Also, a screen saver does not *have* to consume significant overhead-CPU 
cycles. For example the XP "3 D Pipes" screensaver uses 45% of my Pentium-M 
CPU, whereas "Beziers" uses an imeasurably small amount of CPU. I'm sure it 
would be possible to design a screen saver that delivered some useful 
information about the LL-test/FFT etc, but without consuming a significant 
resources on top of the prime test itself.
 
It appears to me that a screen saver version of P95 would offer a simple 
benefit of providing a zero-performance-impact version of the program for 
people (like me!) concerned about the effects of using CPU saturation when I'm 
also trying to use the PC.
 
Peter Moreton
 

________________________________

From: [EMAIL PROTECTED] on behalf of Brian Beesley
Sent: Mon 30/10/2006 08:16
To: The Great Internet Mersenne Prime Search list
Subject: Re: [Prime] Possible method to get more 
prime-searchingparticipantsonline



On Monday 30 October 2006 04:48, Tim Sloane wrote:
> On 10/29/06, david eddy <[EMAIL PROTECTED]> wrote:
> >
> > We don't need "eye candy" - at least not time consuming stuff.

I agree, _we_ don't _need_.
>
> I recall this being discussed ad nauseam on the Mersenne Forums a long
> time ago.  A screensaver, which would always take up some cycles, goes
> against the spirit of the project. 

Who says? Is this just your personal opinion, or is it written down somewhere?

> This project is to make the most
> efficient use of spare CPU cycles,

Yes ...

> and using a screensaver to run the
> program goes against that idea.

I don't follow.

So long as the screensaver is not compulsory, then, if having it available
induces extra people to donate their spare cycles, surely the converse is the
case.

Even if the screensaver were to waste 50% of the donated CPU cycles, then we'd
still be getting the other 50% which are at the moment going to waste.

The real question is, whether the effort associated with generating
screen-saver versions (or front ends) is worth the contributions from the
extra users we'd bring in. I don't know how to answer that question, but I
think it is fair to point out that the really successful distributed
computing projects (in terms of numbers of contributors) have all been
screensaver based, or have had screensaver options.

Regards
Brian Beesley
_______________________________________________
Prime mailing list
[email protected]
http://hogranch.com/mailman/listinfo/prime


_______________________________________________
Prime mailing list
[email protected]
http://hogranch.com/mailman/listinfo/prime

Reply via email to