[FAQ] Seti@home Frequently Asked Questions
Mark Taylor
Posting-Frequency: Monthly
Last-modified: 09 November 2003
Version: 3.0.10
URL: http://setifaq.org
Copyright: (c) 1999-2002 Mark Taylor, Alfred Das
Maintainer: Mark Taylor <maint...@setifaq.org>
SETI@home FAQ V3.0.10 09 November 2003
for newsgroups alt.sci.seti and sci.astro.seti
FAQ Home : http://setifaq.org
Text-only : http://setifaq.org/faq.txt
Html-enhanced: http://setifaq.org/faq.html
Anon. FTP : ftp://setifaq.org/faq.txt
To save this FAQ on you local drive click File/Save on your browser
or newsreader and choose the location where you wish it saved.
Those listed with e-mail addresses are also listed with sections for
which they are working on. If you want to supply any input to these
sections, e-mail them and put CC to Mark Taylor.
Author :Mark Taylor <maint...@setifaq.org>
Html-version :Alfred Das <ad...@home.nl>
Contributors :SETI@home team, Peter Alfredsen, Frank J. Perricone,
M. Stilgar, Arthur Schain, Ed H, Neil Rieck,
Thomas Martin, Malcolm Pack, James Birchfield,
Roelof Engelbrecht, Allen Cleveland, Chris Johnson,
Carl Sagan, Eric J. Korpela, Terry Lee, Sqiz,
David Woolley, Jan Knutar, Peter van der Kort,
David Schilling, Alfred Das, Peter Yackel, Lior Fainshil,
Eric Heien, John Pike, Steve Willner,Alfred A. Aburto Jr.
--- Comments from the Author ---
Don't hesitate to contact me if you see something in the FAQ that you
think is wrong. Suggestions, comments, additions, corrections, etc are
more than welcomed. I'll reply to every email so that you'll know if
your addition/correction will be included and if not, why. People who
contribute to the FAQ will be given credit if they wish so. This latest
version of the FAQ increases to version 3.0.0 to match the current major
version of the Seti@home Client.
Sincerely
Mark Taylor <maint...@setifaq.org>
--- Comments from the Author ---
--- Legal Chit-chat ---
This document is subject to copyright. It may be copied, distributed,
and otherwise electronically transferred, if you agree to the following
terms:
1. If made publicly available, it must be updated regularly, hereby
meaning every 30 days.
2. You agree, if making publicly available this document, or parts
hereof, to link to the official FAQ-pages:
http://setifaq.org
http://setifaq.org/faq.txt
http://setifaq.org/faq.html
ftp://setifaq.org/faq.txt
3. It would be preferred, if you want to copy this FAQ, that you notify
the author.
Further information for mirroring the html FAQ is available as html
comments in the html-version.
--- Legal Chit-chat ---
0 INDEX
1 About SETI@home
1.1 What is SETI@home?
1.2 Background
1.2.1 The Drake Equation
1.2.2 The Fermi paradox
1.2.3 How far away could we detect radio transmissions?
1.2.4 The quest for EXTRATERRESTRIAL INTELLIGENCE
1.2.5 Setup of the SETI@home project
1.2.6 What is a Gaussian?
1.2.7 Analysis of the end data from the SETI@home project
1.2.8 What are pulses and triplets?
1.3 The history and customs of alt.sci.seti and sci.astro
1.3.1 Charter for alt.sci.seti
1.3.2 Charter for sci.astro.seti
1.3.3 Naming convention
1.3.4 .sig convention
1.3.5 Labeling posts
1.3.6 Patching or cracking SETI@home
1.3.7 What is a 'vcard' and why do people tell me not to use them?
1.3.8 What is PST and PDT?
1.4 What will happen if an extraterrestrial signal is detected?
1.5 How is data collected from the telescope and transmitted to
other machines for analysis?
1.6 Are earth signals strong enough to be detected?
1.7 What if my computer finds a signal -- how will I know?
1.8 How can I hear the signal?
1.9 Is there something in it for me?
1.10 Why doesn't SETI@home release the sources for the clients?
2 Problems and questions concerning SETI@home
2.1 Speed improvements
2.1.1 What's the fastest computer to use for this project?
2.1.2 Can I make it run any faster?
2.1.3 REMOVED March 2000, was:
Can I run the SETI@home text-client on Win95?
2.1.4 Will SETI@home run faster with more RAM (e.g., 256 MB instead of
128 MB)?
2.2 REMOVED April 2000, was:
I'm using a proxy server, and I can't connect - what do I do?
2.3 I had a work unit that got returned after only 5 minutes. What's
wrong?
2.4 I heard I was getting the same work unit as everyone else. Is
the program wasting my time?
2.5 My computer wanted to upload to the SETI@home server but said it
couldn't connect or reported error 10065. Are they still there?
2.6 What if someone fakes a result to make it seem like they found a
signal?
2.7 SETI@home keeps getting a 'Bad Header' error. What can I do?
2.8 Suddenly, without warning my system crashes - what should I do?
2.9 I can't see the new WU's I've processed in the status area. Have
they been registered at SETI@home?
2.10 I want to run the text-client as a service in NT - how do I do
that?
2.11 Can I run the client invisibly on Win95/98?
2.12 Sometimes the size of the workunit.txt file differs in size.
Sometimes it's 340, sometimes 341, and yet other times 351. Is
there something wrong?
2.13 I don't have a permanent Internet connection, and have to pay
for all my phone calls and net usage. Can I run SETI@home
without going bankrupt?
2.14 I already run the distributed.net RC5-64 client. Can I run
SETI@home as well, or do I have to choose which project to
support?
2.15 What happened to the gaussian information display in the new Mac
and Windows clients? The client is finding gaussians with lower
fits, do the 2.x clients find more aliens or something?
2.16 Can I run SETI@home 24/7 if I don't have a permanent Internet
connection?
2.17 Is this bad for my processor, or my harddrive?
2.18 Does it use up a lot of electricity? Is this costing me money,
or doing damage to the environment?
2.19 How can I keep appraised of what's going on lately?
2.20 General CL client issues
2.20.1 What is the CL client?
2.20.2 What CL options are there?
2.20.3 HELP, it stops at baseline smoothing!!!
2.20.4 How can I check up on the client to see how it's doing, if it
has found any signals, etc?
2.20.5 How do I tell the CL client to use a proxy?
2.20.6 I just found a bug in the -stop_after_ switches!!!
2.21 Running the CL client on Windows
2.21.1 What client should I download for Windows 95/98/2000/NT?
2.21.2 How do I start it?
2.21.3 How do I stop it?
2.21.4 How do I make Windows 2000 autoconnect?
2.21.5 Do I have to uninstall the screen saver version if I use the CL
version?
2.22 A short guide for the Linux newbie trying to run the SETI@home
client
2.22.1 What client should I download?
2.22.2 How do I uncompress the .tar file?
2.22.3 How do I start it?
2.22.4 How do I stop it?
2.22.5 How can I run it in the background rather than in a window?
2.22.6 How can I have it automatically restart if it dies?
2.22.7 What is 'nice' and how do I set it?
2.23 Why does the client timeout before windows has dialed up my ISP?
2.24 I just got a gaussian with a score of 0.50 and power 1.2, is
this good?
2.25 I looked at my stats of returned units at SETI@home's webpage,
the top gaussian I got for the last unit is missing, why?
2.26 My email address is about to change, what do I do?
2.27 Re-ordered to 2.30 (10-5-2000)
2.28 Why does the new client (3.X) take more time to complete a WU?
2.29 What are pulses and triplets?
2.30 What is an interesting pulse/triplet?
2.31 Why do certain WU's take longer to process?
2.32 Why are the most pulse searches done at a chirp rate of 0?
2.33 There's something strange with the power reported on pulses in
outfile.sah vs. state.sah?
2.34 Is the first half of the pulse graph identical to the second?
3 Third-party software
3.1 JSETITracker
3.1.1 Programmer's comments
3.2 Tk-SETI@home
3.2.1 Programmer's comments
3.2.2 Tk-SETI@home installation
3.2.3 Tk-SETI@home startup
3.3 SETI Spy
3.3.1 Programmer's comments
3.3.2 Processing efficiency
3.4 SETIWatch
3.4.1 What is SETIWatch?
3.4.2 Some background
3.4.3 Where can I get it?
3.4.4 How to install SETIWatch
3.5 SETILog
3.5.1 What is SETILog?
3.5.2 How does SETILog work?
3.5.3 RunSETI.bat
3.5.4 Where can I get it?
3.5.5 How to Install SETILog
3.6 SetiTEAM
3.6.1 Description
3.7 SETIBuf
3.7.1 Legal notice and stuff
3.7.2 General description
3.7.3 Where can I get it?
3.8 SETI Monitor
3.8.1 Description
3.8.2 Some more details
3.8.3 Where can I get it?
3.9 SETI UniT Manager
3.9.1 About SUM
3.9.2 Cost
3.9.3 Requirements
3.9.4 Where can I get it?
3.10 Setimgr
3.10.1 Prgrammer's comments
3.10.2 Setup
3.10.3 Operation
3.11 Seti4Net
3.12 Multi-SETI@home Monitor - Msetimon
4 Homepages
4.1 Homepages concerning SETI@home
4.1.1 SETI@home home
4.1.2 SETIweb
4.1.3 SETIforum
4.1.4 SETI @ SixDegrees
4.1.5 SETI@home Speedup Tips
4.1.6 Derived statistics for SETI@home @ Rovingmouse
4.1.7 SETI STATION
4.1.8 SETI: The Drake Equation
4.1.9 Sci.astro FAQ about SETI
4.1.10 Team Canada
4.1.11 The Planetary Society
4.1.12 Patch-free-Processing
4.1.13 Sky & Telescope
4.1.14 SETI Institute
4.1.15 SETI League
4.1.16 Removed 11 Feb 2001 (was SETI & Beyond)
4.2 SETI utilities
4.2.1 SETIwatch & SETIlog
4.2.2 SETI Manager
4.2.3 TKSETI@home
4.2.4 SETISPY
4.2.5 JSETITracker
4.2.6 SetiTEAM Homepage
4.2.7 SETIBuf homepage
4.2.8 SETI Monitor homepage
4.2.9 SETI UNiT Manager homepage
4.2.10 RunCache & FetchCache
4.2.11 Seti@home Service
4.3 SETI fun
4.3.1 Carolyn's Clinic
5 Acknowledgements
5.1 Sci.astro FAQ
5.2 People who have worked with the FAQ
-----------------------
1 About SETI@home
1.1 What is SETI/SETI@home?
If we assume that our alien neighbors are trying to contact us,
we should be looking for them. There are currently several
programs that are now looking for the evidence of life elsewhere
in the cosmos. Collectively, these programs are called SETI (the
Search for Extra-Terrestrial Intelligence.) SETI@home is a
scientific experiment that harnesses the power of hundreds of
thousands of Internet-connected computers in the Search for
Extraterrestrial Intelligence (SETI). You can participate by
running a free program that downloads and analyzes radio
telescope data. There's a small but captivating possibility that
your computer will detect the faint murmur of a civilization
beyond Earth.
1.2 Background
1.2.1 The Drake Equation
<This is only one possible guess at, how the Drake Equation may
be. If you want to guess for yourself, look under 4.1.8 SETI:
The Drake Equation>
Our sun is only a single star in a collection of over 400 billion
we call the Milky Way galaxy. The Milky Way is only 1 of billions
of galaxies in the universe. Seems like there should be lots of
life out there! Can we make an initial estimate? The first to do
so was the astronomer Frank Drake. He came up with a simple
equation, now called the Drake Equation, that maps out the
possibilities. The equation is quite easy to understand, so don't
tune out, even if arithmetic isn't your strong suit! Here it is:
N = R * f(p) * n(e) * f(l) * f(i) * f(c) * L
"N" here represents the number of communicating civilizations in
our Milky Way galaxy. This number depends on several factors.
"R" is the rate of "suitable" star formation in the galaxy.
"f(p)" is the fraction of stars that have planets.
"n(e)" is the number of these planets around any star within the
suitable ecosphere of the star. An "ecosphere" is a shell that
surrounds a star within which the conditions are suitable for
life to form. Too close and it's too hot; too far and it's too
cold.
"f(l)" is the fraction of those planets within the ecosphere on
which life actually evolves.
"f(i)" is the fraction of those planets on which intelligent life
evolves.
"f(c)" is the fraction of those planets where intelligent life
develops a technology and attempts communication.
"L," is the length of time that an intelligent, communicating
civilization lasts.
Let's briefly look at each of these factors separately and try to
put some reasonable numbers to them. Although the rate of
suitable star formation was undoubtedly much higher when our
galaxy formed, one can still see where stars are being born
today. In the last couple of years, several teams of astronomers
have announced the discovery of planets surrounding nearby stars.
This exciting discovery increases the likelihood of other planets
around many stars. Let's estimate conservatively that one-half of
the stars form planetary systems; the other half form binary star
systems, so
f(p) = 0.5.
The n(e) factor is a little tricky. Small stars are cool and red.
Planets would have to orbit very close to be in the ecosphere.
Also, this ecosphere would be very narrow; like the skin on an
orange. Not much room for planets. Planets that orbit very close
to their parent star are often tidally locked and present one
face to the star at all times. The atmosphere of such a planet
would freeze on the cool side that faces away from the star; this
does not promote life. On the other hand, huge hot blue stars
have a farther and wider ecosphere. Of course, judging from our
solar system, planets are spaced further apart the farther they
are from the star, so the wider ecosphere is cancelled by this
effect. These larger stars also burn their fuel faster and don't
last very long. They are usually so short- lived that life does
not even get a chance to start before the star goes nova or
supernova and destroys everything in the system. In our solar
system, with our average-sized yellow sun, we have two (Earth and
Mars) or maybe three (Venus) planets within the ecosphere. A
conservative guess for the number of planets within the "life
zone" or ecosphere is one.
n(e) = 1.
The next factor, f(l), is where things become a little sticky.
The problem is that we only have a few examples of planets where
conditions are right for life to evolve. As stated above, Venus,
Earth, and Mars all could have had, at one time, proper
conditions. We know life evolved on Earth, and there is now
tantalizing evidence for primitive life existing on Mars
billions of years ago. A conservative guess for this number is
0.2, or one in five planets with proper conditions will evolve
life.
f(l) = 0.2.
How many of these planets will evolve intelligent life? Tough
question, but if we really believe the evidence for natural
selection and survival of the fittest, most scientists would put
this number at 100 percent -- that intelligent life is a natural
outcome of evolution. Of course, here we have only one example,
earth.
f(i) = 1.
How many of these intelligent species will develop technology and
use it to communicate? If we look at the earth, we see humans
doing it, but we also see whales and dolphins, who may also
possess a moderate level of intelligence but never developed
technology. We'll set this number to 0.5 as a first guess.
f(c) = 0.5.
Now we get to the hardest number to determine. "L" is the number
of years that a technologically adept and communicative
civilization lasts. We've only been in this phase of our
evolution for about 50 years. Do advanced civilizations blow
themselves up after discovering the technology to do so? Or do
they get together and solve their problems before this happens?
For now, let's not assign a number to L. Let's plug in the other
numbers and see what we get.
N = R * f(p) * n(e) * f(l) * f(i) * f(c) * L
N = 20 * 0.5 * 1 * 0.2 * 1 * 0.5 * L
Do advanced civilizations use their technology to solve their
problems or do they destroy themselves? On earth we've survived
the first 50 years. Multiplying all the numbers gives us N = L.
