(last updates 28. Aug 2004 and 31 Mar 2005, look there ) New own BOINC/SETI client download page
help to find a signal!
You may proceed directly to the overview with statistics and downloadable software on a site of the famous Berkeley University of California (this about 100,000 citizens counting city lies at the East of the San Francisco Bay), which recently changed to a new client software and website:
or you may read my explanations about the project, in which I also participate now for a while --- I have nearly 20000 work units calculated at 28. Aug 2004.
Further News about SETI you can read at
what will happen?
After downloading the appropriate software for your hardware and operating system you can start the program, which logs you automatically in to the Berkeley server. There you type in some informations (email address... and account key after signing up) in a short dialog and how much or if at all these informations may be announced on that site. Than you can get a source data packet. All files are simple text files with extension .txt and therefore quite transparent. Time of processing depends also of the hardware, software release and operating system, it may vary in a range about 1 hour to 30 hours per work unit or even more. Each work unit has a file size of about 340 to 350 kBytes, but it may be difficult to download it without stalling connection - due to the Internet overload especially in the "rush hours" and the heavy load of the Berkeley server. They try now to upgrade their resources, but it may take a while to complete. With the newer BOINC client (since 14. May 2004) this behaviour becomes more diverse, but also better steerable.
After completion (you can influence the exact behaviour by a few option, read the README file) of the unit the result.txt file (about 1 to 3 kBytes in size, so far I have encountered them yet) is transmitted back and the next work unit is due. If you have the resources, you can run the program in different directories parallel (on UNIX systems only, all MS platforms have clear disadvantages for such a project!) to process also more than one unit at a time. This is especially interesting for users of high end hardware like SPARC servers or SGI workstations, I guess.
Until now I received exclusively Arecibo data (the giant 300 m radio dish, located in a natural valley), which are collected in regions of Taurus, Orion etc. These locations are somewhat restricted, because the Arecibo facility isn't itself movable and has therefore a limited observation range in the sky.
For the interested user I explain shortly the most important parts of these result.txt files, despite they are relatively transparent (you may store these files with low size locally as your personal record, you may also login at the server and read about your statistics...). They contain the cpu_time in seconds (probably only on UNIX systems an accurate number), the range of rectascension in hours.minutes from start_ra to end_ra and also the declination range from start_dec to end_dec in degrees (old), the total angle covered in degrees (angle_range), the time as Julian date (day number with decimals since an arbitrary, historical chosen and ancient starting point) and also as usual time, at what the data were collected, datas about the measured frequencies in Hz (around the "hydrogen" frequency of about 1,420 MHz) and also data about the used Fast Fourier Transformation (fft). Than a set of coordinates again together with the Julian date is following: Julian date rectascension declination (above units).
Finally the spike data follow, these are the relevant data, filtered out by the program and processed further in Berkeley for scrutiny, if they are really important. Through the filtering this last step becomes ultimately feasible... (the other data given together with the power should be now familiar by the above explanations).
The higher the power value, the more likely the spike is real, but this means not necessarily an alien origin! If it would be as simple, the research would be far more easy.
Hoping, that I could convince you to join, go back to the top
of the page...
This is the most active group on my list so far (in terms of delivered results).
A global collaboration of fans of the outstanding science fiction series SPACE
above and beyond, it aims to get counterparts of the "Chigs" "online"! If you
liked it, simply choose join (of course you have to be already a member of
the SETI@home community!). Or can I eventually convince you to join this group?
Hier ist die Gruppe, der ich seit einiger Zeit angehöre: die der Schwäbischen Sternwarte e.V. . Wer hier im Raum mittlerer Neckar lebt oder gar Mitglied des Vereins ist sowie an SETI@home teilnimmt, ist herzlich eingeladen, sich hier mit einzuklinken. Mehr Details dazu findet sich auf meiner neuen Gruppenseite. Dazu muss man einfach nur join wählen und die entsprechenden Felder ausfüllen. Als Gründungsmitglied haben sich nur langsam weitere Leute zu mir hinzugesellt und bis Mitte August 2000 war ich stets allein Führender in der Gruppe (anfangs als einziger natürlich noch zwangsläufig) --- inzwischen gibt es zwei Leute, die ähnlich viel wie ich an Ergebnissen liefern und daher mit mir das "Rückgrat" unserer Gruppe bilden.
Als Absolvent der Universität Stuttgart und in Kenntnis der beträchtlichen
Rechenleistung nicht nur am Rechenzentrum der Uni tut es mir in der Seele weh,
zu sehen, wie wenig dort SETI@home unterstützt wird. Daher der Aufruf an alle
Leute, die zur Uni Stuttgart gehören: macht mit! Übrigens ist für
das UNIX-dominierte Umfeld an Universitäten der SETI-client besonders
günstig; denn als nice-Prozess ohne grafische CPU-Zeit-Verschwendung
belastet er weder die Maschinen gegenüber normalen Prozessen, noch wird
ein Extra-Geldbetrag für die Datenübertragung erforderlich (ist in der
Netzwerkpauschale alles enthalten).
increase your SETI@home performance! back to SETI overview back to main
questions, suggestions etc. to: email@example.com