There's apparently a SETI app (SoG?) written by a third party that runs SETI tasks much much faster provided you have Linux and a newish Nvidia card. Any such thing been done for Einstein? Or optimizations on command line?
If this page takes an hour to load, reduce posts per page to 20 in your settings, then the tinpot 486 Einstein uses can handle it.
Copyright © 2024 Einstein@Home. All rights reserved.
No.
)
No.
Cheers,
Gary.
There's nothing to stop you -
)
There's nothing to stop you - the source code is available under the GPL (https://einsteinathome.org/application-source-code-and-license), and the same page contains guidance for deployment under Anonymous Platform.
But you will have to work very hard to better the optimisations already in the project's own applications, courtesy of Akos Fekete: his work was so spectacular that they hired him and made him a staff consultant (https://einsteinathome.org/science/contributors). Search the forum for that name.
Richard Haselgrove
)
There is one thing stopping me, I'm a rubbish programmer.
So how come SETI hasn't been optimised as well as Einstein? Or is it just that the type of instructions that Einstein WUs do don't need optimizing for specific hardware?
If this page takes an hour to load, reduce posts per page to 20 in your settings, then the tinpot 486 Einstein uses can handle it.
Peter Hucker wrote:So how
)
Largely because of Senator William Proxmire and his followers. Their USA Federal budget amendments in 1982 and 1994 effectively blocked any public funding of SETI - long before the @Home branch was set up, of course, but the political animosity lingers. It wasn't helped when Yuri Milner's Breakthrough Listen initiative initially concentrated on data capture, while poaching a member of staff (never replaced), and initially not contributing to the back-end costs of SETI@Home processing (that's been corrected more recently).
SETI@Home runs on a shoestring, with hand-me-down kit (some from commercial sponsors) and volunteer donations. Unlike Einstein, SETI@Home has never developed a GPU application in-house from scratch: the original SETI CUDA app was developed and donated by NVidia as a loss-leader for their Christmas sales in 2008. Every GPU advance since then has been contributed by volunteers.
But I think we'd better take this conversation back to where it started.
Richard Haselgrove
)
That's a shame and if I had spare cash I'd certainly donate to SETI.
If this page takes an hour to load, reduce posts per page to 20 in your settings, then the tinpot 486 Einstein uses can handle it.