:40 +0000

I would also be interested in seeing the code or at least some explanation about the process used in this demo.

Thanks.

Comment on GPU Metaballs by gregory

gregory — Sun, 08 Feb 2009 14:13:53 +0000

indeed i would fancy looking at the code to understand how it’s done

cool demo
cheers

Comment on Branch free Clamp() by mmack

mmack — Tue, 13 Jan 2009 22:34:38 +0000

Thanks, I should have realised that looking at the disassembly.

I did some more timing today and am less convinced it’s always faster than a standard implementation, although as you suggested converting some of the parameters to doubles did seem to help.

I’ll post some more detailed stats when I get a chance.

Comment on Branch free Clamp() by Jaymin K.

Jaymin K. — Tue, 13 Jan 2009 08:46:25 +0000

I haven’t looked at the generated assembly, but if you are calling this in a loop it may also help to do a bit of “pipelining,”
i.e. do the conversions/casts for iteration n+1 before you do the clamp for iteration n.

Anyway, its one thing to look into and you can tell really easily if it helps ( or makes it slower! )

Comment on Branch free Clamp() by Jaymin K.

Jaymin K. — Tue, 13 Jan 2009 08:35:26 +0000

on the PPU there is no real fsels instruction. Its just fsel with some casts to float done for you for convenience. This MAY ( and I say may because I haven’t tested it out yet ) be faster if you make t a double and change your first __fsels to an __fsel. That way you can avoid one double to float cast and another float to double cast.

Comment on Sound by Si

Si — Fri, 15 Aug 2008 11:53:59 +0000

You’re gay. Love Si. x