[Danometer]

See http://setiathome.ssl.berkeley.edu/s...am_type_4.html

According to the Jim-Dandy-handy spreadsheet, we should:

Pass Scarfy for #59 in about 12±2hours.
Pass ET phone FCSM for #58 on 19 October.
Pass alt.comp.hardware.overclocking for #57 on 20 October.
Pass SETI-Hispano for #56 on 27 October.

Over the past seven days, we've averaged 635 WU/d (one WU every 2 min 16 sec). Overclockers Network (#55) did 731 WU/d. Hardware Upgrade - Set@home (#68) is doing 774 WU/d. Only 22 teams are outpacing us, including all of the top 16.

FYI: Team Ars Technica Lamb Chop is knocking out one WU every 10.4 seconds.

Keep-a-crankin'.

[This message has been edited by Danometer (edited 10-14-2000).]

[Danometer]

We've had a significant slowdown, due to the loss of a couple of high-placing members and migration to SETI 3.00. While the first factor hurts, it is temporary. The second factor is affecting all teams equally; it just means that it will take longer to pass people.

From 14 Oct at 10:14 UTC to 20 Oct at 04:11 UTC, we averaged 432 WU/d. There are twenty-seven teams crunching more than us. Something happened to Hardware Upgrade - Set@home (#68). They had been outpacing us, but they did only 169 WU/d since the last Club Teams update.

Due to the recent slowdown, we should pass (based on current crunching rates):

ET Phone FCSM for #58 on 22 Oct.
alt.comp.hardware.overclocking for #57 on 23 Oct.
SETI-Hispano for #56 on 03 Nov.
Brasil for #55 on 18 Nov.

For those obsessed with the Overclockers Network, they are still leaving us behind. They are doing 679 WU/d, thus extending their lead by 247 WU/d.

More power…

[proux]

Thanks for the update.

What program do you use to track that?
Maybe it would be a good idea to automatically post these numbers somewhere on the site.

Good crunching

[Danometer]

SETI Checker from http://www-users.rwth-aachen.de/stef...ICheckerUS.sit gets the data and Appleworks 5 runs the spreadsheet, plus a little statistical trick I learned from a retiring math professor that has no math proof behind it but works great for forecasting.

Off note: do NOT "downgrade" to Appleworks 6. I saw it "action" at work and was underwhelmed. It ran slower on a G4 (Sawtooth-500 MHz and 256 MB RAM) than Appleworks 5 on my iMac Rev B (233 MHz and 160 MB RAM).

------------------
I am an evolutionist, and they can have my opposable thumb when they pry it from my cold dead hand.

[Danometer]

Apparently the Club Teams list at http://setiathome.ssl.berkeley.edu/s...am_type_4.html is now being updated weekly.

We passed SETI-Hispano for #56 yesterday. We averaged 907 WU/d for the past week, one WU every 95 seconds; only 24 teams surpassed our crunching-rate. Hardware Upgrade - Seti@home (#65) got back on track and are again outcrunching us, but it would take until July 2001 for them to pass us at current rates.

At current rates, we should pass:

Brasil for #55 on 05 Nov
alt.games.diablo for #54 on 10 Nov
Nordlichter for #53 on 19 Nov
Team 3DNow! for #52 on 23 Nov
TidBITS for #51 on 25 Nov

Keep it up folks, the Top 50 is in sight! Crunch-crunch-crunch-crunch-crunch!

[Danometer]

Just when you THINK you have Berkeley's pattern figured out, they take another hit of acid and change their pattern. The Club Teams list was updated six hours ago.

We are still #56. We crunched 714 WU/d since Saturday. Only 24 teams outcrunched us, and this time Hardware Upgrade-Seti@home(#65) was not one of those. The number was good, but down almost 200 WU/d from Saturday, thus delaying our passing of other teams.

Forecast

Keep in mind the forecast projects for all teams to remain at the current rates. Other teams will be changing places along the way also. Significant rate changes will cause significant forecast changes.

Brasil for #55 on 06 Nov.
alt.games.diablo for #54 on 15 Nov.
Team 3DNow! for #53 on 24 Nov.
Nordlichter for #52 on 28 Nov.
TidBITS for #51 on 01 Dec.

[cwasko]

go-Danometer-go...

Thanks for the regular stats!

[seanyepez]

Expect twenty work units or so from me daily. I'm going to seriously start crunching.

[Psych]

any updates yet?

