FFFF - Fast Floating Fractal Fun / News: Recent posts

FFFF 3.2.3, faster, adds IRIX, Linux and OS X i386 support !

- Now also runs on Linux, IRIX, Mac OS X (universal binary: PowerPC+i386).
- Supports GPU Fragment Programs on Mac OS X and Linux.
- Asm optimized MIPS4 code path (R8000 or better).
- Faster SSE/SSE2 on Intel (Peter Kankowski's MOV-ADD method).
- Lots of bugs squashed.

Posted by Daniele Paccaloni 2006-03-18

FFFF 3.2.2 OsX+AltiVec release is out !

New version adds support for OsX using FPU and QuadPixel AltiVec calculation for cross benchmarking ! Peace and love to all PC and Mac users ;)

Posted by Daniele Paccaloni 2004-12-22

1 GigaIters/sec barrier broken ! v3.2.1 is out.

40% speed increase on ATI R3xx Fragment Programs, with up to 30 iters per pixel.
Enjoy.
.Daniele

Posted by Daniele Paccaloni 2003-11-18

Soon v3.2.1, > 1 GigaIters/sec on the R300 (thanx B.Lipchak)

Thanks to Benjamin Lipchak swizzle optimizations, you'll get 31 iters per pixel and more than 1 GigaIters per second on an ATI R3xx !!!
[Latest ATI drivers required].
Stay tuned :)

Posted by Daniele Paccaloni 2003-11-16

New v3.2.0 using FragmentProgram reaches 700 MegaIters/sec !

Reach 700 MITS (MegaIters/sec) on the ATI R300 GPU !
Let me know if/how it runs on the NV30.

.Daniele

Posted by Daniele Paccaloni 2003-03-02

ATI improves Radeon 9x00 VP performance

Also profiling FFFF, ATI improved the performance of Radeon drivers on complex Vertex Programs.
The new drivers will be released very soon.

Posted by Daniele Paccaloni 2003-02-04

FFFF 3.1.2 X-Mas release is here with 30% speedup and SSE2

- Optimized SSE calculation code [GB].
- Added SSE2 support (DualPixel, double precision) [DP].
- Calculation mode keys are changed (see help text), press '0' for GPU calc [DP].
- Minor fixes [DP].

Posted by Daniele Paccaloni 2002-12-23

Soon Gerard's 30% speed optim on SSE code for new CPUs

Gerard did optimize the SSE calculation routine using AMD's Pipeline analyzer.
The speed improvements make FFFF ~30% faster on P4 and newer Athlons.
My tests on a P4@2.4GHz results in 375 MegaIters/sec (about 6.5 clock cycles per iteration). On my P3, the speed improvement is ~5%.

Posted by Daniele Paccaloni 2002-10-07

New v3.1.1 - FIXED benchmarking and minor issues

Sorry for the stupid bug, i am developing on an Athlon, but there are no excuses :)
Please download this fixed version which includes also other minor fixes.

Posted by Daniele Paccaloni 2002-10-05

3.1.1 soon available, with fix for benchmark on Intel CPUs

The benchmark does crash trying to execute 3DNow! code on Intel CPUs.
Will be fixed asap

Posted by Daniele Paccaloni 2002-10-05

Realtime Orbit plot in v3.1.0

Version 3.1.0 supports realtime orbit draw.
Just move the mouse on the image while pressing the "O" key (the O letter, not the zero !).
I forgot to mention it in the doc.. sorry :)

Posted by Daniele Paccaloni 2002-10-02

NEW RELASE 3.1.0

New in this release:

- Added AMD 3DNow! dual pixel calculation [GB].
- Added Radeon GPU pixel calculation using GL_VERTEX_SHADER_EXT (see notes). Maybe it works on other GPUs, tell me ! [DP].
- Added experimental GPU benchmarking (does not work on my Radeon 9000..) [DP].
- Slightly improved speed of SSE2 quad pixel calculation by constant data alignment on 64 bytes boundaries [DP].
- Slightly improved speed FPU code.
- Code restructured to resemble an object oriented program (quite not yet :).
- Fixed minor bugs [DP, thanks for your bug reports].

Posted by Daniele Paccaloni 2002-10-02

Radeon GPU calculation support in next release

The Radeon GPU Mandelbrot code (Vertex Shaders) is ready.
It will be present in the upcoming version 3.1.0.

Posted by Daniele Paccaloni 2002-10-01

3DNow support soon ! (Thanks to Gerard Basler)

Gerard coded the 3DNow! routine that will be integrated and released asap. :)

Posted by Daniele Paccaloni 2002-09-27