In other words, the number of intelligent communicating
civilizations in the galaxy equals the number of years such a
civilization lasts! The figure about which we know the least
bears a great significance in our calculations. Most scientists
hope that if a civilization can overcome its initial tendency to
destroy itself with its own technology, then that civilization
is likely to last for a very long time. Let's hope those
scientists are right. In any case, there should be at least 50
(the number of years WE'VE been around communicating) and if a
communicative civilization lasts for millions of years, there
may possibly be millions of civilizations we can look for.
1.2.2 The Fermi paradox
By John Pike and Steve Willner
One of the problems that the Drake Equation produces is that if
you take reasonable (some would say optimistic) numbers for
everything up to the average duration of technological
civilizations, then you are left with three possibilities:
1. If such civilizations last a long time, "They" should be
_here_ (leading either the the Flying Saucer hypothesis--they
are here and we are seeing them, or the Zoo Hypothesis--they
are here and are hiding in obedience to the Prime Directive,
which they observe with far greater fiqdelity than Captain
Kirk could ever muster). -or-
2. If such civilizations last a long time, and "They" are not
"here" then it becomes necessary to explain why each and
every technological civilization has consistently chosen not
to build starships. The first civilization to build
starships would spread across the entire Galaxy on a
timescale that is short relative to the age of the Galaxy.
Perhaps they lose interest in space flight and building
starships because they are spending all their time surfing
the net. (Think about it --- the whole point of space flight
is the proposition that there are privileged spatial
locations, and the whole point of the net is that physical
location is more or less irrelevant.)
-or-
3. Such civilizations do not last a long time, and blow
themselves up or otherwise fall apart pretty quickly
(... film at 11).
Thus the Drake Equation produces what is called the Fermi
Paradox (i.e., "Where are They?"), in that the implications of
#3 and #2 are not terribly encouraging to some folks, but the
two flavors of #1 are kinda hard to come to grips with.
An alternate version of 2 is that interstellar travel is far more
difficult than we think it is. Right now, it doesn't seem much
beyond the boundaries of current technology to launch "generation
ships," which amount to an O'Neill colony plus propulsion and
power systems. An alternative is robot probes with artificial
intelligence; these don't seem so difficult either. The Milky
Way galaxy is well under 10^5 light years in diameter and over
10^9 years old, so even travel beginning fairly recently in
Galactic history and proceeding well under the speed of light
ought to have filled the Galaxy by now. (Travel very near the
speed of light still seems very hard, but such high speed isn't
necessary to fill the Galaxy with life.) The paradox, then, is
that we don't observe evidence of anybody besides us.
1.2.3 How far away could we detect radio transmissions?
By Al Aburto and David Woolley
Representative results are presented in Tables 1 and 2. The
short answer is
(1) Detection of broadband signals from Earth such as AM radio,
FM radio, and television picture and sound would be
extremely difficult even at a fraction of a light-year
distant from the Sun. For example, a TV picture having 5
MHz of bandwidth and 5 MWatts of power could not be detected
beyond the solar system even with a radio telescope with 100
times the sensitivity of the 305 meter diameter Arecibo
telescope.
(2) Detection of narrowband signals is more resonable out to
thousands of light-years distance from the Sun depending on
the transmitter's transmitting power and the receiving
antenna size.
(3) Instruments such as the Arecibo radio telescope could detect
narrowband signals originating thousands of light-years from
the Sun.
(4) A well-designed 12 ft diameter amateur radio telescope could
detect narrowband signals from 1 to 100 light-years distance
assuming the transmitting power of the transmitter is in the
terawatt range.
What follows is a basic example for the estimation of radio and
microwave detection ranges of interest to SETI. Minimum signal
processing is assumed. For example an FFT can be used in the
narrowband case and a bandpass filter in the broadband case (with
center frequency at the right place of course). In addition it
is assumed that the bandwidth of the receiver (Br) is constrained
such that it is greater than or equal to the bandwidth of the
transmitted signal (Bt) (that is, Br >= Bt).
Assume a power Pt (watts) in bandwidth Bt (Hz) radiated
isotropically. At a distance of R (meters), this power will be
uniformly distributed (reduced) over a sphere of area: 4 * pi *
R^2. The amount of this power received by an antenna of effective
area Aer with bandwidth Br (Hz), where Br >= Bt, is therefore:
Pr = Aer * (Pt / (4 * pi * R^2))
If the transmitting antenna is directive (that is, most of the
available power is concentrated into a narrow beam) with power
gain Gt in the desired direction then:
Pr = Aer * ((Pt * Gt) / (4 * pi * R^2))
The antenna gain G (Gt for transmitting antenna) is given by the
following expression. (The receiving antenna has a similar
expression for its gain, but the receiving antenna's gain is not
used explicitly in the range equation. Only the effective area,
Aer, intercepting the radiated energy at range R is required.)
Gt = Aet * (4 * pi / (w^2)), where
Aet = effective area of the transmitting antenna (m^2), and
w = wavelength (m) the antenna is tuned to.
f = c / w, where f is the frequency and c is the speed of
light.
c = 2.99792458E+08 (m/sec)
pi = 3.141592654...
For an antenna (either transmiting or receiving) with circular
apertures:
Ae = <eta> * pi * d^2 / 4
<eta>r = efficiency of the antenna,
d = diameter (m) of the antenna.
The Nyquist noise, Pn, is given by:
Pn = k * Tsys * Br, where
k = Boltzmann's constant = 1.38054E-23 (joule/kelvin)
Tsys = is the system temperature (kelvins), and
Br = the receiver bandwidth (hertz).
The signal-to-noise ratio, snr, is given by:
snr = Pr / Pn.
If we average the output for a time t, in order to reduce the
variance of the noise, then one can improve the snr by a factor
of sqrt(Br * t). Thus:
snr = Pr * sqrt(Br * t) / Pn.
The factor Br*t is called the "time bandwidth product," of the
receive processing in this case, which we'll designate as:
twp = Br * t.
We'll designate the integration or averaging gain as:
twc = sqrt(twp).
Integration of the data (which means: twp = Br * t > 1, or
t > (1 / Br) ) makes sense for unmodulated "CW" signals that are
relatively stable over time in a relatively stationary (steady)
noise field. On the other hand, integration of the data does not
make sense for time-varying signals since this would distroy the
information content of the signal. Thus for a modulated signal
twp = Br * t = 1 is appropriate.
In any case the snr can be rewritten as:
snr = (Pt * Gt) * Aer * twc / (4 * pi * R^2 * Br * k * Tsys)
Pt * Gt is called the Effective Isotropic Radiated Power (EIRP)
in the transmitted signal of bandwidth Bt. So:
EIRP = Pt * Gt, and
snr = EIRP * Aer * twc / (4 * pi * R^2 * Br * k * Tsys)
This is a basic equation that one can use to estimate SETI
detection ranges.
#######################################################################
# If Rl is the number of meters in a light year (9.46E+15 [m/LY]), #
# then the detection range in light years is given by #
# #
# R = sqrt[ EIRP * Aer * twc / (4 * pi * snr * Br * k * Tsys) ] / Rl #
# #
# If we wanted the range in Astronomical Units then replace Rl #
# with Ra = 1.496E+11 (m/AU). #
#######################################################################
Note that for maximum detection range (R) one would want the
transmit power (EIRP), the area of the receive antenna (Aer), and
the time bandwidth product (twp) to be as big as possible. In
addition one would want the snr, the receiver bandwidth (Br), and
thus transmit signal bandwidth (Bt), and the receive system
temperature (Tsys) to be as small as possible.
(There is a minor technical complication here. Interstellar
space contains a plasma. Its effects on a propagating radio wave
including broadening the bandwidth of the signal. This effect
was first calculated by Drake & Helou and later by Cordes &
Lazio. The magnitude of the effect is direction, distance, and
frequency dependent, but for most lines of sight through the
Milky Way a typical value might be 0.1 Hz at a frequency of 1000
MHz. Thus, bandwidths much below this value are unnecessary
because there will be few, if any, signals with narrower
bandwidths.)
Now we are in a position to carry out some simple estimates of
detection range. These are shown in Table 1 for a variety of
radio transmitters. We'll assume the receiver is similar to
Arecibo, with diameter dr = 305 m and an efficiency of 50%
(<eta>r = 0.5). We'll assume snr = 25 is required for detection
(The META project used a snr of 27--33 and SETI@home uses 22;
more refined signal processing might yield increased detection
ranges by a factor of 2 over those shown in the Table 1.) We'll
also assume that twp = Br * Tr = 1. An "educated" guess for some
of the parameter values, Tsys in particular, was taken as
indicated by the question marks in the table. As a reference note
that Jupiter is 5.2 AU from the Sun and Pluto 39.4 AU, while the
nearest star to the Sun is 4.3 LY away. Also any signal
attenuation due to the Earth's atmosphere and ionosphere have
been ignored; AM radio, for example, from Earth, is trapped
within the ionosphere.
The receive antenna area, Aer, is
Aer = <eta>r * pi * dr^2 / 4 = 36.5E3 m^2.
(Scientific notation is being used here; 1E1 = 10, 1E2 = 100,
1E3 = 1000, so 36.5E3 is 36.5 times 1000.) Hence the detection
range (light years) becomes
R = 3.07E-04 * sqrt[ EIRP / (Br * Tsys) ].
Table 1 Detection ranges of various EM emissions from Earth and
the Pioneer spacecraft assuming a 305 meter diameter
circular aperture receive antenna, similar to the Arecibo
radio telescope. Assuming snr = 25, twp = Br * Tr = 1,
<eta>r = 0.5, and dr = 305 meters.
-------------+--------------+-----------+--------+--------+-----------+
Source | Frequency | Bandwidth | Tsys | EIRP | Detection |
| Range | (Br) |(Kelvin)| | Range (R) |
-------------+--------------+-----------+--------+--------+-----------+
AM Radio | 530-1605 kHz | 10 kHz | 68E6 | 100 KW | 0.007 AU |
-------------+--------------+-----------+--------+--------+-----------+
FM Radio | 88-108 MHz | 150 kHz | 430 | 5 MW | 5.4 AU |
-------------+--------------+-----------+--------+--------+-----------+
UHF TV | 470-806 MHz | 6 MHz | 50 ? | 5 MW | 2.5 AU |
Picture | | | | | |
-------------+--------------+-----------+--------+--------+-----------+
UHF TV | 470-806 MHz | 0.1 Hz | 50 ? | 5 MW | 0.3 LY |
Carrier | | | | | |
-------------+--------------+-----------+--------+--------+-----------+
WSR-88D | 2.8 GHz | 0.63 MHz | 40 | 32 GW | 0.01 LY |
Weather Radar| | | | | |
-------------+--------------+-----------+--------+--------+-----------+
Arecibo | 2.380 GHz | 0.1 Hz | 40 | 22 TW | 720 LY |
S-Band (CW) | | | | | |
-------------+--------------+-----------+--------+--------+-----------+
Arecibo | 2.380 GHz | 0.1 Hz | 40 | 1 TW | 150 LY |
S-Band (CW) | | | | | |
-------------+--------------+-----------+--------+--------+-----------+
Arecibo | 2.380 GHz | 0.1 Hz | 40 | 1 GW | 5 LY |
S-Band (CW) | | | | | |
-------------+--------------+-----------+--------+--------+-----------+
Pioneer 10 | 2.295 GHz | 1.0 Hz | 40 | 1.6 kW | 120 AU |
Carrier | | | | | |
-------------+--------------+-----------+--------+--------+-----------+
It should be apparent then from these results that the detection
of AM radio, FM radio, or TV pictures much beyond the orbit of
Pluto will be extremely difficult even for an Arecibo-like 305
meter diameter radio telescope! Even a 3000 meter diameter radio
telescope could not detect the "I Love Lucy" TV show (re-runs) at
a distance of 0.01 Light-Years!
It is only the narrowband high intensity emissions from Earth
(narrowband radar generally) that will be detectable at
significant ranges (greater than 1 LY). Perhaps they'll show up
very much like the narrowband, short duration, and non-repeating,
signals observed by our SETI telescopes. Perhaps we should
document all these "non-repeating" detections very carefully to
see if any long term spatial detection patterns show up.
Another question to consider is what an Amateur SETI radio
telescope might achieve in terms of detection ranges using
narrowband FFT processing. Detection ranges (LY) are given in
Table 2 assuming a 12 ft (3.7 m) dish antenna operating at 1.42
GHz, for various FFT binwidths (Br), Tsys, snr, time bandwidth
products (twp = Br*t), and EIRP values. It appears from the
table that effective amateur SETI explorations can be conducted
out beyond approximately 30 light years provided the processing
bandwidth is near the minimum (approximately 0.1 Hz), the system
temperature is minimal (20 to 50 Degrees Kelvin), and the EIRP of
the source (transmitter) is greater than approximately 25
terawatts.
Table 2 Detection ranges (LY) for a 12 foot diameter amateur
radio telescope SETI system, operating at 1.420 GHz.
+-------------------------------+
| EIRP |
+-------+--------+------+-------+
| 100TW | 25TW | 1TW | 100GW |
-------+-------+----------+------+-------+--------+------+-------+
Br | Br*t | Tsys | snr | Detection Range |
(Hz) | | (kelvin) | | (LY) |
-------+-------+----------+------+-------+--------+------+-------+
0.1 | 2 | 50 | 25 | 28 | 17 | 3.4 | 1.1 |
-------+-------+----------+------+-------+--------+------+-------+
0.1 | 1 | 50 | 25 | 20 | 12 | 2.4 | 0.76 |
-------+-------+----------+------+-------+--------+------+-------+
0.5 | 2 | 50 | 25 | 12.7 | 6.4 | 1.3 | 0.4 |
-------+-------+----------+------+-------+--------+------+-------+
0.5 | 1 | 50 | 25 | 9 | 4.5 | 0.9 | 0.3 |
-------+-------+----------+------+-------+--------+------+-------+
0.1 | 20 | 50 | 25 | 90 | 54 | 11 | 3.4 |
-------+-------+----------+------+-------+--------+------+-------+
1.0 | 200 | 50 | 25 | 90 | 54 | 11 | 3.4 |
-------+-------+----------+------+-------+--------+------+-------+
REFERENCES:
Radio Astronomy, John D. Kraus, 2nd edition,
Cygnus-Quasar Books, 1986, P.O. Box 85, Powell, Ohio,
43065.
Radio Astronomy, J. L. Steinberg, J. Lequeux, McGraw-Hill
Electronic Science Series, McGraw-Hill Book Company, Inc,
1963.
Project Cyclops, ISBN 0-9650707-0-0, Reprinted 1996, by
the SETI League and SETI Institute.
Extraterrestrial Civilizations, Problems of Interstellar
Communication, S. A. Kaplan, editor, 1971, NASA TT F-631
(TT 70-50081), page 88.
(this section taken from sci.asto FAQ, see 5.1 for the copyright
statment of sci.astro FAQ)
Also see section 1.6
1.2.4 The quest for EXTRATERRESTRIAL INTELLIGENCE
By Carl Sagan
Cosmic Search Magazine Vol. 1 No. 2 May, 1978
Through all of our history we have pondered the stars and mused
whether mankind is unique or if, somewhere else out there in the
dark of night sky, there are other beings who contemplate and
wonder as we do - fellow thinkers in the cosmos. Such beings
might view themselves and the universe differently. Somewhere
else there might exist exotic biologies, technologies and
societies. What a splendid perspective contact with a profoundly
different civilization might provide! In a cosmic setting vast
and old beyond ordinary human understanding we are a little
lonely, and we ponder the ultimate significance, if any, of
our tiny but exquisite blue planet, the Earth. The Search for
Extra-Terrestrial Intelligence (SETI) is the search for a
generally acceptable cosmic context for the human species. In the
deepest sense the search for extraterrestrial intelligence is a
search for ourselves.
