Archives

September 2000

Sci-Fi Fans
If you happen to be a fan of good Science Fiction television (well in the US anyways).  The Sci-Fi Network started yesterday to show the series Babylon 5.  This is a series, which is actually a series...a beginning, middle and an end.  The story was meant to be shown in a 5 year span...and each episode isn't a disjointed mess in which most Sci-Fi series tend to be.  I highly recommend it for anyone that has access to the Sci-Fi network.  Yesterday's episode was the first part of a two part pilot....you didn't miss *too* much, but if you catch it at the beginning, then you will not be wondering what happened in all the previous episodes :)

Sorry...
For the Bastard downtime. As many of you probably are aware, the benchmark page used to be hosted by Guru. With his departure from the team, we lost our really nice nifty table. Since I have no skill, and no time, I haven't been able to come up with anything to get it back up. So, for you Benchmarking enjoyment, I've cut and paste it together over several different sheets. It doesn't sort, and don't get it wet, or the paste holding it together will dissolve. If anyone needs me, I'll be in the corner. Eating the rest of this paste..  -Rat

(hrmmm...i'm gonna teach Rat to update the front page with the current news on it... hehehehe things looked a bit retro for a bit ;)

Planned Server Downtime
This just in from the S@H servers.........servers will be down for an hour and half tomorrow:

OUTAGE NOTICE : we will be having a 1.5 hour outage on 9/26/00 in order to apply patches and update the firmware on our RAID subsystem. This will begin at 20:00 UT.

Be Prepared...

An Explanation or Two
Twice within the past week there have been Berkeley server problems in which people were getting strange errors when trying to upload work units.  Things have been quiet about what was going on, until a post today on the technical news section of the SETI website.  Their explanation:

September 25, 2000
Last night (Sunday, September 24th) and a few nights ago (Thursday, September 21st) we experienced two sudden server outages, both due to running out of swap space. Users were getting "Unknown Error #1" messages during these times. We are re-tuning the server now.

There was another outage on Saturday for the team page here.  The problem turned out to be a DNS server crash for the site.  Unfortunately the warning system that should alert Jason when the server would go down didn't work, so it took some time for the news to get to him.  Sorry for the problem.

3,000 Members
That is cool.  Welcome to the new members who joined recently to push the team total members over the 3000 member mark.  Each of ya help make this team what it is and thanks for the help!  We are still kickin butt and taking names....there are three more teams to go and lets haul em in!

Weekly Roundup
The weekly stats are up for this past week and there has been quite a bit going on.  There has been a bit increase in the members for the week, and quite a bit of milestones hit for the week. 

Axl squeeks into the top 300 this week posting the top mover for the week moving up 46 spots.  SpeedGeek and Ross -TLC were battling it out for 2nd and third this week with SG coming in second.  Tops this week in work unit production was again Admins spend CPUs.  Knight tried as he might but fell just about 100 work units short of tops this week.  Try harder this week Knight!  Moving in at third and cranking things up again is Vomesh.  He hopped in at third and hopes to stay there a while!

Take note at the larger graph for members in the top 1000....things were getting kind of suffy on the graph so I made it a bit larger to make it easier to see what is going on.  TLC now has 21 members in the top 1000 S@H members overall....That ain't bad!

There has been some moving in the ranks of the top 40 in the past week....leading off things Knight has moved up to the team #7 spot passing Gonzo several days ago.  OoklaTheMok passed up aenima, but he himself was passed by both Vomesh....leaving him in the #12 spot.  Vomesh made a big move today upping production and leaving 7 others in his wake moving into the #11 spot.  Greens in Oz moved up several spots and into the #13 spot.  Virus - TLC moved up passed NSteussy and into the top 20.  

Joker moved past Turbo into the #24 spot, R.Marvin moved up one to 27, while Dan Dan the Linux Man and 77rickyli battled it out and DDTLM moved passed and into the #28 spot.  MrCatbert cruised up one spot to #31.    -zAmboni

Where to Begin?
There appeared to be some server problems over at Berkeley last night.  Alot of people could not send in results, but could download new work units....weird eh?  Not sure why this happened, but they haven't posted anything on the SETI homepage for an explanation.  Things seem to be ok right now though.

