Genefer - Family of GFN Testing Programs
| Genefer.zip | - Version 2.2.1, November 26, 2011 | - 32 bit Win program using 64-bit X87 assembly FFT |
| Genefer.exe | - Version 2.2.1, November 26, 2011 | |
| Genefer80.zip | - Version 2.2.1, November 26, 2011 | - 32 bit Win program using 80-bit X87 assembly FFT |
| Genefer80.exe | - Version 2.2.1, November 26, 2011 | |
| GenefX64.zip | - Version 2.2.1, November 26, 2011 | - 64 bit Win program using 64-bit SSE2/SSE3 assembly FFT |
| GenefX64.exe | - Version 2.2.1, November 26, 2011 |
Update Log
Version 2.2.1
- Added updated checkpoint file format to match the other Version 2.2.1
genefer family implementations
- Added a Control-C event handler
Genefer Programs
Genefer is a family of programs, based on algorithms by Yves Gallot, for finding large probable GFN primes. His original download page is http://galloty.chez.com/primes/pgm/
- Genefer uses a 64-bit floating point FFT which is fast but can only handle a limited range of lower b values.
- Genefer80 uses an 80-bit floating point FFT and is considerably slower, but can handle a much wider range of b values.
- GenefX64 is a complete rewrite using SSE2/SSE3 and x64-64. It is 25% faster than
any other known CPU based implementation (with a similar lower limitation in b values).
With an appropriate graphics card, GeneferCUDA is by far the fastest
implementation.
David Underbakke developed GenefX64 for the 64-bit Windows environment and targets the hardware improvements in the AMD Family 10h Processors (Phemon and newest Athlon X2) and Intel Core 2 Processors (or above). It is the fastest program for testing probable GFN primes. The minimum processor requirement is SSE2/SSE3 and x86-64 with a Windows 64 bit OS. GenefX64 utilizes algorithms similar to Yves Gallot's genefer.c program which can be found at http://pagesperso-orange.fr/yves.gallot/primes/download.html
Genefer and Genefer80 were originally developed by Yves Gallot. For the 64-bit FFT version, careful usage of the X87 FPU in 80-bit mode for internal calculations resulted in a 25% wider effective b range (versus 64-bit FPU SSE2) with very little performance impact.
They were further optimized by David Underbakke and modified to include new features such as a checkpoint file, a residual value for composite numbers, and progress reporting.
Mark Rodenkirch assisted with many of the new features such as the checkpoint file, quick calculate mode, progress reporting, and residual values.
Linix, Mac Intel, and PPC releases of Genefer and GenefX64 are provided by Iain Bethune.