~ ~ ~ ~ ~
There are some who look on our global problems here on Earth - at
our vast national antagonisms, our nuclear arsenals, our growing
populations, the disparity between the poor and the affluent,
shortages of food and resources, and our inadvertent alterations
of the natural environment of our planet - and conclude that we
live in a system which has suddenly become unstable, a system
which is destined soon to collapse. There are others who believe
that our problems are soluble, that humanity is still in its
childhood, that one day soon we will grow up. The existence of a
single message from space will show that it is possible to live
through technological adolescence: the civilization transmitting
the message, after all, has survived. Such knowledge, it seems to
me, might be worth a great price.
~ ~ ~ ~ ~
There will surely be differences among civilizations which cannot
be glimpsed until information is available about the evolution of
many civilizations. Because of our isolation from the rest of the
cosmos, we have information on the evolution of only one
civilization - our own. And the most important aspect of that
information, the future, remains closed to us. Perhaps it is not
likely, but it is certainly possible that the future of human
civilization depends on the receipt and decoding of interstellar
messages...It is difficult to think of another enterprise within
our capability and at relatively modest cost which holds as much
promise for the future of humanity.
1.2.5 Setup of the SETI@home project
This was taken from a.s.s, and written by Eric J. Korpela,
responding to a post by <deme...@iquest.net>
The Ultra 450 (4 cpu) is the science database server which stores
results, does analysis, and also runs a splitter process.
One Ultra 10 is the user database server which stores user
information.
One Ultra 10 workstation is the server machine which handles
connections and directs them to the appropriate database This
machine also has storage for the workunits themselves.
One Ultra 10 is a full time splitter.
Two Ultra 10s (one fast and one slow) are development
workstations and an after hours splitters.
One slow Ultra 10 is the web server and Dan's workstation. (Dan
Werthimer)
One Sparcstation 10 used to be a splitter (that's the old really
slow one) and is used as a development workstation.
1.2.6 What is a Gaussian?
A gaussian is a mathematical function, mostly commonly describing
the sort of distribution of values you get around the nominal
value of some property or measurement as a result of measurement
(and production errors). I would expect the maximum speeds of
CPU chips to show this sort of pattern.
It is often described as a bell curve, as it starts off rising
slowly, then accelerates before starting to level off and come
down in a mirror image of its rise, something like the cross
section of a church bell.
1.2.7 Analysis of the end data from the SETI@home project
First thing, they'll be run through some RFI (Radio Frequency
Interference for the newbies out there) rejection routines.
There are a few different algorithms used. If a signal at the
same frequency, but from a different place on the sky comes in
within a few minutes, it's likely to be RFI.
There are certain frequencies where continuous RFI is received,
that will also be rejected. If a signal comes in at a chirp rate
of zero, it's also likely to be RFI (extraterrestial signals
should show a chirp signature due to the rotation of the earth
and/or the rotaion of the ET's planet.) RFI rejection will
probably eliminate the vast majority of the candidates (>99.99%).
From there, the probablility that the candidate signals are just
a random peak in the noise in the reciever, will be calculated.
Then there'll be created a priority list of candidates based upon
this probability, the signal strength, frequency width, goodness
of gaussian fit, etc and pointed observations of the best
candidates will be proposed.
Somewhere in this chain, there'll also be looked for repeaters.
Signals that show up at the same place in the sky at about the
same frequency, but widely separated in time. Repeaters will
likely get bumped to the top of the priority list.
Another thing that will be looked for is signals with decent
gaussian fits that show up at different frequencies, but at the
same time. That might boost a candidates' priority as well.
1.2.8 What are pulses and triplets?
In the contexts of SETI@home, they are spikes that are repeated
many times. Triplet means that there are three evenly spaced
spikes. The triplet detection takes very little time to do, but
it can only detect strong signals.
The pulse finding algorithm can find very weak pulses, the more
pulses the better the algorithm can find them. The pulse seeking,
however, takes alot more time to do.
1.3 The history and customs of alt.sci.seti and sci.astro
1.3.1 Charter for alt.sci.seti
The original documents can be found here:
The first post:
http://x31.deja.com/[ST_rn=ap]/getdoc.xp?AN=484562021&CONTEXT=
938880213.74252397&hitnum=3
The control message, that created the group:
http://x31.deja.com/[ST_rn=ap]/getdoc.xp?AN=484419771&CONTEXT=
938880213.74252397&hitnum=2
This was posted as the first post ever in alt.sci.seti on June
1st by Chris:
Welcome to "alt.sci.seti" This group will probably be a little
barren at first until it begins to propagate more thoroughly Why
not post a message and get things rolling?
Hopefully someone will one day create a FAQ for this newsgroup,
but until then here's the charter...
Charter:
Discussion about the SETI (Search for Extra Terrestrial
Intelligence) project and the search for extra terrestrial life
in general.
Also, discussion of the "SETI@home" project which allows
individuals to utilize their computer's idle time to assist SETI
in processing its overwhelming amount of recorded data.
Should discussion about the SETI@home project and the SETI
project in general begin to crowd eachother, a second newsgroup
devoted solely to SETI@home will be created with the name
"alt.sci.seti.at-home" leaving "alt.sci.seti" for discussion of
the SETI project.
Binaries are not permitted and should instead be posted to the
appropriate binary newsgroup or FTP site where they may be
accessed.
Justification:
The SETI project has been going on for some years now and has
amassed a great deal of interest. A quick search on DejaNews will
show that there are thousands of posts regarding SETI, yet there
is no currently available newsgroup to keep these discussions
from getting lost in numerous other non-specific forums.
"alt.sci.seti" will address this lacking.
It will also provide a place for the enormous and growing number
of people who have begun to participate in the SETI@home project
to discuss problems and solutions in assisting SETI to process
all of its raw data. While the SETI@home project is expected to
end around 2001 or so, it is likely that SETI will seek to call
upon the public again in a similar way. This newsgroup will
therefore continue to be timely and useful.
This newsgroup was proposed, discussed, and approved in
"alt.config" at the end of May 99.
Created 01 Jun 99
1.3.2 Charter for sci.astro.seti
RATIONALE: sci.astro.seti
The Search for Extraterrestrial Intelligence (SETI) is the
scientific discipline of searching for electromagnetic evidence
of extraterrestrial civilizations. SETI has received a lot of
attention recently due to the SETI@home project. The SETI@home
project has shown that at least several hundred thousand
individuals are willing to dedicate computer resources to the
search for alien radio signals. This has brought an increase in
the amount of discussion of SETI and the possibilities of
extra-terrestrial intelligence (ETI). Which has increased the
number of posts about SETI in related newsgroups (sci.astro,
etc.) by a large amount.
The SETI@home project is a distributed computing project which
harnesses the computing power of hundreds of thousands of
Internet connected computers to search for radio evidence of
extraterrestrial civilizations. It is the newest and most public
SETI project to date. Currently it has attracted almost a million
people willing to donate computer time to this search. However,
SETI@home is not the only SETI project, nor will it be the last
new one. Several SETI projects are on the drawing board (1HT,
etc.) and many of them will require as much or more computing
power as the SETI@home project uses currently. It would be
surprising if none of these new SETI programs use the distributed
computing model that has allowed SETI@home to harness computing
power equivalent to multi-million dollar super-computers for very
low costs.
This newsgroup will serve as a forum for discussion of SETI in
general, and any SETI projects in specific. This includes
discussion of SETI@home, both it's scientific aspects, as well as
the use, configuration, and troubleshooting of the SETI@home
client software and any similar software by future SETI projects.
Additionally, it will serve as a place to discuss the technical
specifics of all current and future SETI projects, and as a place
for teachers who are developing curricula around SETI projects
(such as SETI@home).
CHARTER: sci.astro.seti
This group will be unmoderated and distributed worldwide. This
newsgroup is intended for the discussion of the Search for
Extra-Terrestrial Intelligence. Appropriate topics for discussion
include the following:
1) Discussion of SETI projects (such as SERENDIP, Phoenix,
SETI@home, BETA, ARGUS, etc.)
2) Installation and configuration of the SETI@home client
or other SETI projects using distributed computing.
3) Trouble shooting the use of the SETI client programs.
4) The possibilities of Alien life (Drake equation, planetary
abundance and its relavance to SETI, etc.)
5) Discussion of statistical results for SETI projects.
6) The potential content of alien messages and how to decode
them, as well as any messages we (humans) have / will /
could send into space that are intended for ETI's (such as
the Voyager record, the Arecibo message to M13, the Encounter
2001 project, etc.)
7) Potential alien technology in the context of detection /
communication by / with humans (using visible light lasers
instead of radio, for example).
8) Discussion of school curricula built around a SETI
program
Inappropriate posts include:
1) Commercial advertisements of any kind, including those for
items related to SETI or any SETI project.
2) Binaries, with the exception of cryptographic signatures.
3) Discussions concerning UFOs, "alien abductions", etc,
which should take place in other groups.
1.3.3 Naming convention
When talking about your computer(s) on the newsgroups, the
following information is the minimum for other people to be able
to determine if the machine is working optimally.
CPU-TYPE, for example Pentium MMX, AMD Athlon, Pentium III....
CPU-SPEED, 133Mhz, 600Mhz....
Memory size
Memory type
FSB-speed (Front side bus speed)
Also useful to include:
Motherboard chipset
Operating System
1.3.4 .sig convention
When reading the SETI newsgroups, you will find, that many use a
* or # in their signatures. This strange way of bragging was
first seen when the first people in alt.sci.seti started getting
close and passing 100 Workunits done. They wanted to give
themselves something for their effort, so they started giving
themselves medals, putting them in their sig. Initially there was
only one character used, the star (*) which you 'earned' for
every 100 WU's. As time passed by, people started using more and
more complex notations, which evolved into the complex system we
have today.
The newest .sig convention is as follows:
The following notation is a way of expressing your personal or a
group's contribution to the S@H program.
PRECISION NOTE:
These formats imply a certain range.
2.3* would for instance mean the interval of 230-239, whereas
2* would mean 200-299. The same goes for the symbolic notation.
For instance ** would mean the interval of 200-299. More decimals
added will imply greater accuracy for FORMAT 1 and more
characters added would do the same for FORMAT 2.
FORMAT 1:
d....@d.ds
Where "d" is a digit and "s" one or a combination of the
following symbols:
!=10
*=100
#=1000
!#=10000 (ten thousand)
*#=100000 (hundred thousand)
##=1000000 (thousand thousand) etc...
Notes:
This table can be used for workunits and cpu-time alike.
"@" only as a separator.
Examples:
9@9.8! (9wu/98hr)
1.0!@2.5* (10wu/250hr)
2.1#@3.4!# (2100wu/34000hr)
3.0*#@3.6## (300000wu/3600000hr)
FORMAT 2 - a more symbolic notation:
Where the notation is composed of only symbols:
!=10
+=50
*=100
#=1000
Notes:
This table can be used for workunits and cpu-time alike.
"@" only as a separator.
Example:
****+@### (450wu/3000hr).
All symbols are counted for their value and finally added up.
Here 100+100+100+100+50=450wu in 1000+1000+1000=3000hr.
It is preferred to sort the symbols. The greater first, then the
smaller.
FORMAT 3 - only work units:
Format 1 or 2 where the cpu-time portion has been omitted.
Example:
2.1# (2100wu), ****+ (450wu).
FORMAT 4 - the most simple and understandable:
(60WU/660hrs)
DERIVED FORMATS - not preferred but no less understandable:
Mixed schemes may occur. E.g.: 4*+@3# (450wu/3000hr).
Today, there's even a program that will automatically make and
update the signature for you! The program is available for
Windows at http://www.geocities.com/theTFZ/SETI/SETIsig.html
(requires VB6 runtime libraries)
In addition many people have begun marking their posts with a
short statement, indicating their opposition towards Olli (1.3.6)
It usually goes something like this:
+++++++ ONLY USE OFFICIAL SETI@HOME SOFTWARE +++++++
+++++++ DO NOT USE SETI@HOME PATCHES +++++++
It's inserted just before or in the .sig
1.3.5 Labeling posts
As there are many discussions about things not so relevant to
SETI in these newsgroups (a.s.s. and s.a.s.), there have been
developed many schemes, on how to label the posts.
I suggest labeling posts in this newsgroup as follows:
[sci] About science in general with no content about SETI@home
[meta] Discussions about discussing things
[comp] Discussions about computing with no content about S@H
[ot] Off-Topic: has nothing to do with SETI in any way.
[patch] Discussions about patching the SETI@home client program.
[brag] General bragging about your computer's speed, etc.
[join] Invitations to join groups.
[FAQ] Postings of Frequently Asked Questions or comments.
(no label) Discussions about the SETI@home project, client, etc.
Insert the [*] after any "Re:". Some newsreaders use the "Re:"
to display which posts are replies. Please leave a space after
the ":" in "Re: ". Not doing so confuses some newsreaders.
Don't make up new tags. Make the tag general and the text after
the tag specific. If you think that a new tag is needed, start
a discussion about it under [meta] (NOT under the proposed tag!).
If someone doesn't follow the rules. don't criticize. This is
100% optional. Posters are free to do as they choose. Limit
your efforts to gentle persuasion only.
The single most effective thing that you can do to promote this
idea is to change the subject line and write a new one before the
old one, so that the subject will be like this: Subject: <new
subject> Was:<Old subject>.
OFF topic:
1. BINARIES - a big NO.
2. ANY commercial advertising.
3. Number of WU's done. Use your .sig file to brag.
4. Personal Chit-chat use [ot].
5. Posts about how old you are, what you do, etc. use [ot].
6. How fast your CPU is as compared to others. [brag]
1.3.6 Patching or cracking SETI@home
Olli:
When you hear a reference to Olli in the group, this is a
reference to a German who thought, that he could just do as he
wanted with the S@H program. What he did, essentially, was to
decompile the code of the program, and release a new version of
the program, unauthorized. This led to a big discussion in the
alt.sci.seti NG, where he was eventually boo'ed out, because of
his actions.
Microsoft cracking the code:
Microsoft wrote their own version of SETI, highly optimized for
certain Windows hardware. They wanted to turn in the fastest WU
times, to prove how fast Windows is. The SETI people discovered
MS's cheating, and told them they must run the original SETI
software, and threatened to dissolve the MS team, and said they
would refuse results from any WU processed on a non-official SETI
client. SETI had obvious concerns, that their algorithms might be
programmed incorrectly.
SETI@home's (Eric J. Korpela) response to a post concerning the
programming variables of the patch:
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
From: kor...@islay.ssl.berkeley.edu (Eric J. Korpela)
Newsgroups: alt.sci.seti
Subject: Re: seti patch bashing or the truth is...
Date: 28 Oct 1999 16:06:33 GMT
Organization: Cal Berkeley-- Space Sciences Lab
Lines: 147
Message-ID: <7v9sa9$ror$1...@agate-ether.berkeley.edu>
In article <7v7b3b$70l$1...@nnrp1.deja.com>,
******* ******* <*********-deja.com> wrote:
>In article <7v4phg$1vv$1...@agate-ether.berkeley.edu>,
> kor...@islay.ssl.berkeley.edu (Eric J. Korpela) wrote:
><snip>
>> In other words, if I don't get the same results using a hammer
>> on a screw that I do using a screwdriver then something is
>> wrong with the screwdriver?
>>
>> Olli has provided a means by which anyone with a compiler can
>> replace the FFT routine with one that produces random
>> numbers. Your conclusion would be that if the random number
>> generator gets different results than the FFT does, something
>> is wrong with the FFT.
><snip>
>
>Excuse me, but I have to ask. Are you really a programmer?