Complaints?
Well there are still alot of people who are complaining about the SETI clients not being optimized for this or that instruction sets for the CPU of their choice.  There was another post with this sentiment today on alt.sci.seti.  Hiram (who is the organizer for the porting of the client to different systems) responded swiftly and justly with this post:

Good Morning SETI Fans:
The often commented upon "deficiencies" of the SETI@Home project are easily understood in the context of the following:

1.  There is a very long wish list of tasks and desires facing the project.
2.  There is a finite amount of machine, people, and monetary resources available to the project.
3.  The list of tasks and options are ordered according to priority relative to the aims of the project.
4.  The project is a science experiment, not a CPU benchmark.
5.  Within this context, the priority list of tasks has optimizations for particular processors below the cut-off line of possibilities at this time.

Over all, when you consider the paucity of resources that the project is dealing with, I'd say they are doing a most amazing job.

--Hiram - WB6RSS

I tend to agree.  Even with the recent money influx, earmarked for the SETI projects, that money hasn't been available to the SETI@Home project (yet).  AFAIK there are only a total of 4 people who help keep the SETI@Home project going.  And all 4 of those staff members do so part-time.  When anyone has any complaints or gripes should take a good look at the above post and think again about those complaints. 

Alternative Olympics
Tired of the (day old) Olympics on NBC?  Want to compete in the Olympics yourself, but aren't quite athletic enough?  Well there is hope for you after all.  Enter the Geek Olympics.   You too can compete in thrilling events such as the Software Buyathalon, Code Wrestling, or even the CPU Tower Jumps.  There may be some time left to enter some events....so go check it out now!

Welcome!
There was a big addition of members in the past 24 hours, and I wish a good welcome to all.  I hope you enjoy your stay with Team Lamb Chop! 

Stats Pages Problem
No not the stats pages here!  It seems that there is a problem on the updates on some/all of the stats pages over on the Berkeley servers.  I haven't really noticed that much of a problem since the pages I pull the stats for seem to be ok (for me at least).  But Matt Lebofsky chimed in with the following:

Due to the natural swelling of the database, aggravated by recent database backups slowing things down, the stats pages at setiathome.berkeley.edu have been kinda screwed up. Pretty much a new stats-page process would start before the last one finished.  I'm going to fix this so that it won't happen again in the future. It'll take a day or two for all the pages to get fixed - until then, thanks for your patience.
- Matt -

Beta 2.76 Reports
    I personally haven't installed the 2.76 beta on my machines (I'm going to wait till the Version 3.0 client hits the Berkeley servers), but there are some early reports on the latest beta to be released.  Roelof posted that it seemed on his machines that the 2.76 client may be about 17% faster than the 2.74 client.  I for one didn't expect it to be faster because they only seemed to be some minor bug fixes in the client, but Lawrence Kirby chimed in with this:

>> 2.76 should be faster at gaussian searches.
>Why is that Lawrence?

An inefficiency in the gaussian search code has been eliminated.

>I found a healthy 17% speed increase for a 0.417 angle range, as reported on alt.sci.seti.

If it is down to the gaussian code then very large and small angle ranges should be unaffected.

Well that is a good thing I guess.  The inefficiency noted must be the thing that caused the work units crawl to a slow at the beginning and end of work units....plus the extremely long times on work units with very large and very small angle ranges (sometimes processing twice the normal times).  I do want to point out that the 17% reduction in times is compared to the 2.74 client.  The 2.70 client seemed to be the fastest of all the betas released, but that isn't surprising since the 2.70 client did less searching than any of the betas released after that.  Lets just hope that there are no major problems with this last beta release and the full version 3.0 is released soon.

1.5 Million and Counting
Team Lamb Chop passed the 1.5 million work unit mark today.  That is a heck of alot of work units!  Congrats to each and every member of TLC.  Each work unit that you have crunched has gone into that total, and without the collective effort of the team, we would have never been able to pass this milestone.  Now everyone go around and pat each other on the back.  The team also passed another milestone several days ago.  The team passed the 2000 years total processing time...  I know we have some old timers on the team, but I really doubt that they have been around that long ;^).  A third milestone that should be reached in the next week or so will be the total members mark.  Soon we should pass the 3,000 team member mark.  Definitely alot of members and alot of SETI crunchin'.  Way to go...and crunch on!