[Danometer]

Not to the Club Teams list at http://setiathome.ssl.berkeley.edu/s...am_type_4.html .

See http://forums.macnn.com/cgi-bin/Foru...ML/000132.html for details on current position based on data from the Overall Teams list http://setiathome.ssl.berkeley.edu/s...am_type_0.html .

[tadd]

We're #53!

------------------
Tadd Torborg

[slboett]

SETI 3.0 effin' blows! If I can find 2.0.4, I'm going back! I almost want to disable this piece of %^&*.

Scott

[reader50]

I agree. I have the 2.03 installer, if you want it I will email it. I would like to have 2.04 myself.

[DocWest]

I've got 2.04.
I just put it on my web site.

GET IT HERE

[reader50]

Thanks.

[keyser_soze]

well, they updated the club teams page, we're now officially #37
congratulations!
wonder if OCN finally believes that the seti stats page was screwy now that both of our teams jumped way up in the standings.

[slboett]

Thanks Doc!

Scott

[reader50]

Club teams has been updated again. We dropped to #39. Nordlichter (#37) blew past us for one of those places - they may have added some heavy hitters. The other party that pushed us down is probably rnoggin "The Guru" (#23) that we have been hearing about playing games at the top.

[Danometer]

We're back up to #38 as of 0915 UTC Monday with the updating of the Overall Teams list.

28 Quebec 233053
29 Overclockers Network 223546
30 BP6/VP6 User Group 215840
31 Yahoo SETI Club Team 204685
32 Hungary 202170
33 Royal Astronomical Society of Canada 191564
34 Austin Shaggers 183354
35 Team Denmark 183120
36 Team RUSSIA 180859
37 Nordlichter 174823
38 Team MacNN 172311
39 COMETS&CATS!@SETI 171950
40 The IsraAliens 166151
41 Project Blue Book 165906
42 FurSETI 163859
43 Sluggy Freelance 163254

With the adjustment of Nordlichter, I need another datapoint to make a new forecast. Maybe there will be an update tomorrow.

[Danometer]

Sure enough, they provided another datapoint, and actually a hour before I wrote the above; they just hadn't posted it yet.

At 797 WU/d, we are the 21st most productive club team. This is a much better rate for passing other groups, but even more would be even better.

30-Day Forecast

Now that SETI seems to have some things straightened up and if "guru" will pick a group and stick with it, a forecast can now be projected. At current rates, we will pass:

Nordlichter for #37 on 15 Dec.
Team RUSSIA for #36 on 29 Dec.
Austin Shaggers for #35 also on 29 Dec.
Team Denmark for #34 on 03 Jan 2001.

200k WUs on 09 Jan 2001. If we can get 975 WU/d (a 22% increase), we can make 200k on 01 Jan 2001.

<Insert usual disclaimer here.>

[This message has been edited by Danometer (edited 12-06-2000).]

If anyone can help me...

I'm kinda new to the team and just kinda went with the defaults that the SETI program had. I didn't get much done. Then I changed some of the settings and things moved much quicker.

Question:
What would be my best set-up to get units done fastest?

System:
Dual 450's
384 RAM
DSL (Does this even matter!?)
SETI 3.00

Are there any set-up secrets that will make my box work faster?

Thanks

[druber]

Welcome aboard, Catatomic. Sure, dsl will help, if only because you won't have to be there when your computer finishes a work unit (wu). dl/ul speed isn't much of a factor.
The basic speed tips are to set the screensaver to go to a black screen after one minute. Also, putting your seti data onto a ramdisk will give you a substantial boost. The seti station (http://www.flex.com/~daniel/SETI/) has some applescripts to help you set this up and maintain it more easily. Hasn't been updated in a while, but it still works great.

So that's a start anyway. Of course, speed matters a little less every time you turn your computer off, so just leave your Mac on 24/7 and crunch away..

How much of a boost does using a RAM disk typically give you? 10%? 20%?

[reader50]

The SETI Team lists are not updated as of this post, but it was time for an update. By checking the individual team listings:

In Club Teams we were #36. We have since passed Team RUSSIA (#35) and Austin Shaggers (#34) to take the #34 position. Team Denmark is dead ahead at #33.

In the All Teams list, we were #66. We have since passed Nordlichter (#65), Alcatel (#64), Texas A&M University (#63), Team RUSSIA (#62), Austin Shaggers (#61), and University of Twente (#59) to take the #60 position. Team Denmark passed University of Twente before we did and currently occupies the #59 position.