Well, that depends upon how you define "programmer." I'm
actually a scientist. My profession requires me to be a
capable optical engineer, electrical engineer, mechanical
engineer, and programmer. Above all it requires me to be a
proficient systems engineer. And one of the things Olli's patch
is missing is any consideration of the system aspects.
Let me tell you some of the system aspects. The bottlenecks in
SETI@home are currently (in order of size):
1. The speed of the user database machine. This limits the
number of connections we are able to handle per second. Sun
has decided to give use another two Enterprise machine with
two CPUs each, so this bottleneck will be going away soon.
It will probably be another month before these machines
arrive. The effect of a faster client on this bottleneck
would be a higher rate of rejected connections and a lower
system efficiency.
-1 for Olli.
2. The rate at which work units can be split. The arrival of
the two Enterprize machines will allow two more splitters to
be used, to this bottleneck will go away, too. The effect
of a faster client on this is nothing.
+0 for Olli.
3. The fraction of time the S@H recorder is operating at
Arecibo. We have no control over this parameter. When very
RFI sensitive
experiments are carried out at Arecibo, the SETI@home data
recorder is shut off to prevent interference. For the first
10 months of the year(1999 ed.), this fraction was about 1/2.
The effect of a faster client on this is nothing.
+0 for Olli.
4. The speed of the data recorder at Arecibo. Again, there's
nothing to be done here but add another recorder working at
different frequecies. That may be done at some point. The
effect of a faster client on this is nothing.
+0 for Olli.
5. The speed of the SETI@home client. Note that this appears
BELOW the previous four. Because of this, improving it
doesn't improve systems efficiency. Let's assume #1 and #2
are solved and that we release a client that does the work
in 1/4 the time. What is the response of the system to this
optimization? Because data isn't coming in any faster, any
speed increase in the client just increases the number of
times a work unit is processes. Increase the speed of the
client by a factor of 4 and you've increased the processing
redundancy by a factor of 4. So there's no net processing
efficiency increase. You've still got to store all the
incoming results, so you're actually reducing efficiency
slightly. So this is actually a negative for Olli.
-1 for Olli
The obvious conclusion is that Olli's patch, while increasing
the efficiency of a specific instance of the SETI@home client
decreases the system efficiency. But Olli doesn't care about
that.
There is a way around number 5, that is to add more
processing capability to the client. This is what we were
planning to do in the next release before we were so rudely
interrupted.
>I have seen
>several posts in these threads that appear to be from you
>that suggest you don't understand some things that I have
>always considered pretty basic.
I understand a bit more than you appear to.
>If you are a scientist in general and a computer scientist
>in particular, then I would think your primary concern would
>be in getting the best tools to resolve the research
>question in hand.
Sorry, you don't seem to understand science. Faster doesn't mean
better. And trustworthy is far better than faster. (Did you
notice Olli's message stating that he would add malicious code to
an employer's system in case he was fired "for the wrong reason."
I would guess he considers adding malicious code to an employer's
system "the wrong reason.")
In addition, a scientist doesn't add unnecessary variables to an
experiment. A different FFT algorithm for every platform is an
unnecessary variable.
>If you are not competent to assess his optimizations, that is
>merely an unfortunate technical gap that you can choose to
>address.
Pardon me if I take offense to your insults. I never said we
weren't competent to assess his optimizations. We don't have the
time to deal with the patches every yahoo with a debugger and a
compiler throws our way.
>I'm not suggesting you should give up control of your project.
Look at the SETI@home web page... I never had control of the
project.
>But neither is it the case that your enthusiasm is holy and
>other people's enthusiasm is cursed.
When his enthusiasm leads him to give everyone with a compiler
control of the science code, I'd call that cursed.
>1) Optimize your inner loops better.
We're working on it.
>2) Learn about code signing.
I think you're misunderstanding code signing. Code signing is
used as proof to the user of who created the code, not as a
means of preventing the user from tampering with the code.
Even Olli will tell you that even if the code checks it's
signature before running, that check ends in one or more
conditional jumps that are easily removed. The other option is
encrypting the entire executable, but even then you need to
provide in the executable a decryption routine and a key.
If you give the hacker those, you've given him the equivalent
of a decrypted executable. Actually you don't even need those
as the code decrypts itself to memory where it is easily
acessible.
>3) Consider modularizing your client.
No way.
>The last suggestion is most complicated, but if implemented
>properly, it would actually allow you
and everyone else on the planet
> to reprogram the clients as your needs
or their desires
>required without even distributing new versions.
Eric
--
Eric Korpela | An object at rest can never be
kor...@ssl.berkeley.edu | stopped.
http://sag-www.ssl.berkeley.edu/~korpela
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
And another post made by Matt Lebofsky of the S@H team,
concerning the ethics of patching:
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
From: ma...@albert.ssl.berkeley.edu (Matt Lebofsky)
Newsgroups: alt.sci.seti,sci.astro.seti
Subject: Re: Found Olli's Patch !
Date: 13 Dec 1999 17:31:52 GMT
Organization: Space Sciences Laboratory
Lines: 61
Message-ID: <833ai8$pd1$1...@agate.berkeley.edu>
Despite all the reasons below being completely valid, I'll give
you another one:
SETI@home doesn't allow unauthorized access to our data server.
Patched clients which look like real clients that contact our
server could, in theory, do any number of things that cause
harm. In short, it's a form of hacking.
Of course people believe the patch is safe and accurate. We here
at SETI@home don't know that, nor has the patch creator been
willing to prove it. The author of this patch even goes as far
as to completely *refuse* to identify his patch so our server
can recognize it as a patched client. To me, this is an obvious
affront.
In short, we can't tell if results are from patched clients or
not. The patch could easily be modified to fix this. It hasn't
been. Case closed.
Too bad I didn't know about Olli before I went touring in
Germany for five weeks this past summer. Hopefully I'll be back
again in fall of '00. If anybody knows where he lives/works/hides
let me know. I'd like to discuss the patch in person.
This is my first AND last message on the matter.
- Matt - SETI@home
In article <8327ak$ln4$1...@nnrp1.deja.com>,
Daviddth <davi...@my-deja.com> wrote:
>In article <3853bf65...@uutiset.nic.fi>,
> j.k.@gotmail.com (Jan Knutar) wrote:
>
>>>Found Olli's Patch !
>
>> DO NOT USE ILLEGAL PATCHES!
>
>Please tell us why not. In your answer, please do not include:
>
>Morality - Your morals may not be others
>
>Scientific results - The patch has proven itself accurate here
>in multiple tests
>
>Scientific "purity" - if the patch is accurate, but quicker,
> then what is the problem in using it.
>
>Dislike of Olli - There are plenty of people that do not like
> others, but respect their work. Learn to live with this anger.
>
>I await your reply.
>
>--
>David
>http://setiweb.org
>http://www.lisp.com.au/~daviddth/king/
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
Please note that these Usenet posts that have been quoted here
aren't meant to represent the absolute truth, they are meant to
give you SETI@home's opinion on patching, and to give you some
interesting reading.
There's an excellent page, dedicated to this very subject here:
http://home.hccnet.nl/a.alfred/p-free-p1pfp.html
1.3.7 What is a 'vcard' and why do people tell me not to use them?
A vcard is only readable by a newsreader capable of rendering
html. For this reason and because vcards are 'attached' to the
Usenet posting, they are not recommended. They are also very
annoying in that they very rarely change. This is analogous to
giving everyone you talk to your business card each and every
time you talk with them... after a while, we have enough of your
cards, and we really don't want any more. A signature is the
preferred method to communicate pertinent personal information.
Please see the section on 'signatures' or '.sig' for more
information.
1.3.8 What is PST and PDT?
They're timezones ;)
PST - Pacific Standard Time (GMT - 8)
PDT - Pacific Daylight Time (GMT - 7)
(GMT = Greenwich Mean Time)
1.4 What will happen if an extraterrestrial signal is detected?
A procedure has been agreed upon by SETI researchers around the
world. First, other SETI researchers will independently verify
the signal. If the signal is real and can't be explained by
man-made sources (satellites, reflections, etc.) then press
agencies and governments will be notified in a systematic way.
1.5 How is data collected from the telescope and transmitted to other
machines for analysis?
Data is recorded on high density tapes at the Arecibo telescope
in Puerto Rico, about one 35 Gbyte tape per day, then mailed to
Berkeley, then divided into 0.25 Mbyte chunks which get sent from
the SETI@home server over the internet to people around the world
to analyze. Arecibo does not have a high bandwidth internet
connection, so data must go by snail mail to Berkeley at first.
1.6 Are earth signals strong enough to be detected?
What sort of spectrum is currently being emitted by earth? Is
that signal visible say 10 or 50 light years away? If SETI were
on a planet say 10-50 light years from here and running this
project there, would it be able to detect earth's signal
(assuming it was looking in our direction)?
Earth is polluting space with radio and television signals that
might be detected by nearby advanced civilizations, but it would
be difficult for such a civilization to discover these signals if
they only have Earth's current level of technology (eg: if they
have an Arecibo like telescope and SETI@home like search).
Early TV shows like I Love Lucy and Ed Sullivan left the earth
about 40 years ago, so have gone out 40 light years, reaching
several thousand nearby stars. But these signals are relatively
weak and SETI@home is not likely to detect the equivalent of
Earth type TV transmitters, even on the nearest stars.
Earth's strongest transmitters might be somewhat easier to
detect, such as those emitted by military radars, or some radio
telescopes. The Arecibo telescope transmits very powerful signals
when it is used as a radar system to study planets, asteroids and
the ionosphere. These radar signals are powerful enough to be
detected 10,000 light years away by searches like SETI@home,
except for three big caveats:
a) The Arecibo transmissions are in a very tight beam (they are
not omnidirectional, like TV and military radar), so they only
cover a very small part of the sky at once (about a millionth
of the total sky). It's is unlikely another civilization will
be within one of these narrow beams.
b) The Arecibo transmitter's oldest signals left Earth about 30
years ago, so have only travelled 30 light years.
c) SETI@home is not searching the band of frequencies that the
Arecibo transmitters utilize (although the older SERENDIP III
program did survey one of those bands).
Also see section 1.2.3
1.7 What if my computer finds a signal -- how will I know?
You won't know, because your computer can't find a signal all by
itself. All it can find is bits of pattern that are worth further
investigation and correlation with other bits of pattern in other
work units. These will be flagged for the SETI@home staff to look
into, and when they've verified it by various methods with
scientific rigor, then they'll make the announcement. Don't worry
-- they'll give you co-credit.
1.8 How can I hear the signal?
The long answer: the data isn't sound, it's radio waves. You can
make up an arbitrary set of rules to "map" radio waves into
sound, but since you picked the rules, you really decide what
you're hearing, not the signal. As an analogy, imagine if you
wanted to make a picture of the melody of a song was. You could
decide (ala "Close Encounters") that a middle C turned into a
teal light, and the G above middle C turned into a red light.
Then any given melody becomes a set of colors. But when you're
done, the flashing lights you see tell you more about the
particular rules of mapping you made up, than they did about the
melody you started with.
The short answer: And even if you did, it'd just sound like white
noise. So turn on some speakers without any signal hooked up to
them, or tune your TV to a channel you don't get, and listen do
that. It's about the same thing.
The SETI@home team has decoded one, and it's located her:
http://setiathome.ssl.berkeley.edu/audio.html
Nothing but white noise.
Jan Knutar has made a program that maps the radio waves into
sound atleast one way, the result sounds, not surprisingly, like
noise. The program, which is available for both Linux and Windows
PC's, is downloadable at http://gamma.nic.fi/~jknutar/wu2wav/
1.9 Is there something in it for me?
No. Unless you count the chance to be the first one to make
contact with "The little green men"
1.10 Why doesn't SETI@home release the sources for the clients?
The sourcecode is not released because of both security and
scientifical reasons. If the code was available freely, anyone
could replace the core analyzing algorithm with some superfast
random number generator, for example. SETI@home is a scientifical
project, speed is not everything.
In the newsgroups, it has been hinted that the source will be
released when the project is over.
There are actually small pieces of the sourcecode available. Some
of the early clients were GPL'd, and you can find the sourcecode
for them on the Internet.
In a post to alt.sci.seti by Eric Heien, additional details of
the code used was given. Parts of the post below:
----clip-------------------------------------------------------
In the old versions, we used the four1 procedure for FFTs from
Numerical Recipes in C. You can see the exact code and
scientific and mathematical derivations for it in Numerical
Recipes in C. It's available at www.nr.com, or you can just jump
directly to
http://www.ulib.org/webRoot/Books/Numerical_Recipes/bookcpdf/c12-2.pdf
for the specific section.
The new FFT used in the beta versions (and soon version 3.0) is
the Ooura FFT library. You can get the code and benchmarks at
http://momonga.t.u-tokyo.ac.jp/~ooura/fft.html.
The new pulse finding code is called the Fast Folding Algorithm
(FFA). I'm sure there are several sources and papers for it on
the Internet, but the first I saw was
http://www.mpifr-bonn.mpg.de/staff/peter/ffades.en.html. This
particular analysis routine was written by us, but was based on
code that is publicly available (for example, from the above
link).
The triplet code is based on ideas developed by some NASA
scientists (I don't know their names offhand). The code was
actually entirely written here rather than based on other code,
but I'm sure there are papers and sample code available somewhere
out there for it.
----/clip------------------------------------------------------
See also SETI@home's official FAQ at
http://setiathome.ssl.berkeley.edu/faq.html#q1.9
2 Problems and questions concerning SETI@home
2.1 Speed improvements
2.1.1 What's the fastest computer to use for this project?
The computer you have. If it can run SETI@home, running it will
make more contribution than not running it. Even if you're only
doing half as many work units per month as the guy sitting next
to you, you're still doing more than you would if you weren't
doing them at all. If you want to see, what the fastest computer
possible is, check the following sections.
2.1.2 Can I make it run any faster?
1. Make sure you've got it set up to run continuously (a machine
with at least a 200 MHz clock is desirable; if an Intel
Platform then at least a BX or JX chip set is even better,
otherwise the performance of the system might be untolerable)
2. Each time SETI@home launches, it optimizes itself for the
current monitor color depth (256 colors, Thousands, Millions).
If you change the color depth while SETI@home is running,
it may slow things to a crawl.
3. It should run reasonably at any screen depth, but it will run
somewhat faster at lower screen depths (256 colors) than
higher ones. Screen resolution (800X600, 1024X732, etc.)
should have negligible effect on speed.
4. Make sure the graphical display window is never showing (run
the window collapsed on the task bar)
5. If running Windows 95/98/NT, make sure your screen saver is
set up for "Blank Screen". For some reason when screen saver
is set to "SETI@home" with "continuous run" enabled the client
seems to waste time fighting with itself. It can turn out a
work unit in half the time by doing this.
6. If running on WindowsNT 4 then try running the client at a
higherpriority. To do this you must do the following:
4a. do a 3 finger salute (ctrl-alt-del) to bring up the
"Windows NT Security" panel
4b. click the "task manager" tab
4c. locate the task called SETI@home
4d. right click on it
4e. clink task priority (low is the default)
4f. select either "medium" or "high" (but not "real time" or
you might need to reboot in order to regain control of
your machine)
On windows 9X, you can use the shareware program taskinfo.
Start Taskinfo, right click SETI@home, change priority,
realtime. Taskinfo can be found at http://www.iarsn.com).
Be warned, changing the priority to realtime is for the
purists only. It will give nearly all CPU time to the S@H
client, making your system unresponsive.
7. Use the text-client. Even though it's not as much fun as the
graphical, it does run faster. It will run on any win98/NT
system (NT calls have been ported to win98. The text client
will also run on Windows 95 if you upgrade Winsock to
version 2. You can find the upgrade at Microsoft's website
(http://www.microsoft.com).
