Interesting Tidbit

Don't know how accurate this is but it was fun reading it anyway:The following prediction is based on SETI@home's all teams pages last updated on Wed Apr 10 18:45:35 2002 UTC.
The updating process begins at 0030hr UTC and takes about 2 hours to complete. Therefore, please do not be alarmed if you find the following info inaccurate during the updating period.

Last team results: 685306
Last average WU/day: 4524.67

Rank in all teams: 31
Overtaking rank 30 Athlonmb.com Seti Team (710175WUs @3237.71WU/day) in 19.32 days.
Overtaking rank 29 OS/2 Warp (738741WUs @987.5WU/day) in 15.11 days.
Overtaking rank 28 Cisco Systems (750888WUs @592.86WU/day) in 16.68 days.
Overtaking rank 27 Unisys (768519WUs @570.86WU/day) in 21.05 days.

Rank in Club teams: 21
Overtaking rank 20 Athlonmb.com Seti Team (710175WUs @3237.71WU/day) in 19.32 days.
Overtaking rank 19 OS/2 Warp (738741WUs @987.5WU/day) in 15.11 days.

As for the total team WU, I have the data found in this page: 686,339 WU doing in average 2,628.5 WU/day.

As for your intercept date, I have that we:

Stomp OS/2 Warp in 25-29 days.
Stomp Cisco Team in 26-30 days.
Stomp Unisys in 32-37 days.

And a long wait to reach AMDmb.com, as we might get near them in 130 days.

But considering that the server is currently 6,696 WU behind (if you use SETI Team it shows 693,035 WU) this number might change a little as soon as they are refreshed.

did you do the math or did you find it on the Seti Site? Nice!!

Mictlan said:

As for the total team WU, I have the data found in this page: 686,339 WU doing in average 2,628.5 WU/day.

As for your intercept date, I have that we:


And a long wait to reach AMDmb.com, as we might get near them in 130 days.

Click to expand...

I Warned of Accuracy

I make no claim to having anything to do with the computation of these figures or their accuracy. I simply thought team member might enjoy seeing someone elses juggling of the figures. Our own stats team is doing a top notch job and this was in no way intended to imply that these were in any way superior to our own people's figures just a chance to read something different.

These figures are the sole property of Seti@Work.

SkyHook

Knowing that the data is from SETI@work gives me some insight of the source of the data. I haven't found how do they gather they data, seems they do a query to the server like the one used by SETITeam. The only problem is they predictor, as the one used by Team Ars Technica, are very sensitive to huge dumps or server lags.

For example, if a new member with 5,000 WU joins a team it creates a big difference in the tendency (step ¿?) of the line. If you are estimating that the output is a line you usually fit the data to the following equation:

y=mx+b

where m is the tendency or step and b in the origin of the line.

In anyway I meaned to sound offensive or upset about this thread. If you develop a good fitting program or subroutine you are more than welcome to post your develop and I gladly incorporate it to the main body of my gathering-fitting spreedsheet .