New Beta Out
Hot on the heels of the news posted late last night, follows the release of the version 2.76 beta client.  Here is a snipet of the readme file in the beta email. 

The SETI@home beta client version 2.76 fixes the remaining minor bugs that the beta testers have reported, as well as tunes the final bits of scientific code.

NEW BUG FIXES:
There isn't much difference between 2.74 and 2.76, except for some science tweaking (adjusting thresholds so the result files didn't fill up with too much garbage) and some small bug fixes which were mostly cosmetic. We did make one interim build (2.75) and found that there was an actual bug in a Windows time library routine that caused an invalid page fault when given a negative number. This was not a visible problem for most users, but we added a fix to make sure this routine was never called with negative values.

At any rate, we felt enough has changed to warrant one final beta. If everything is working as expected, then this exact version will be renamed 3.0 and released to the public.

By the way, a couple testers complained about "program file corrupted" errors. We believe this is a fault on the server side. If you see this
happen, please send e-mail with information regarding what platform you are using, and if you are using any third-party "add-on" software.

All Is Well
Looks like everything is all fine and dandy with the site now.  I have been getting direct access to the site since Saturday.  Hope that the access for y'all is without problems now.  Jason notified me that he is working on becoming a web registrar,  when that happens, HE will have the say on when and how things will get done and if there is any type of changes in the web site they will be done IMMEDIATELY.  Lets just hope that nothing like what happened last week will happen again  (crosses his fingers).

Version 3.0 News (finally)
I ran across this on the alt.sci.seti newsgroup tonight....Eric Korpela is here to bring the news:

Well, things got bogged down for a little while.  We're tweaking the gaussian chi-squared threshold a bit to cut down some of the low-quality detections.  We've also fixed a dozen or so bugs in 2.74, and we're hoping to distribute a new beta in the next couple days. 
I'm trying to push hard for the full release to go before the end of the month.
Eric

DON'T expect this to mean shorter run times in the next beta released.  I believe that this only means they are changing the threshold for sending back results to the Berkeley servers.  The client would still do the same searching, but they will be more selective with the results from those searches being sent back. 

Anniversary Almost Forgotten
I almost forgot this one milestone that happened a little over a week ago.  September 8, 1999 was the launch date for the team statistics website for Team Lamb Chop.  Back then, the stats were limited and the site was hosted on my personal web space on my ISP.  Sometimes it is hard to believe that I have been doing the stats for the team for over a year, but I guess that I have :^).  The team has grown and the content in the statistics have grown also.  Unfortunately, detailed stats news that I used to do have declined in the past year.   The additional stats that I pull have taken away from the time I would need to do that detailed news.   If you want to stroll down memory lane, and check out some old stats pages, check out the old stats news archives.  Some of the formats of the pages are strange, I started to convert the old page into the format this page has here, but I never really did finish them :).  Here's to another year of stats....bigger and better and hopefully #1 this time next year!

Weekly Stats
Yup, they are up!  Check 'em out here.   There are some swift movers from the past week....they were lead by SpeedGeek and Anrky.  The top movers have been dominated by members in the 200-300 range, but they will find it a bit tougher to move once they crack the top 200.  Knight moved back into his usual #2 spot in Work Units for the week behind Admins spend CPUs.  On the milestones front Admins spend CPUs leads the way breaking the 60,000 work unit barrier!  Other notable milestones include Virus - TLC and Action both breaking the 10,000 WU mark.  Virus - TLC, also had another milestone, breaking into the top 1,000 users overall in the SETI@Home project!   He currently weighs in at #994....That brings Team Lamb Chop to a total of 21 members currently in the top 1,000 overall.   Look for Action to break into the top 1000 sometime this or next week.  Congrats to all!