2.1.3 REMOVED March 2000, was:
Can I run the SETI@home text-client on Win95?
Now appears in section 2.21.
2.1.4 Will SETI@home run faster with more RAM (e.g., 256 MB instead of
128 MB)?
SETI@home uses about 16 MB of RAM while it's running. Beyond a
certain point (typically 64MB, more if you run memory-itensive
applications) more RAM won't make it run faster.
2.2 REMOVED April 2000, was:
I'm using a proxy server, and I can't connect - what do I do?
Was removed because the SETI@Home client's hugely improved proxy
support. Consult 2.20 for help on the CL client.
2.3 I had a work unit that got returned after only 5 minutes. What's
wrong?
The SETI@home program found enough noise that it determined the
packet was messed up with it. It's like if you're trying to hear
an egg being dropped to the ground on the other end of a football
field, and someone blares a megaphone in your ear. No point in
continuing to listen for the egg. You wont get credit if it took
under 10 minutes to complete the workunit. This is to eliminate
results from the buggy mac clients that finished all workunits in
no-time.
There are a few examples of excessive noise in workunits at the
SETI@home website:
http://setiathome.ssl.berkeley.edu/rfi/index.html
2.4 I heard I was getting the same work unit as everyone else. Is the
program wasting my time?
Nope, because the only time you're giving it is time your
computer would have wasted anyway. Yes, early in the program
there were times when the same work units went out over and over,
due to overloading of the SETI@home servers that were supposed to
be making new ones to send out. (They didn't expect half a
million people to sign up, and they don't have enough staff or
computing power to keep up with it.)
And since then, the same work units are still sent out to several
people, for various reasons (for instance, more than half the
people who signed up have never returned their work units, and
probably dropped out) But new work units are being sent out too,
so just leave your SETI@home program working and it'll take care
of the details.
Note:
If workunits are sent out multiple times, they can be
doublechecked by SETI@home.
2.5 My computer wanted to upload to the SETI@home server but said it
couldn't connect or reported error 10065. Are they still there?
Yes But they're sometimes swamped with traffic. Just try again
later. Error 10065 is a winsock error - means the same.
The page http://setiathome.ssl.berkeley.edu/sstatus.html is
auto-generated and tells you whether SETI@home's data server is
running or not.
Sometimes the error might be between your ISP and Berkeley,
sometimes at your ISP. Most operating systems have tools to help
you determine where the error is. If you want to try to find out
what's wrong when you can't connect, try using the 'traceroute'
command (in Windows: tracert). You bring up a command prompt and
type in the command followed by the site you wish to connect to,
in this case shserver2.ssl.berkeley.edu (for the S@H server).
If you get "Cannot resolve" or similar error, then your computer
was unable to translate the address into an IP number. The cause
of that could be that your ISP's DNS server is down or not
working properly.
2.6 What if someone fakes a result to make it seem like they found a
signal?
The SETI@home staff will be reviewing the actual data that
produced the result, and if they don't find the same results,
they will discard the fake. Besides, while it's not impossible,
it might be harder than you think to fake a result file.
Since some workunits are sent out more than once, SETI@home can
detect errors by comparing the results. During the time of the
project, the sky will be scanned several times. It's very
unlikely that a cheater would get a workunit from the same
location in the sky more than once.
2.7 SETI@home keeps getting a 'Bad Header' error. What can I do?
1st Possibility
First close the SETI@home client. Open the work_unit.sah file in
the SETI@home directory and delete all the lines that appear
before the 'type=work unit' line, but do not delete this line.
Save the work unit file then restart the SETI@home client.
2nd Possibility
If you installed the client software from the FreeBSD ports
collection, install the highest numbered version available from:
http://setiathome.ssl.berkeley.edu/unix.html.
If you then get an error with ld.so, go to /usr/src/lib/compat
and make, make install compat22.
3rd Possibility
There's a problem at Berkeley with the servers. Try again later.
If you look closely at the GUI client while it returns, you might
or not might see "all data sent". If that happens, then the
results were sent back allright. If you did not see that, then
you can try to use the following to get a new workunit.
Close SETI@home. Make sure it is closed, right click the icon in
the systray and exit the client.
Go to the SETI@home folder. Move the files "outfile.sah" and
"result.sah" to a temporary folder.
Start SETI@home. The client should get a new workunit. If this
does not work and you get bad header again, then the problem is
probably one of the other possibilities.
When you wan't to try to send the results back again for the
workunit that got the bad header error, then:
Exit the client.
Move the text files "work_unit.sah", "result_header.sah",
"outfile.sah" and "state.sah" to another temporary folder. Move
back the files from the first temporary folders.
Start notepad, select save as, move to the SETI@home folder
(usually c:\program files\SETI@home), save the file as
"stop_after_send.txt". Start SETI@home. The results should be
sent now. Exit SETI@home and delete the files "result.sah",
"outfile.sah" and "stop_after_send.txt". You can now move back
the files you moved to the temporary folder.
If you are unsure which files should be moved away, then move all
the text files.
2.8 Suddenly, without warning my system crashes - what should I do?
Make sure you have the latest video (try first) and peripheral
drivers for your system. This is often the cause of lockups and
crashes, at least for Win9x.
If your machine suddenly reboots itself or you get a blue
screen, then it could be that the processor or some other
part of your computer is overheating. Check that you have
enough cooling for your processor. Most new computers have
built in sensors for measuring the temperature of various
parts in your computer.
2.9 I can't see the new WUs I've processed in the status area. Have
they been registered at SETI@home?
Probably. Sometimes you first get Stats at next WU. You can also
check with the personal stats available at the SETI@home website
(http://setiathome.ssl.berkeley.edu). Those stats are currently
updated almost immediately so they should be the most accurate.
2.10 I want to run the text-client as a service in NT - how do I do
that?
Method nr.1
You need to be administrator to do it. The easiest way is if you
have IE4 or IE5 and Task Scheduler. It comes with IE5 and is an
option in IE4. Set up a task to run when your PC boots to launch
it. When you boot your PC, let it sit at the logon screen for 30
seconds or so to verify the Task Scheduler service has started
and it has launched the task. Now it will run in the background
and the only way to stop it is with kill.exe from the NT resource
kit.
If you don't have IE4/5, then use the Schedule Service built into
NT, but you have to be an administrator to do this. Make sure it
is set to run and log on as you. Then schedule it to run in about
2 minutes using the AT command. Type AT /? from a command prompt
for help. You do NOT want to use the /interactive switch. It will
then run in the background. This is easier to use if you have the
Resource Kit as well because you could use the SOON.EXE command
in a batch file in your startup group.
*Note: You must run the seti client manually the first time to
configure it. Afterwards you may allow it to start automatically.
Method nr.2 provided by Peter Yackel:
You need two files from the NT resource kit: srvany.exe and
instsrv.exe. You'll use these files to install SETI as a service.
Here's the procedure:
Copy srvany.exe to the SETI@home directory. (I'll use c:\seti
in this example. I also assume the seti executable to have
been renamed to seti.exe)
Copy instsrv.exe to the winnt directory.
Click Start, Run, and type "cmd" to open a DOS window.
Type: instsrv SETI c:\seti\srvany.exe
Type "exit" to return to NT
Click Start, Run, and type "regedit"
Go to the following registry key:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services
Expand it by clicking the +
Highlight SETI, then right click and select New, Key
Type: Parameters
Highlight Parameters, right click and add the following New
String Values Names:
Application
AppDirectory
Now double-click the Application value name you just created
and enter the following Value Data:
c:\seti\seti.exe
Do the same for AppDirectory but enter c:\seti for the Value
Data.
Exit the Registry Editor.
Go to Control Panel and double-click Services.
Go to SETI and double-click on it.
Set the startup to Automatic. Click OK. This will cause the
service to automatically start at the next boot.
To start the service immediately, highlight Seti and click Start.
*Note: You must run the seti client once manually to configure
it. Afterwards you may allow it to start through services.
Method nr. 3
Use a third party utility. (See 4.2.11)
2.11 Can I run the client invisibly on Win95/98?
Nr 1.
This has been found to work in the past but has failed with some
Client/OS configurations.
Before you do this, please bear in mind that you have to have
permission to run SETI@home on the computer. Do not use this
method to run SETI@home on other's computers. It is a violation
of the license agreement to run SETI@home on computers you do not
have permission to run SETI@home on.
Install and setup the client in the normal way. After you have
completed the setup of the client making sure that it is running
all of the time and not just in screen saver mode. Run regedit
and search for SETI@home It will probably be in there more than
once, so the one you are looking for looks like this:
seticlient C:\Program Files\SETI@home\SE...@home.exe -min
This string value will be in the key Run. Move it to the
RunServices Key and remove it from the Run key Restart and the
next time it comes up it will run even before you log in with no
icon visible.
Nr 2.
This method works well with the Command Line version and Win 98.
SetiLog is a third party utility that creates and maintains a
file of completed work units. It can be used in conjunction with
SetiWatch to monitor the progress of the client.
Use SetiLog to start the client using the command line switch /H
This will start the client hidden.
Once the client starts and runs with this method create a
shortcut to SetiLog in the Start Up folder.
The client should now start on boot and run invisibly. In order
to shut down the client you will have to ctrl-alt-del and select
SetiLog.
You can also create a registry entry to start the client
automatically. I had to use the Shortcut to start SetiLog instead
of the path directly to SetiLog. The registry entry should go in
HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Run
and can be named anything, but should contain the value of the
path to the Shortcut to SetiLog, not the path to SetiLog.
Other third party utilities may also have options to start the
clients invisibly.
2.12 Sometimes the size of the workunit.txt file differs in size.
Sometimes it's 340, sometimes 341, and yet other times 351. Is
there something wrong?
The difference between 340k and 341k is most likely a difference
in the number of telescope position strings reported in the
header. The 351k is an benign bug in the portion of the splitter
which determines where the work unit ends. It basically tags on
an extra 10.67k (IIRC) of data, that the SETI@home client
ignores. It has to do with the timing relationship between
position information from the telescope and the start of a block
on the tape. All of the data in the work unit is still OK.
2.13 I don't have a permanent Internet connection, and have to pay for
all my phone calls and net usage. Can I run SETI@home without
going bankrupt?
SETI@home will only connect to the Internet when you want it to.
The GUI (slow, pretty graphics) clients have an option under
"Preferences" to "Ask me before connecting to the Internet", and
the CL (fast, no graphics) clients have a switch
"-stop_after_process". In each case, this will prevent an
internet connection being made until you're ready, and means the
clients can be left safely unattended. When you are ready to
connect to the Internet (say, for a normal browsing, usenet or
mail session) you can make the client send results and retrieve a
new Work Unit. In the GUI case, it will ask you to make a
connection. For the CL client, stop the process, then restart it
without the "-stop" switch, and it will connect automatically.
Depending on the load at the Berkeley servers, within 5 minutes
you will have sent your results and received new work to do.
Programs have been developed, that will buffer the WU's for you,
see sections 3 and 4.
If you are still concerned that the clients will connect when you
don't expect, make sure your system is configured *not* to
connect "on demand" without prompting you for confirmation (a
good idea anyway if you are concerned about unwanted
connections), switch off your (external) modem, or pull the
telephone plug from the socket (internal modem).
2.14 I already run the distributed.net RC5-64 client. Can I run
SETI@home as well, or do I have to choose which project to
support?
Both clients can run simultaneously. In screen-saver mode the
SETI@home client seems to take priority. For those that don't
allow the clients to connect automatically, the advantage is that
the RC5-64 client can download multiple blocks to work on as
opposed to SETI@home's single work unit. This means that while
the SETI@home client waits for you to connect next, the RC5-64
client can continue working on its own tasks.
2.15 What happened to the gaussian information display in the new Mac
and Windows clients? The client is finding gaussians with lower
fits, do the 2.x clients find more aliens or something?
The 2.x GUI clients introduced new gaussian curve fitting
graphics. To not make the new display blank all the time,
SETI@home decided that the client should show all gaussians
found, no matter how weak they were. The gaussian power is lower
than in 1.x, too low to be reported back to SETI@home.
The 3.x clients alternate between gaussians, pulses and triplets.
2.16 Can I run SETI@home 24/7 if I don't have a permanent Internet
connection?
There are a couple of ways of running multiple instances of
SETI@home such that, if one instance finishes its Work Unit,
another will take over, so making sure that your system is
working flat out most of the time.
Under Windows 98/NT/2000 Command Line (non-graphic), and various
*ix flavours, multiple clients in different directories may be
"chained" to run consecutively by specifying the
"-stop_after_process" switch for each. When one Work Unit is
finished, that client will stop and another will take over. When
all clients are done, an Internet connection can be made to send
all results and receive new Work Units. The process may then be
repeated.
Also, there's the possibility of using one of the WU caching
add-ons, such as SETIBuf or SETI Manager for Windows, SETI Unit
Manager for Mac, or Hiram Clawson's RunCache & FetchCache for
Unix like systems (for example Linux). See section 3 and 4.
Running several clients simultaneously on a single-processor
machine, or running more clients simultanesouly than the number
of CPU's in your machine is not recomended, running them after
eachother will give you better performance.
2.17 Is this bad for my processor, or my harddrive?
Most technicians agree that turning the computer on and off is
worse for the lifespan of the parts inside, than leaving it
running. Of course most computers are obsolete long before the
processor gives out, even when it's being run all the time. Your
processor might get a little warmer, but not dangerously so,
except if it has already been overclocked.
If you are worried, make sure that your computer has sufficient
cooling.
2.18 Does it use up a lot of electricity? Is this costing me money, or
doing damage to the environment?
Many SETI@home users would have been leaving the computer on
anyway. For some computers, starting it up uses as much
electricity as running it for a while anyway. With the monitor
turned off, the average computer uses less electricity than a box
fan, more along the same lines as a bright light bulb. Even less
for laptops. Odds are running SETI@home all night while you are
sleeping costs you a few pennies a day at most, and probably
less. The millions of people whose computers are on to run this,
who wouldn't have had them on otherwise, are adding a tiny
fraction of a percent to the energy usage of the world -- not
enough to have a measurable environmental impact. But turn that
monitor off, when you're not using it. It probably uses more
electricity than the rest of the system put together.
The harddrive is not likely to take any damage either, as the S@H
software only accesses the HD at most every 60 seconds.
2.19 How can I keep appraised of what's going on lately?
Read the SETI@home web site at:
http://setiathome.ssl.berkeley.edu/ and especially
http://setiathome.ssl.berkeley.edu/tech_news.html where news
about the software is posted. Also you should check out the links
at the bottom.
Alt.sci.seti and sci.astro.seti are also good newsgroups to read
since members from the SETI@home team posts there quite often.
Look for posts by Matt Lebofsky and Eric J. Korpela, they're part
of the SETI@home 'staff'.
Also, look for posts by Hiram Clawson
Hiram manages the UNIX porters and maintains the page containing
the command line clients,
http://setiathome.ssl.berkeley.edu/unix.html
"I am merely a volunteer at the edge of the project, not part
of the SETI core team, although I do have a relationship with
the team. I was also one of the members/programmers of the first
SERENDIP team in 1974/75 at Berkeley."
2.20 General CL client issues
2.20.1 What is the CL client?
The Command Line client is a version of the SETI@home client that
completely lacks pretty graphics and looks just like a DOS
program. Because of that, it is also faster.
2.20.2 What CL options are there?
Just start the client with a bogus option such as '-help' and it
will tell you.
2.20.3 HELP, it stops at baseline smoothing!!!
Version 2.4 of the commandline client is 'quiet' during
processing. It still processes the data, but it doesn't tell you.
