Fastest seti@murcers July 31-Aug 7

[paulcs]

Being able to upload WU's without losing them would be nice as well.

Just lost two. I believe that's eight in a weeks time.

Paul
paulcs@flashcom.net

[Rattledagger]

Ranking by most cpu-hours doesn't work, because win9x doesn't use cpu-hours but run-time-hours. The gui-version doesn't use cpu-hours either, not even under NT.

If you wants to get many hours, run the graphic version alongside the cmd-line. If you're using win9x, you can start many instances of the cmd-line and get many hours.

[Pace]

I suppose - but can we not rely on people to cheat? We could always spot the guy on a P3 with 139(!) hours unit average.

Paul.

[ScooterX]

Thanks for noticing Pace
I am trying to get more machines running to secure myself a position in the top 100
As I mentioned in another thread, Diablo2 is killing my WU production.

[Ees]

Guys,

Just to see if we have a business case here...

This is what the "ranking by effort" would look like (for the same week):

<TABLE BORDER=1><TR><TH>name</TH><TH>hours</TH><TH>weeks</TH></TR><TR><TD>breezer</TD><TD>7538.76</TD><TD>44.87</TD></TR><TR><TD>GnrlData</TD><TD>2279.16</TD><TD>13.57</TD></TR><TR><TD>He cubus</TD><TD>1490.22</TD><TD>8.87</TD></TR><TR><TD>Alastair</TD><TD>1440.52</TD><TD>8.57</TD></TR><TR><TD>Charles</TD><TD>1227.24</TD><TD>7.30</TD></TR><TR><TD>Bob & Sons</TD><TD>1227.24</TD><TD>7.30</TD></TR><TR><TD>Ees</TD><TD>1051.92</TD><TD>6.26</TD></TR><TR><TD>Brian</TD><TD>1051.92</TD><TD>6.26</TD></TR><TR><TD>Radioman</TD><TD>1015.27</T D><TD>6.04</TD></TR><TR><TD>JSB</TD><TD>964.26</TD><TD>5.74</TD></TR><TR><TD>Alpha Centauri</TD><TD>964.26</TD><TD>5.74</TD></TR><TR><TD>paulcs</TD><TD>964.26</TD><TD>5.74</TD></TR><TR><TD>box</TD><TD>964.26</TD><TD>5.74</TD></TR><TR><TD>JJc</TD><TD>876.60</TD><T D>5.22</TD></TR><TR><TD>Maggi</TD><TD>788.94</TD><TD>4.70</TD></TR><TR><TD>bert</TD><TD>701.28</TD><TD>4.17</TD></TR><TR><TD>Mark F.</TD><TD>701.28</TD><TD>4.17</TD></TR><TR><TD>DirtFarmer</TD><TD>701.28</TD><TD>4.17</TD></TR><TR><TD><-TT-></TD><TD>701.28</TD><TD>4.17</TD></TR><TR><TD>NauTiLUS</TD><TD>701.28< /TD><TD>4.17</TD></TR><TR><TD>hbitsch</TD><TD>701.28</TD><TD>4.17</TD></TR><TR><TD>Andy Watling</TD><TD>701.28</TD><TD>4.17</TD></TR></TABLE>

One obvious problem is that many people appear to have exactly the same number of hours. That's because on Seti's page, "Total CPU time" is listed in years, with two decimals precision. Not much I can do about that. That's also the reason why this is actually a top-22 (hbitsch and Andy Watling fall just below the top-20, but have the same number of hours).

So: interesting? I can just include it in the "weekly" (ahem ) reports I post here.

Martin

(Edit: whoops, that didn't come out quite right... This better?)
(Re-edit: buggery fsck...)
(Re-re-edit: waaaaaah, if it still isn't right, you'll have to live with it . UBB keeps messing with my table.)

[This message has been edited by Ees (edited 10 August 2000).]

[Rattledagger]

Martin, your comment seems to indicate using the wrong approach on top cpu-hours. 0,01 years is 87,6 hours, a much too big error. For Rags, this is 30 units.

But seti@home also rapports average cpu-time per unit. This is in 0,1-seconds. For a user with 8000 units, 0,1 second difference is only 13m20s, or 0,000025 years. Only then you've completed more units whan SGI SETI will this approach give a bigger error.

[cjolley]

Why am I arguing for a measurment that would put me way down the list?
chuck

[Brian R.]

Cool, I finally made some list or another. How about a list of handles in alphabetical order?