Are We there Yet?
Hopefully the site issues are starting to clear up.  I still can't access the site directly as of the time of writing this, but some ppl. are having more luck.  The problem was found last night while talking to the site's admin contact, Jason.  The TLC site is hosted virtually on one o the Nimbus Networks servers.  The problems arose from the TLC site's secondary DNS server was originally submitted different than the other (save one) virtually hosted sites.  When the server was moved and the new DNS information was submitted, the info didn't mesh with the previous TLC info, and the IP address didn't get updated for the TLC site.  The information has been resubmitted with the correct info and things should be straightened out soon.  I am sorry for the problems that the site has seen in the past couple of days, and thank you for your patience in this matter.  It has been a giant headache for Jason, me and the rest of y'all.

A Blinding Sun?
With Sun Microsystems taking over the top spot in the team standings, many of you have been wondering how this effects the plans for Team Lamb Chop SETI@Home Universal Domination.  Even with the addition of some heavy hitters on the Sun team, TLC is still producing more work units per day than every team, and we are cranking out 2,000 - 2,500 WU more than even Sun right now.  When would we pass them?  A rough estimate would put things somewhere at July/Aug 2001.  It may be a long haul to do it, but we eventually will dominate!

VBA 4 U
The fruits of my labor are starting to be realized.  I have been working on some more VBA coding for Excel to give you all some more and better stats for the site.  The major push is to get some charts going for the Overall Team data.  What I am planning on doing is to keep track of the top 50 teams and some various vital stats.  The minor push is to get everything integrated into a fast, efficient MegaWorkBook that will not take that long to update the stats on my end.  I have gotten a couple of early test pages for you to see....these aren't final, but will help give you some idea what I am working on.  These are only a few...there will be more.  Here are the pages:  Overall Teams,  TotalWUs,  WU/Day.  Enjoy!

Where is Version 3.0?
Your guess is as good as mine.  Things are quiet on the newsgroups from the Berkeley guys, so I cannot give you and update on that.  I can only assume that the silence is a good thing.  If there were problems with the last beta that was released, then I guess they would have had a subsequent beta released by now.  I have a feeling that they are working on getting their databases ready and other things set for the 3.0 release.

CBC Rulez!
And NBC sucks.  What if they held an Olympics and you knew the results before they ran a race?  If you live in the US and want to watch the Olympics, then that is what you have.  NBC has decided that there isn't one event in the Olympics that is important enough to show live to the public.  Luckily I live close enough to Canada to receive the local Canadian CBC station.  Already tonight I have seen more live coverage of the Olympics than you will see on NBC total.  There is one good thing about not having to watch NBC though.....I won't have to see the wall to wall gymnastics coverage, and I won't have to hear the words "Up Close and Personal".

Oh and by the way Jenny Thompson just posted an Olympic Record in the 100m Butterfly heats...

A New #1
For those of you who haven't noticed, SGI is King no more. Sun passed SGI today for the #1 team overall in the SETI@Home project.  How did they pass them so suddenly?  They gained two members SpeedRacer, and Cobra which pushed them over the top today.

Both SpeedRacer and Cobra work for Sun, but they also have had stints on our team.  Will Sun stay on top?  Who knows.  As many Team Lamb Chop members already know, the SpeedRacer and Cobra's team loyalties are pretty fickle.  They may be on the Sun Team Today....and gone tomorrow....ya never know.

DNS Problems Still Abound
I have no idea when things will get straightened out with the DNS and the IP address issues that have been going on since Saturday.  Jason who takes care of the hosting, says that everything is OK on his end, but it still looks like there is a DNS propagation problem somewhere down the line.  The strange part of it is that www.nimbusnetworks.com (216.139.208.171) is being hosted on the same box and same IP address as this site.  The Nimbus site will come up without a problem, but the team site doesn't.  About the only reliable way I have found to get to the team site here is to first go to the Nimbus site, and then retry the team site.  I usually can get through this way (or until the browser etc loses the DNS settings....)