You can use some of the add-on programs (section 3) to check up
on the client, or you can give the -verbose option to the client
2.20.4 How can I check up on the client to see how it's doing, if it has
found any signals, etc?
You can either examine the .sah files, or use one of the third
party add on programs, found later in this FAQ.
2.20.5 How do I tell the CL client to use a proxy?
You start the client with the command line option
-proxy proxyserver:port
For example
seti.exe -proxy my.isp.proxyserver.net:8080
where seti.exe is the name of the CL client.
2.20.6 I just found a bug in the -stop_after_ switches!!!
No you didn't. Here's what they do:
-stop_after_process
The client will process the current workunit, then exit. If the
client has already processed a workunit, but you haven't sent it
yet, then the client will send the results, fetch a workunit and
process that workunit and _then_ stop.
work_unit.sah present: Process workunit, delete
work_unit.sah.
work_unit.sah not present: Get new workunit, process workunit,
delete work_unit.sah
-stop_after_xfer
The client will send in the results AND retrieve a new workunit.
If work_unit.sah is present, it will do nothing.
stop_after_send.txt
If this file is present in the current directory, the client will
SEND ONLY when it connects to the server. Delete the file if you
want to get a new workunit, and create the file if you want the
client to not get another workunit the next time it connects.
The stop_after_send.txt file is empty. Works with the GUI client
too.
The presence of this file will not prevent the client from
processing a workunit, if there is one in the directory.
Other:
work_unit.sah and result.sah will never exist in the same
directory at the same time. If you're making some scripts for
managing the clients, you can use this fact for retrieving info
about the current status of the processing in a directory.
2.21 Running the CL client on Windows
This sections purpose is to give a short guide on how to run the
command line client on windows.
2.21.1 What client should I download for Windows 95/98/2000/NT?
There are currently two clients available:
i386-winnt-cmdline.exe for Intel processors
alpha-winnt-cmdline.exe for Alpha processors
The clients are at http://setiathome.ssl.berkeley.edu/unix.html
Despite the name 'winnt', the clients will run on Windows 95, 98,
2000 and Windows NT. For Windows 95, you have to download the
Winsock 2 upgrade. You can get it from Microsoft:
http://www.microsoft.com/windows/downloads/bin/W95ws2setup.exe
According to posts by hir...@sco.COM to the seti newsgroups, the
i386 client detects the processor it runs on and optimizes to it.
Since some processors do not work well with that, an i486 version
was compiled, optimized for 486 processors. Some Cyrix and AMD
processors have troubble with the i386 client.
Rumors tell it that there's very little performance difference
between the two clients. Some say the i486 is faster, some say
the i386.
If the i386 client crashes your computer, try the i486 instead.
2.21.2 How do I start it?
Every client should have its own directory, you can't run the CL
client in the same directory as the GUI client.
Starting the CL client can be as simple as double-clicking on the
exe file from Windows Explorer, a window will then pop up and ask
you a few questions if it's the first time you run it.
To pass options to it, you normally start an MSDOS window and
start the program from there. If you can't find DOS from the
start menu, click start, run and type in command.com. That should
bring up a CLI (command line interface). For ease of use, I
suggest you rename the setiathome executable to something that's
short and easy to type, like 'seti.exe' (without the quotes).
Once you have to MS-DOS window in front of you, navigate your way
through the directory structure by using the CD command.
In this example, we'll pretend that the client is installed in
F:\seti. To get there, you first type f: and press enter. Don't
forget the colon. Then, it might be that the screen looks like
this:
C:\Windows\>f:
F:\something\>
Now you type cd.. (yes, two points at the end).
This will bring you to
F:\>
Now you type cd seti
and should now be in the seti directory.
If you've renamed the executable to seti.exe you'd type something
like this
seti -verbose -stop_after_process
2.21.3 How do I stop it?
You press two keys simultanesouly, the keys are CTRL and C.
So, just press CTRL-C and the client will stop. As easy as that!
Some of the add on programs can handle the starting and stopping
of the CL client for you.
2.21.4 How do I make Windows 2000 autoconnect?
Solution taken from a post by "Jedi"
Goto Network and Dial Up Connections
Under Advanced - Dial Up Preferences - Auto Dial
Enable Current Locations
Uncheck Always ask before
Uncheck Disable For Current Session
Goto Services
Set Remote Access AutoConnection Manager to Automatic
Set Remote Access Connection Manager to Automatic
As is the norm with Windows, you must now reboot for this to take
effect. For immediate results, you can manually start these
services.
2.21.5 Do I have to uninstall the screen saver version if I use the CL
version?
No, but running both at the same time will give you worse WU
throuput per day than running only one client at a time. An
exception to this is computers that have more than one CPU's,
then running one client for each processor is the most optimal.
The CL client can not continue on a work unit that the GUI
version has started on but not finished, likewise the GUI client
can not continue on a work unit that the CL client has started to
process. If you try to make them do that, they'll start over from
0 %. So, don't have the CL client and GUI client installed in the
same directory.
2.22 A short guide for the Linux newbie trying to run the SETI@home
client
Many people seem to try a Linux distribution these days,
including running SETI@home on it. This section answers some of
the most common questions asked in the SETI newsgroups.
2.22.1 What client should I download?
If you're using a newer Linux distribution, you'll probably want
to download one of the gnulibc2.1 clients. Some older
distributions might not work with those two clients, in that
case, download one of the gnulibc1-static clients instead.
If you've got a Pentium Pro, Pentium II, Pentium III, Celeron or
AMD Athlon, the i686 clients will work on your machine (and will
probably run slightly faster too.)
All processors older than those mentioned above are usually not
686 processors, a common mistake is to think the the AMD K6/K7 is
a 686 processor. It is not, use the i386 client instead.
2.22.2 How do I uncompress the .tar file?
You type tar -xvf filename.tar
Where filename.tar is the name of the 'tarball' you want to
uncompress. Uncompressing the file creates a directory of the
same name. Within that directory will be the executable also
with the same name. Many rename the executable to setiathome or
simply seti.
2.22.3 How do I start it?
You navigate to the directory where you've got the setiathome
client, then type './filename' (without the quotes, where filename
is the name of the executable). The ./ is important so that linux
knows you want to run something in the current directory.
2.22.4 How do I stop it?
If you've got it running in a console or Xterm, just press
CTRL-C.
If you started the client in the background you should use:
kill `cat /path/to/seti/pid.sah`
You can also follow the instructions in the readme file that
comes with the client. The instructions in it are quite easy to
follow and it even works too!
Some third party add ons can handle the starting and stopping of
the client for you, and are highly recommended.
2.22.5 How can I run it in the background rather than in a window?
Start the client with the options to direct output to /dev/null
and use the switch & to release the command window. i.e.
./setiathome > /dev/null &
2.22.6 How can I have it automatically restart if it dies?
You can ensure that it is always running by creating a cron job
to start it. If it is already running then the cron job will not
start an additional instance. It will only start the client if
it is no longer running. My cron entry is in /etc/crontab and
looks like this:
30 * * * * mark cd /home/mark/Seti@home; ./setiathome > /dev/null &
The user (mark in this example) must be allowed to use cron processes
by an entry in the /etc/cron.allow file.
2.22.7 What is 'nice' and how do I set it?
Mama said "Always be nice."
Nice is the unix term for priority. A process that is nice will use
the available processor but will give way to other processes that
are not so nice. The range of nice is -20 to 19 (depending on
your distribution. Negative numbers have the higher priority and
positive numbers have a lower priority. You can set the niceness
of the client with the -nice parameter.
./setiathome -nice 19
2.23 Why does the client timeout before windows has dialed my ISP?
The SETI@home client does not set a timeout delay, it's windows
that does this. Unfortunately, Microsoft has set this value too
low so that you get a timeout before your modem has dialed your
ISP and logged on.
You can hope that the next patch from Microsoft will fix this, or
use one of the excellent WU buffering/caching programs available,
some mentioned in this FAQ and on SETI@Home's link page.
http://setiathome.ssl.berkeley.edu/links.html
2.24 I just got a gaussian with a score of 0.30 and power 1.2, is this
good?
Considering that the SETI@home client does not record and report
back gaussians with a power as low as that, the answer should be
no.
There has been some discussion in the newsgroup about what is an
interesting gaussian and what is not. First of all, the client
shows you on screen the best gaussian found so far based solely
on the score it has, this value is also stored in state.sah and
used by many add-ons. Gaussian score is calculated as power/fit
and spike score as log10(power/40).
The score is not good enough for determining if the gaussian is
interesting, since its power might be way too low. The gaussians
that the SETI@home client sends back to Berkeley are much more
interesting, these are recorded in outfile.sah and result.sah.
To see if the client has recorded a gaussian, open outfile.sah
or result.sah and check for lines beginning with "gaussian:"
(without the quotes). If such lines are present, the SETI@Home
client has recorded the gaussian and will report it back to
Berkeley. If you use windows, then you might have to use
'Wordpad' or 'quickview' instead of Notepad, since Notepad has
some problems with the format used in .sah files.
So, what are the requirements for a gaussian to be recorded and
reported by the client? First of all, the fit has to be lower
than 8.8 (10 in versions 2.XX and below), this is always the same.
The power is a bit more tricky, here's what Eric J. Korpela said:
---------------clip---------------
>If thats the case, why is the top gaussians page full of entries
>with powers less than 3.2? Are we talking different units of
>power?
Yes, I had forgotten that our threshold is 3.2 in units of what
we call "true mean power" and is integrated over the gaussian,
whereas the reported power is the peak power or the gaussian. So
the conversion between true mean power units and the reported
units depends upon the width of the gaussian.
--------------clip----------------
Roelof Engelbrecht seems to have found out in his latest
version of SETISpy how the SETI@home client determines whether it
should send back a gaussian or not. The magic formula is:
(peak / mean) > 3.2
In state.sah, that is bg_power / bg_true_mean. In result.sah and
outfile.sah, it's peak / mean. Note that mean=7.283651e-01 is
0.7283651, not 7.283651. The fit is called bg_chisq in state.sah.
The criteria is also mentioned in SETI@home's science paper:
http://setiathome.ssl.berkeley.edu/sciencepaper.html
It says "In order to confuse and obfuscate the public, we
utilize substantial technical jargon", luckily Roelof
Engelbrecht has translated this for us mortals in his excellent
SETISpy add-on. See section 3.3.
2.25 I looked at my stats of returned units at SETI@home's webpage,
the top gaussian I got for the last unit is missing, why?
The top-gaussian is not the same as what is returned to
SETI@home, see 2.24. This goes for spikes too.
There's an add-on, called SETI Monitor, that provides information
on all returned gaussians and spikes. See section 3.8.
2.26 My email address is about to change, what do I do?
Go to the SETI@home web page. There's a tool there that allows
you to change your email address. You'll need access to your old
email address as you need a password to change the email address.
When you've changed the email address, you need to tell the
SETI@home client to use the new one. With the GUI clients, you
can do this from the settings menu. Make sure you don't create a
new account, but log in to an existing one. The commandline
clients have a switch for this, -login.
If you have already created a new account with your new email
address instead of changing it, you can't transfer the credits
from your old email address to your new one.
2.27 Re-ordered to 2.30 (10-5-2000)
Was: What is an interesting pulse/triplet?
2.28 Why does the new client (3.X) take more time to complete a WU?
Because more science is done. The client now searches a wider
range of drift rates, from -50Hz to +50Hz. The client also
searches for pulsed signals.
2.29 What are pulses and triplets?
In the contexts of SETI@home, they are spikes that are repeated
many times. Triplet means that there are three evenly spaced
spikes. The triplet detection takes very little time to do, but
it can only detect strong signals.
The pulse finding algorithm can find very weak pulses, the more
pulses the better the algorithm can find them. The pulse seeking,
however, takes alot more time to do.
2.30 What is an interesting pulse/triplet? (re-ordered: was 2.27)
For Pulses, the threshold is at a score of 1.0. A score of 1.04
means that the pulse is twice as unlikely to be caused by random
noise.
For triplets, the threshold is at a score of 7.75.
2.31 Why do certain WU's take longer to process?
The Arecibo radio observatory can track stars to a certain
degree, even if it's basically a big hole in the ground. If the
telescope is moving more slowly, you get more data from a smaller
area of the sky, simply more time on one target. This makes it
possible for the pulse finding algorithm to use a larger chunk of
the WU data at a time, making it more sensitive to possible
pulses received.
2.32 Why are the most pulse searches done at a chirp rate of 0?
Shorter pulses have a larger bandwidth, and therefore less
affected by doppler shift. This is why you don't have to search
for pulses at an FFT length of 128 and chirp of 2.3Hz, for
example.
2.33 There's something strange with the power reported on pulses in
outfile.sah vs. state.sah?
Yes, the power is measured differently, here's a post by Eric
Korpela of SETI@home explaining it:
-------clip-----
This part is actually by design. The internal value of the pulse
power if the peak power measured from the zero point. The
reported value is the peak power measured from the mean power.
Since they are both normalized to the mean power, the reported
power should be always 1.0000 less than the value in the state
file. Sorry for the confusion.
-------/clip----
2.34 Is the first half of the pulse graph identical to the second?
Yes, this is because of how the pulse finding algorithm works.
Since the number of pulses could be over 2000, it's impossible to
draw a graph of the entire duration of the pulses (they do this
with the gaussian graph), so they draw the result of the folding
algorithm twice to make it look periodical.
Here's a description taken from a post by Eric Korpela of
SETI@home:
------CLIP-------------------------------------------------
The pulse finding algorithm works like this (in general, not
quite in detail):
The algorithm is called a folding algorithm. Suppose we have a
data stream of 66 points that looks like this:
001100010010010011010001000100001111010110000110011101010011111100
The folding algorithm looks first for things with a period of N/3
or 22 samples by adding up the points in groups of 22
0011000100100100110100
0100010000111101011000
0110011101010011111100
----------------------
0221021201221212232200
Now we look in this folded stream for an event above a threshold.
If there was a strong signal with a period of 22 samples, the
peaks from those signals would line up and we would see a peak
in the folded array. Now we take the folded array and fold it in
half again to get a period of 11 samples:
02210212012
21212232200
-----------
23422444212
And we look for peaks above a threshold. And again to get an
(average) period of 5.5 samples (it gets tricky with non-integer
periods, and I'm really not attempting to do a good explanation
of this part)
234224
444212
------
678436
And again to get a period of 2.75
6 8 7
4 6 3
--------
10 14 10
Then we go back to the original data and search on a slightly
smaller period, in this case 21+2/3 ~= 21.6667. We do this by
shifing our end point by one sample.
Here's the original data again:
001100010010010011010001000100001111010110000110011101010011111100
Here's the samples we add together, note the last row was shifted
by one.
0011000100100100110100
0100010000111101011000
0011001110101001111110
----------------------
0122011210312202232210
Now we fold that in half to search on a period of (21+2/3)/2 or
about 10.83333
01220112103
12202232210
-----------
13422344313
And so on. And so on.
For a given array of length N, we search periods of
N/(3*2^n) to N/(4*2^n) in period steps of 1/(3*2^n)
with n=0 to log_2(N/3)-1
N/(4*2^n) to N/(5*2^n) in period steps of 1/(4*2^n)
with n=0 to log_2(N/4)-1
N/(5*2^n) to N/(6*2^n) in period steps of 1/(5*2^n)
with n=0 to log_2(N/5)-1
In principle, you could go on from there to N/(6*2^n) and onward,
but you've reached a point of diminishing returns, most of the
periods you would search would have already been covered. You
only gain in sensitivity to pulse duration much smaller than the
sampling rate, and since SETI@home is designed to be insensitive
to things with large bandwidth, we probably wouldn't detect
signals of that short a duration anyway.