For the last ten days we have averaged 862 WU per day. At our current stats we are closing a 2,971 WU gap with Team Denmark at a rate of 388 WU per day and are projected to pass Team Denmark on Wednesday, December 27.

[reader50]

Berkeley has updated the Club & All Team listings. Trouble is, the listings do not agree with each other.

In Club Teams, we are #36 now, with Nordlichter #35 & Austin Shaggers #33 somehow ahead of us again. Nordlichter in particular is just ahead, it looks like we are about to pass them for the third (or fourth?) time. Hopefully in the next few days, but who knows the way they jump around. Maybe they have friends in Berkeley. It will take a few more updates before there are enough data points to plot Nordlichter accurately.

Austin Shaggers apparently had not had their team page updated lately. We had not passed them after all, I was looking at old data.

The All Teams list is nearly 6 hours older and shows us ahead of Austin Shaggers. Austin Shaggers had not beed updated yet at that time. While the All Teams list shows us as #60, we are probably #61 now, with Nordlichter #60, Team Denmark #59, & Austin Shaggers #58 ahead of us.

The Forecast
Assuming current data is reliable, we will pass:
Team Denmark on Tuesday, December 26
Austin Shaggers on Saturday, January 5
Nordlichter, who knows.

Take this Forecast with a grain of salt. I expect erratic SETI production from all teams during the holidays.

ps - Danometer, get back here!

[reader50]

Berkeley posted an update to the Club & All Teams today. I spot checked some individual team totals near MacNN with my spreadsheet and the numbers add up OK this time.

In Club teams we are now #34 having passed Team Denmark #35.
Austin Shaggers #37 has fewer units now than during the last update and is apparently behind us. Either a bunch of people high in their team bailed, or their stats are suspect. We have passed them in the official stats again, but I can't tell if it will hold.
Nordlichter #33 is ahead of us again by 2,000 units because our daily average fell to 612 while Nordlichter only fell to 794. Before the holidays began Nordlichter was producing 996 units per day. If our production returns to 1000+ after the holidays we should presently pass them, but no forecast can be made. Right now they are adding to the lead while the holidays continue.

In All Teams we are now #59 with Nordlichter #58 & University of Texas at Austin #57 ahead. Both are producing about the same as us during the holidays or somewhat more.

Until our production returns to normal, no Forecast can be made. However, the future looks bright, we are over 500 team members and growing faster than any of the teams near us. I expect our production will increase faster than the competition's production does.

Happy New Year's everyone and let's bury some more teams in 2001!

[reader50]

January 13, 2001

We have jumped to #32 in Club Teams having passed Royal Astronomical Society of Canada #33 and Nordlichter #34 again. In All Teams, we are now #56 having passed Royal Astronomical Society of Canada #57, Nordlichter #58, and the University of Texas at Austin #60.

This time we crunched 1,324 units per day, the best I have seen so far. We outcrunched all but11 teams. The top 7 teams are all still outcrunching us. Of greater concern, three teams behind us crunched even faster than we did. The IsraAliens (#35 in Club Teams) crunched an amazing 2,588 units per day, if they keep it up they will pass us by Tuesday. DSL Reports #41 crunched 1,762 units per day, we don't have to worry about them until late February. And OcUK - Overclockers UK #48 crunched a huge 3,195 units per day. At current rates, they will pass us around the first of February.

For those interested, we outcrunched OCN #24 by 304 units per day, Team MacObserver #19 by 761 units per day, and O'Grady's Go2Mac.com Team #13 by 923 units per day. All three would be months away if current crunching rates continued. Team MacAddict #3 had incorrect numbers this time but are probably outcrunching us by at least 2,000 units per day.

The Forecast

• January 16 - IsraAliens #35 may pass us. Not enough info at this time to know if they had someone large join, or if they are really crunching that fast.
• January 30 - we will pass Yahoo SETI Club Team #31.
• February 1-5 - OcUK#48 will pass us.
• February 6 - we will pass Hungary #30.
• Late February - we will pass Quebec #29 and be passed by DSL Reports #41.

More info will be available after Berkeley's next update. Munch.

[Misha]

The "huge" work unit crunching weeks are due to members joining our team who already have units... we got a guy last week who had 3000+ units, so that boosts our average for that week by about 350.

I imagine the same is true for the other teams that suddenly jumped in the weekly output.