For Your Viewing Pleasure
I just finished doing the weekly stats.  (Is it Friday already? :^).  Things have picked up in the movement category this past week.  There are several ppl tearing up the charts...leading the pack is ichijo moving up 58 spots this week sneaking in at #296....he/she was followed by Anrky and tasm both tearing up the 200 level looking for more.  Knight gave Admins spend CPUs a break this week letting them stand alone at the top work unit chart....but beware, he gave all his work units to Rat Bastard today (all 1000 plus of them)...he will be back in the hunt soon.  Memnoch and svdsinner were second and third for the week.  Anytime you can post 1000+ work units a week is a good thing!

Sorry...
For the server downtime.  They had to move the server (physically) so it was down for a period today.  It also looks like they had to change the server IP address so it had to propagate.  If you can read this it has propagated to you!   If not then you aren't reading this right now :).

SETI Gets Some Platters
The SETI servers were down for a bit on the 7th.  They had to shut down the servers to add a couple more hard drives.  You guys produce so much data that they were running out of space!  Hopefully they got some of those kick ass ultra fast platters and it may help speed up some of their processing.  One can only hope.

Hrmmm..
Has it been that long since I have updated the news?  Woah.  I finally have up the weekly stats for this past week.  I also noticed that I goofed with a couple of the previous graphs deleted some that I shouldn't have, and mislabeled another graph on the weekly stuff.  You may want to check the previous weeks stats to see if you missed anything!

Can You Spare a Work Unit Mac?
After laying waste to Microsoft, Team MacAddict passed Intel a couple of days ago.  Congrats!  Don't worry, they aren't a threat to Team Lamb Chop...at the moment at least :^).

No News?
There hasn't been any rumblings about the release of version 3.0 yet.  I haven't really seen any type of serious bugs on the newsgroups, and even less feedback from the SETI guys over at Berkeley.  I guess you can say this is a "good sign".  If there aren't any problems with the last beta, expect the version 3.0 release sometime soon.  Again, I am not sure how they are going to roll it out...so stay tuned.  Version 3.0 is going to put a hurtin' on average work unit times...the current estimates put the version 3.0 client running 25-50% slower than the 2.04 version.

More Stats Goodness???
I'm working on it.  I am getting some more goodies for the overall team stats coming up.  I am in the organization phase, and about to start the coding phase.  Look for some new stuff in the next couple of days.     -zAmboni

Woah...
With yesterday's update, the SETI folks FINALLY flushed the backlog of work units that needed to be updated into the team totals.  The totals on the SETI pages actually matched the totals of the team individual members.   It has been quite a long time since those numbers were accurate.  Needless to say, the numbers I have had for the teams were accurate numbers :^).  Unfortunately with today's update, some more weirdness from their numbers show up again.  When updating the overall team numbers.  The totals that the SETI pages had for each team are actually OVER the totals of the individual members...to the tune of about 1000-1500WUs.  Weird.  On top of that, again, it doesn't look like members joining a team with work units have those numbers added into the totals, and if a member leaves with work units they aren't taken away.  Bummer.  I am trying to get a grip on what is going on over there, because I want to do some VBA mojo on the team numbers for the stats pages, but I may be limited to what I do.   They seem to have flushed the backlog, but again, there is still some stats database issues that haven't been rectified.  

Patience...
Some of you may have noticed that with the last update 7 - 3 does not equal 3 like the stats main page shows.  I guess my new stats on that page are still in beta and I need to do some debugging to figure out.  Hopefully I will get that done soon.

Version 3.0 Command Line???
One of the questions with the upcoming version 3.0 client being released soon, is how long will it take them to get some of the other version 3.0 clients out.  Yes most of the people will be using either the Win9x/Mac GUI versions (on whole not necessarily with Team Lamb Chop)...but when can we expect the command line/unix clients?  there was an interesting post yesterday by Eric Korpela  when someone asked him about a version 3.0 CLI.

>Also, what's the status of v3 command line for winnt?

Last I heard we had lost our NT porter.  I don't know if Hiram has found a replacement.  If he's reading this, maybe he'll have an answer.
Eric

Hiram hasn't answered back....so there may be a delay for a command line client after the version 3.0 client hits the shelves.

Weekly Stuff
I forgot to point out that the weekly stats were updated a couple of days ago.  We did get SouthPark back for a couple of days, but looks like they are back into production...hopefully we can get them to crunch some more soon! :)

-Front Page