-----/CLIP-------------------------------------------------
What you see on the graphics in the new S@H client is the folded
data.
3 Third-party software
3.1 JSETITracker
By James Birchfield
3.1.1 Programmer's comments
JSETITracker is an add-on client for the SETI@home project
software. It provides a vast array of information that is either
not found in the SETI@home software, or is hard to find.
JSETITracker, in addition, provides logging of all work units,
and two different visualization methods to view your data. The
first and simplest is the SkyMap. The SkyMap plots each work unit
against a whole sky map to show you where your work units have
come from. Each work unit shown is selectable and information
about each is readily available with the click of a mouse. The
second is a JSETITracker exclusive, CoordinateTracker.
CoordinateTracker requests the detailed image of the area of sky
that the work unit was recorded from SkyView, a NASA website. The
image is then placed on the CoordinateTracker panel, and the work
unit's coordinates are plotted on top of this image. This
provides the user with a detailed path that the work unit
followed as the receiver traveled along the sky. As the work unit
processing progresses, a small square travels along the plotted
line to indicate which part of the sky the current processing is
currently looking at. The user may also at anytime choose anyone
of the 20 or so coordinates and view their location in the sky
with a different square.
JSETITracker is written entirely in Java, and requires Java 1.1.6
or higher, as well as JFC(Swing) 1.1 or higher. JSETITracker acts
as a passive monitor to the 'state' files that the SETI@home
software produces. It polls these files on a set interval and
updates the display accordingly.
JSETITracker has ben known to run successfully on a variety of
Java enabled platforms including: Windows 95/98/NT, Linux,
Solaris, OS/2, and Macintosh. JSETITracker should work on any
other Java enabled platform as well.
JSETITracker is deployed using Zero G's InstallAnywhereNow
product. There are are platform specific installers for Windows
95/98/NT, Macintosh, and Unix. There is also an 'other' installer
for any other Java enabled platform. JSETITracker is also
distributed as a single zip file that contains all the files
necessary to run JSETITracker, including the JSETITracker .jar
file and all associated images.
JSETITracker is free to use by anyone.
Get it: http://zap.to/jsetitracker
3.2 Tk-SETI@home
By Rick Macdonald
3.2.1 Programmer's comments
TkSETI is a GUI front-end to the SETI@home client for UNIX. It is
fully customizable with lots of cool features:
Can passively monitor an already running SETIathome client.
Can start/stop/pause the SETIathome client when TkSETI
starts/exits.
Can manually run/kill/pause/continue the SETIathome client.
Can automatically stop the client during certain hours on certain
days.
Can run your dialup network start/stop scripts when the client
needs to contact the server (even avoiding certain days and
times).
Restarts the client if it dies unexpectedly.
Linux only: can automatically run/kill/pause/continue the
SETIathome client based on system idleness by monitoring any
devices such as keyboard, mouse, etc.
Support for proxy servers.
Tracks your personal best scores for big Spikes and Gaussians,
and notifies you when new bigger ones are found.
Displays your statistics such as work units processed, total CPU
time, progress of current work unit, largest Spike and Gaussian,
client CPU usage, etc.
SkyMap shows the location of all work units processed plus the
location of your best spike and gaussian.
Fully configurable to run on any UNIX platform.
Font selector.
Lets you check the work statistics of your friends to see who is
ahead.
Notifies you if you or your friends make the Top Users, Spikes or
Gaussians lists.
Notifies you when a new version of TkSETI is available.
Contact Rick Macdonald <ri...@vsl.com> with any problems or
enhancements ideas.
TkSETI checks after every work unit for updates and notifies you
when a new version is available. A window is popped up and also a
message is placed in the TkSETI window manager title bar.
The latest version is available from
http://www.cuug.ab.ca/~macdonal/tkseti
3.2.2 Tk-SETI@home installation
Untar the distribution file:
gunzip -qc tkseti-1.38.tar.gz | tar xvf -
and just place the tkseti file anywhere in your path.
There is a "contrib" directory where various scripts and
information has been contributed by TkSETI users.
TkSETI requires Tcl/Tk 8.0 or newer.
See http://www.scriptics.com/download
3.2.3 Tk-SETI@home startup
You must run the SETIathome client once manually from the command
line to get registered. Just answer all the prompts. Once the
client is running properly, you can run tkseti.
TkSETI can be started if the client is already running, or it can
start the client manually or automatically for you. This is
explained in the Setup section that follows.
TkSETI will look for the client files in the directory
~/setiathome. If you've run the client elsewhere, or run more
than one client, just specify the setiathome client directory on
the tkseti command line. For example:
tkseti ~/setiathome
3.3 SETI Spy
By Roelof Engelbrecht
3.3.1 Programmer's comments
SETI Spy is a little program I wrote to "spy" on the progress and
performance of the SETI@home client. I initially developed it for
my own use, but I have decided to make it available to the
general public free of charge.
The graphical SETI@home client displays the progress and status
of the analysis, but generating the graphics uses 60% or more of
the available computing power. Some folks, including myself,
would much rather use all of the available power to crunch data
quicker than look at the pretty pictures. Enter a new type of
software -- the SETI@home tracker -- that displays that progress
an status of the analysis without having to generating the
time-consuming graphics. There are some good SETI@home trackers
available, but I wanted something to display the information I am
interested in -- the progress and especially the performance of
the SETI@home client. This is why I wrote SETI Spy.
You can get SETI Spy at <http://pages.tca.net/roelof/setispy/>
3.3.2 Processing efficiency
I developed SETI Spy to provide a tool that can be used to ensure
that you are running your SETI@home client at peak efficiency.
For benchmarking purposes I developed the following table of peak
efficiencies from work unit speeds I measured and those reported
on various news groups, bulletin boards, and web sites.
Processor Peak Efficiency
(cycles / FLOP)
AMD K6 10.0
AMD K6-2 11.0
AMD K6-III 10.5
AMD Athlon 8.5
Intel 80486DX2 18.0
Intel Pentium 12.0
Intel Pentium MMX 9.5
Intel Pentium Pro 8.5
Intel Celeron 8.5
Intel Pentium II/III 8.0
Intel Pentium II/III Xeon (512kB L2) 7.5
Intel Pentium II/III Xeon (1MB L2) 5.5
Intel Pentium II/III Xeon (2MB L2) 5.0
Sun Enterprise 4000 5.0
Sun Ultra 60 5.2
PowerMac G3 6.5
PowerMac G4 4.5
The peak efficiency of your processor depends on a number of
factors, including:
1. Floating Point Unit design
Since most of the processing is done on floating point
numbers, a very efficient Floating Point Unit (FPU) is
essential for good performance. The Intel Pentium Pro,
Celeron, Pentium II/III (Xeon), and AMD Athlon have pipelined
FPUs which are more efficient than the non-pipelined FPUs of
the other processors.
2. Cache size and cache speed
The most time-consuming part of SETI@home is the FFT routine
which accesses a data set slightly larger than 512kB.
Performance is much improved if this data set fits entirely in
the L2 cache of the processor, as is the case for the 1MB and
2MB Pentium II/III Xeons. In addition, the fast L2 cache like
that of the Pentium II/III Xeon improves performance even
more.
3. Memory size and speed
SETI@home requires about 16 MB of memory. The quicker it can
access this memory, the faster it will run. Low latency memory
will reduce the access time and speed up processing. Having at
least 64 MB of physical memory will avoid swapping of the
SETI@home code and data to slow virtual memory when running
SETI@home together with other software.
4. Operating system
Some operating systems are more efficient than others. For
example, a processor will be slightly more efficient under
Windows NT than under Windows 95/98. Also, more efficient
SETI@home clients exist for certain operating systems. For
example, there is a Linux text client optimized for 686-class
machines, but the Windows clients are optimized only for
386-class machines.
You can use the values in the table to determine if your
SETI@home client is running at optimal efficiency. If your cycles
/ FLOP value is much higher than value in the table for your
processor, you can probably improve your processing efficiency by
using some of the tips in this FAQ.
You can also use the values in the table to estimate the optimal
work unit processing time for your processor, using the following
equation:
Topt = 555 (CpF / MHz )
where
Topt = Optimal WU processing time (hours)
CpF = Cycles per FLOP (from table)
MHz = processor speed in MHz
For example, a 350 MHz Pentium II is expected to process one work
unit in 555 (8.0 / 350) = 12.69 hours.
3.4 SETIWatch
By Mark Loukko
3.4.1 What is SETIWatch?
After using SETI@home for the last few months, I recently
downloaded the command line version for NT. While the command
line version is running it just displays the percentage
completed. I wanted to know a little bit more, so I wrote a
program called SETIWatch. It turns out if you're using the screen
saver version of SETI@home you can also benefit from SETIWatch.
SETIWatch has been tested on Windows NT and 98.
3.4.2 Some background
On June 28, 1999 I released SETIWatch to the general public.
Well, all I can say is WOW, I've been completely blown away by
the response. So many people have sent me their complements and
enhancement requests I've had a hard time keeping up. I've done
my best to complete as many of the requests as I can.
Unfortunately I do have a full time job and some enhancements
will have to wait.
3.4.3 Where can I get it?
Download it from this homepage:
http://members.home.net/mloukko/
3.4.4 How to install SETIWatch
Place SETIWatch.exe into the same directory as SETI@home and run
it.
3.5 SETILog
By Mark Loukko
3.5.1 What is SETILog?
Many people have ask me to include a way to record completed work
units in SETIWatch. This task turned out to be a lot harder than
I thought it would. I wanted a method that works every time, even
when SETIWatch is not running. It turned out SETIWatch is not the
place to capture completed work units. Instead, I developed a
small (8k) program called SETILog.
3.5.2 How does SETILog work?
The key to capturing a completed work unit is to run the command
line version of SETI@home in a batch file. First SETI@home runs
and then SETILog. This way we guarantee when a work unit
completes we also log the results.
When the work unit completes and SETILog runs, it grabs
information about the work unit and places it into a csv (comma
separated values) file called SETILog.csv. SETIWatch can read
this file and displays the results in the "Completed Work Units"
window. This csv file can even be loaded into Microsoft Excel,
Access etc where you can do your own analysis if desired.
3.5.3 RunSETI.bat
RunSETI.bat looks like this:
:Start
seti.exe -stop_after_process
if exist result.txt goto SaveLog
if errorlevel -1073741510 goto Stop
goto Start
:SaveLog
if errorlevel 0 SETILog.exe
goto Start
:Stop
A couple of points regarding the batch file:
1. Wondering what -1073741510 is for? Windows returns this number
when Ctrl-C or Ctrl-Break is pressed.
2. The file name for the command line version of SETI@home is
quite long. Version 1.3 is
"setiathome-1.3.i386-winnt-cmdline.exe" I find this a little
tedious to type at the best of times! I've renamed my file to
seti.exe This is what the above batch file refers to.
3.5.4 Where can I get it?
Download it from this homepage:
http://members.home.net/mloukko/SETI.html
3.5.5 How to Install SETILog
Step 1. UnZip SETILog.zip into the same directory as SETI@home.
Step 2. Run the batch file!
3.6 SetiTEAM
By Sqiz
3.6.1 Description
Current Version 16th April 2000 = 1.6b
SetiTEAM is free software for Windows 95+ downloadable from
http://www.sqiz.co.uk/Seti/setiteam.html. It requires no special
installation, can use a standard internet connection or supports
access via a proxy server.
SetiTEAM interprets the team results and statistics pages on the
SETI@home server to provide a more convenient interface with
additional features and flexibility.
SetiTEAM allows the team / statistics webpages for any group to
be sorted (by Position, Name, WU's, Total or Average time), saved
(in Word, Excel, HTML, Notepad, CSV or Clipboard formats),
printed, or viewed as bar charts.
In addition to remembering the previous session for use off-line
as an aide memoir, results can now be saved for later comparison.
This is useful to spot members joining / leaving teams and can
highlight how different members are performing.
The latest version also has a unique feature which allows you to
predict the status of teams / members up to 3 months ahead. Lots
of fun if you want to know when User A will overtake User B.
Team founders can use a special mode to list the entire
membership of the team along with all the Email addresses.
Lots more features, including built in help with clickable links!
3.7 SETIBuf
By Terry Lee
3.7.1 Legal notice and stuff
SETIBuf is a set of *.bat files and instructions, created by
Terry Lee. They are offered on an as-is basis without charge, and
may be freely redistributed as long as the integrity of the
installation package is preserved If you wish to distribute
SETIBuf with modifications, please include the unaltered
SETIBuf.zip file along with your modifications in your own
package, and call it anything other than SETIBuf.
The batch files provided rely on SETIWatch and SETILog from Mark
Loukko (because they are such nice tools!) However, instructions
for doing the work unit buffering without using these programs
are part of the SETIBuf.doc document.
IMPORTANT: The SETIAtHome command-line client will not run on
Windows 95. You must be running Windows 98, Windows
NT 4.0, or Windows 2000 in order to use the
command-line client or this procedure If you are
running Windows 95 and do not wish to upgrade, then
you can use only the SETIAtHome GUI (screen-saver)
version.
(Note from the FAQ-maintainer, you CAN run the CL
client on Windows 95, see section 2.1.2 subsection 7)
3.7.2 General description
One Work Unit (WU) is kept in each of the active 1 thru 9
folders.
To keep one of the 1 through 9 folders from participating in the
Work Unit cycle, rename the SETI.ok file it contains to SETI.no.
To reactivate such a deactivated folder, rename the SETI.no file
to SETI.ok. The original distribution files have folders 1 thru
9 activated; if upgrading or reinstalling, this scheme retains
the settings you currently have. By renaming the SETI.ok/SETI.no
files in those folders according to the above scheme, you can
control the size of your work unit buffer. WU
sending/downloading/ processing will only be performed if there
is a SETI.ok file present and there is not a SETI.no file
present.
To stop a running SETI instance at any time:
Restore the window if it is minimized
Type CTRL+C
Reply Y to the 'Terminate batch job (Y/N)?' prompt
To stop processing on one WU and move on to the next WU
directory:
Restore the window if it is minimized
Type CTRL+C
Reply N to the 'Terminate batch job (Y/N)?' prompt
When you try to reboot a Win98 computer with a SETI instance
running, you will get a message box advising you that Windows
cannot stop the process. Proceed as follows:
Press the OK button. The SETI window will restore if it is
minimized
Type CTRL+C
Reply Y to the 'Terminate batch job (Y/N)?' prompt
Whenever SETI.bat is started, it first checks all the activated
buffers, sending in any completed Work Units and replacing any
sent in with new ones, and filling any empty buffers. After all
the activated buffers have been filled, it then begins processing
at the point where processing was last interrupted. If no
process was interrupted, it begins with the lowest-numbered
active folder. Whenever a WU is completed, all the activated
buffers are checked again, and refilled as required. Following
the buffer filling, the WU processing resumes with the WU in the
next activated folder. This way, the maximum number of work units
will always be available.
The AutoDial.ok file is a signal to SETI.bat that it should
attempt to connect to SETIAtHome automatically when one or more
of the WU buffers requires attention. You can suppress this
automatic connection by renaming it to AutoDial.no. Automatic
connection will be attempted if, and only if, both of the
following conditionsare True:
AutoDial.ok does exist in the SETI folder
AutoDial.no does not exist in the SETI folder
If you disable automatic connection by deleting the AutoDial.ok
file, instead of by the methods described above, then automatic
connection attempts will be resumed by SETI.bat if you should
ever upgrade or reinstall SETIBuf.
If you have suppressed automatic connection with AutoDial.ok/
AutoDial.no, or if some of the automatic upload/download attempts
have failed, you can try again without interruption of WU
processing by invoking SETICall. SETICall attempts to connect to
SETIAtHome regardless of the AutoDial.ok/no settings.
Additional processing scenarios are supported. See SETIBuf.doc
for details.
Multiple concurrent Work Units (for machines with multiple
processors) are supported by SETIBuf Full instructions are
included in SETIBuf.doc.
3.7.3 Where can I get it?
SETIBuf is available from:
http://www.hallquist.net/SETI/SETIBuf.htm
3.8 SETI Monitor
By Lior Fainshil
3.8.1 Description
SETI Monitor is a free add-on for SETI@home. It allows you to
monitor the activity of your SETI@home client and see what it
finds using almost no CPU power. SETI Monitor shows the signals
found by your SETI@home client and saves information about
completed work units. You can see the signals found in your
current work unit, browse through previous results and see the
totals. I used SETI@home for days with SETI Monitor and without
it and couldn't see any difference in performance. SETI Monitor
is highly optimized and its effect on performance is so
insignificant that it is very hard to measure. SETI Monitor works
with both the graphic and the text clients.
3.8.2 Some more details
SETI Monitor reads the files created by SETI@home. Some of them
contain the results which are going to be sent to the back
server. SETI@home currently searches for two kinds of signals:
spikes and gaussians. SETI Monitor shows these on a graph, where
spikes are shown in red and gaussians in blue. This is in
contrast to the other well known add-ons, which currently can
show only the parameters of the best signal. If SETI Monitor is
in memory when a work unit is completed, it automatically saves
the results and allows you to review them later in the same way
you see the current work unit. You can also see all the found
signals from all work units on one graph.
SETI Monitor has a few
settings. You can set for example if SETI@home is loaded on
startup. There are some hidden settings as well, which are
configured automatically without causing troubles to the user
with questions that only make things hard. It also has a very
small and clean uninstaller.
The best way to see what SETI Monitor is, is of course to look at
the screenshots at
http://www.zrlm.com/highstress/setimon/screen.htm
3.8.3 Where can I get it?
http://www.zrlm.com/highstress/setimon/
3.9 SETI UniT Manager
By Vicksoft, Christian Vick
3.9.1 About SUM
SETI UNiT Manager (SUM) is a very versatile add-on for the
Macintosh client, and offer many features:
* Buffering of workunits
- automatic up-/downloads to adjustable times and...
- automatic up-/downloads after an adjustable amount of
finished units or...
- manually up-/downloads.
* Independece of the SETI-Server:
- no interrupts of crunching during times of bad
connections to the SETI-Server.
- offline crunching during your 4-week vacation.
- comfortable use of computers without
internet-connectivity as SETI-workstations.
* Automatically uses a Ramdisk, if present, to save your HD and
make it less noisy. Optional backup to disk of the data.
* Statistics includes CPU usage in percent, total time, crunch
time, CPU time, all average times, all best/worse times and how
many UNiTs have been crunched.
* After sending results to the SETI-Server, SUM can take
schreenshots of the current stats.
* Can help you connect if you use certain ISP's that sends you
advertisements first, regardless of what you requested.
* Famechecker, checks your units against those on SETI@home's
Top-20 page
3.9.2 Cost
SUM donates 2% of your workunits to the "Magic Village Team",
otherwise it's completely free.
3.9.3 Requirements
* Mac OS 8.6 or 9.x. (8.5.x also, but check the SUM webpage)
* Akua Sweets 1.3.6. or higher (www.akua.com)
* Optional: Text-to-Speech to let SUM speak.
3.9.4 Where can I get it?
http://www.cooc.de/applescript/
3.10 Setimgr
By Bernard Hatt
3.10.1 Programmer's comments
Setimgr is a simple manager for the SETI@home clients on UNIX
machines. It buffers work units, runs multiple instances on
multiprocessor machines and outputs a primitive progress report.
The latest version is 0.03a (BSD style license)
http://www.arkady.demon.co.uk/seti/setimgr-0.03a.tar.gz
3.10.2 Setup
Setimgr requires:
* A setiathome client binary
* A compiled setimgr binary
* A setimgr.conf config file
* A sub-directory for each buffered work unit (called proc[n])
If there isn't a setimgr.conf in the current directory setimgr
will output an example config file with the default values in it.
Further details can be found in the readme file.
3.10.3 Operation
Setimgr can either be setup (in the config file) for a permanent
connection, where it will attempt to upload/download completed/
new work units as soon as one is finished, or for dial-up
connections it will wait until it receives SIGHUP to initiate
transfer.
Details of the config file and setup can be found on the web
page: http://www.arkady.demon.co.uk/seti
3.11 Seti4Net
Andre Starkloff is writing an application to monitor multiple
machines on a net running Seti@home.
You can monitor his progress at http://www.seti4net.de.vu/
3.12 Multi-SETI@home Monitor - Msetimon
http://msetimon.sourceforge.net
A graphical add-on package to monitor seti@home activity that
may be running on multiple computers over a network or multiple
instances on the same computer. Runs on Windows or Linux.
4 Homepages
This document is more than a FAQ, below you'll find a few pages
related to SETI and SETI@home. Consider is at as "guide" instead
of answers in a FAQ.
Dale's "Star Rating"
"0" = The Pits. Don't even bother going there.
* = If you don't have anything to do, well... maybe.
** = Interesting, but has room for improvement.
*** = Very nice site. Interesting, Informative, Could be
spruced up a bit.
**** = Cool site. I was impressed. Go There. Be informed and
pleased.
***** = Way To Do A Site! I'm Impressed! GO HERE!
4.1 Homepages concerning SETI@home
4.1.1 SETI@home home
http://www.setiathome.ssl.berkeley.edu/
Reviewed by Dale Willamson
This is the home-base of operations for the SETI@home scientific
project. It's the official place where you can download the
latest version of the processing client software, but it also has
many interesting areas that are well worth checking out.
Perhaps one of the more interesting aspects of the SETI@home Web
site, is the "News and Statistics" sections. Here you can find
some really interesting information presented in graphical form,
concerning how many new volunteers are joining the search for ETI
each day; The total so far, of work units processed by groups and
individual volunteers; The top 20 "spikes" and "Gaussians" found
so far; Graphs and Maps; an updated report on things relating to
the project from storm threats, to hardware upgrades. You can
even take the "SETI at Home Poll" and give your reasons for
joining the search for ETI and also some of what your own
opinions and thoughts might be about "ET"!
As far as "looks" are concerned, it's a little "dark", but hey --
these guys are "Scientists", and not necessarily polished at
making a Web site look really, really cool! But then again, I'm
glad they know what they're doing in searching for ET instead of
great at putting those lame "flames" all over the place, aren't
you?
One other mention that I'd like to make, is that near the bottom
of the Front page, you'll find a listing of those companies that
have made some substantial donations to the SETI@home Project. If
you can find the time, it wouldn't do any harm to log onto those
donating companies and leave a message telling them that you
appreciate the help they've given to this project. Though there
are no strings attached to these donations they have given, any
business likes to know that they are appreciated and this is a
great way to show them. Too, you will also find there, a place
where YOU can also make a donation to SETI@home. While they
certainly appreciate everybody helping them process their
collected data from the Aricebo Dish Antenna, they really need
some money to help buy more equipment. So, if you could spare
what it costs to go to a movie once, it would make quite a
difference.
4.1.2 SETIweb
Reviewed by Dale Williamson
Hosts the sci.astro.seti pages, where links can be submitted,
and binaries can be posted, as the s.a.s. group do not permit
binaries.
Here it is:
Stan Schonberg is the editor of these pages.
4.1.3 SETIforum
4.1.4 SETI @ SixDegrees
http://www.geocities.com/~kris_j/seti/index.html
Reviewed by Dale Willamson - Dale's Rating: **
The above URL will take you to a very nice site created by Kris
Johnson, however if you click on the Seti@SixDegrees button, you
will get clobbered with some commercials, so be fore-warned. But,
it _is_ an interesting site anyway. For most of the options, you
do need to become a member, but I will note -- It Is Free! This
is most definately a "Seti@Home" type of place, with some good
starting points. I'll try to re-visit this place in a little bit
and see how they've improved things.
4.1.5 SETI@home Speedup Tips
http://www3.sympatico.ca/n.rieck/docs/seti_tips.html
4.1.6 Derived statistics for SETI@home @ Rovingmouse
http://www.roving-mouse.com/setiathome/
4.1.7 SETI STATION
http://www.flex.com/~daniel/SETI/
Reviewed by Dale Willamson - Dale's Rating: **
If you've got a MAC computer, check this site out. If you loose
the pointer to it, just go to the SETI@home site and go to
"Related Web Sites" and it'll be on the top of the list. Just
"click" and you're there. It's even been rated as "Internet Site
of The Month" by My Mac Magazine - Sept. 1999.
It's got some "fun things" at SETI Station, as well as some
serious stuff too, like learning how to speed up your MAC client
processing by using RAM disk. There is even a Poll available to
let them know what kind of MAC you're running. There's Tips,
Teams, Winners & Loosers, and articles taken from various
sources.
Though the site is a bit "dark", it is useful and informative. I
didn't care too much for the pull-down windows, but to each his
own.
4.1.8 SETI: The Drake Equation
http://www.activemind.com/Mysterious/Topics/SETI/drake_equation.html
4.1.9 Sci.astro FAQ about SETI
http://sciastro.astronomy.net/sci.astro.6.FAQ
4.1.10 Team Canada
Reviewed by Dale Willamson - Dale's Rating: ****
I thought I'd mention a Seti "Team" site this time, just because
I was struck by how nice the effort was done in putting this
together. Some things just deserve a special mention when
everything turns out so well. The "Team Canada" site is very
attractively put together and is hosted by Andrew Turi.
While there is always room for improvement, some great work was
put into this site with just a few "reminders" that Canadians had
something to do with it. <grin> If you go to their section called
"What is Seti @ Home?", you'll find some media links to MSNBC,
CNN, and Time coverage of the Seti@Home project, a pretty cool
"Alien" creature created by a guy named Zombieman, and some Team
newsletters which are maintained by Bob Page. It gives you the
impression that these guys are serious about this project!
The Team Canada site is pleasant to the eyes, not overly-done,
but "classy". I like it, and think you will too. If you decide
to stop by for a visit, leave them an email if for no other
reason, to let them know what you thought about their site.
4.1.11 The Planetary Society
Reviewed by Dale Williamson - Dale's Rating: ****1/2*
Right from the start, this is obviously a highly rated site!
It's well put together -- everything works -- and it's a great
place to find scads of information. At the time of this writting,
Planetfest '99 is going on with the Mars landing only 7 days
away. There's a "Headline" section featuring the hotest current
space science topics, and a "Special Sections" area where you
can operate a rover from a classroom! They even have a very nice
search engine for searching "seti" subjects.
The Planetary Society is now in partnership with Geoman.net in
France and is also accessible in Spanish!
It's a nice site, like I mentioned at the top -- well put
together and worth checking out.
4.1.12 Patch-free-Processing
By Alfred Das
http://home.hccnet.nl/a.alfred/p-free-p.html
4.1.13 Sky & Telescope
http://www.skypub.com/news/special/seti_toc.html
Reviewed by Dale Williamson - Dale's Rating: *****
The Sky&Telescope site. What a pleasant and nicely done site!
Full of interesting features such as Current News, News Archive,
Special Reports, and you can cruise through their current monthly
magazine on-line. Of interesting note, the site also features the
artwork of Lynette Cook, a great artist, and there's a link to
some of her artwork. Check that out as well when you visit here.
(Don't forget to click on the "detail" hotlinks so you can see
her pictures in larger detail, too.)
Paying special attention to just one of the articles being shown
when I visited this site, is a Dec 1998 article entitled "The
Chance of Finding Aliens" and in it you'll find a reevaluated
"Drake Equation" along with some interesting photos.
All-in-all, this is a great site (or it wouldn't have my
five-star rating), and is just loaded with interesting "stuff"!
I was impressed and think you will be too!
4.1.14 SETI Institute
(SETI Institute review: 8.04.2000)
Reviewed by Michael Johnson - Michael's Rating: *****
The SETI Institute is a non profit corperation that was founded
in 1984, and serves as an institutional home for scientific and
educational projects relevant to the nature, distribution, and
prevalence of life in the universe.
The website is very nicely laid out and easy to navigate. Just
about every topic imaginable that is related to SETI can be found
here, including scientific and technological aspects of
astronomy, planetary sciences, biological, chemical and cultural
evolution. Their primary focus is on conducting and encouraging
public education on these topics.
The site is not updated as frequently as I would like to see it,
but I did enjoy breezing through the archives and reading some of
the past news articles. The FAQ page is very well done also, and
answers most questions about the website and their mission.
All in all, a wonderful website that I will visit
frequently....as should you.
4.1.15 SETI League
4.1.16 Removed (11 Feb 2001)
4.2 SETI utilities
4.2.1 SETIwatch & SETIlog
http://members.home.net/mloukko/
4.2.2 SETI Manager
http://home.t-online.de/home/steffen.krasselt/seti/
4.2.3 TKSETI@home
http://www.cuug.ab.ca/~macdonal/tkseti
4.2.4 SETISPY
http://pages.tca.net/roelof/setispy/
4.2.5 JSETITracker
4.2.6 SetiTEAM Homepage
http://www.sqiz.co.uk/Seti/setiteam.htm
4.2.7 SETIBuf homepage
http://www.hallquist.net/SETI/SETIBuf.htm
4.2.8 SETI Monitor homepage
http://www.zrlm.com/highstress/setimon/
4.2.9 SETI UNiT Manager homepage
http://www.cooc.de/applescript/
4.2.10 RunCache & FetchCache
http://setiathome.ssl.berkeley.edu/RunCache
http://setiathome.ssl.berkeley.edu/FetchCache
4.2.11 Seti@home Service
http://www.telepath.com/~dennison/Ted/SETI/SETI_Service.html
4.3 SETI fun
4.3.1 Carolyn's Clinic
http://home.columbus.rr.com/theehlens/index.htm
If you ever suffer from WU withdrawal or some other SETI@home
related illnesses, visit the clinic. Wondering whether to
overclock or not or what that science jargon in the Tech news
means, the clinic will help you with that and just about anything
else too.
5 Acknowledgements
5.1 Sci.astro FAQ
From the sci.astro FAQ I have used section 1.2.2 and 1.2.3, and
have hereby agreed to bring this copyright statement:
Subject: Copyright
This document, as a collection, is Copyright 1995, 1996, 1997 by
T. Joseph W. Lazio (jla...@patriot.net). The individual articles
are copyright by the individual authors listed. All rights are
reserved. Permission to use, copy and distribute this unmodified
document by any means and for any purpose EXCEPT PROFIT PURPOSES
is hereby granted, provided that both the above Copyright notice
and this permission notice appear in all copies of the FAQ
itself. Reproducing this FAQ by any means, included, but not
limited to, printing, copying existing prints, publishing by
electronic or other means, implies full agreement to the above
non-profit-use clause, unless upon prior written permission of
the authors.
This FAQ is provided by the authors "as is," with all its faults.
Any express or implied warranties, including, but not limited to,
any implied warranties of merchantability, accuracy, or fitness
for any particular purpose, are disclaimed. If you use the
information in this document, in any way, you do so at your own
risk.
5.2 People who have worked with the FAQ
The current maintainer of the FAQ wishes to thank
the following persons for their past and/or present
work with the FAQ:
Peter Alfredsen, who maintained the FAQ up to version 1.5.
Walter Novacek who did the html porting up to version 1.30.
Dale Williamson who handled the 4.x.x section until his computer
broke down.
Jan Knutar, who maintained the FAQ from version 1.5.2 to 1.9.8