Spikes, Pulses & Triplets: what constitutes a meaningful value?


log in

Advanced search

Message boards : SETI@home Science : Spikes, Pulses & Triplets: what constitutes a meaningful value?

Author Message
hlr
Send message
Joined: 12 Aug 09
Posts: 5
Credit: 1,918
RAC: 0
Canada
Message 926323 - Posted: 15 Aug 2009, 18:44:59 UTC

Hello:
I'm new (3 tasks so far) and in looking at the results of the second task, I see 5 spikes; 1 pulse; 7 triplets. I've read the home page explanation of what these are, but there is no indication about the range of values these tasks are likely to produce. What constitutes an exciting result to the SETI team--or at least one that comes to their attention as worth another look by them?
Thank you,
hlr

John McLeod VII
Volunteer developer
Volunteer tester
Avatar
Send message
Joined: 15 Jul 99
Posts: 24694
Credit: 522,659
RAC: 19
United States
Message 926340 - Posted: 15 Aug 2009, 19:57:28 UTC - in response to Message 926323.

Hello:
I'm new (3 tasks so far) and in looking at the results of the second task, I see 5 spikes; 1 pulse; 7 triplets. I've read the home page explanation of what these are, but there is no indication about the range of values these tasks are likely to produce. What constitutes an exciting result to the SETI team--or at least one that comes to their attention as worth another look by them?
Thank you,
hlr

Unfortunately, you can't tell anything from a single Work Unit. The only thing that will tell anything is a reobservation of the same spot on a different day that has a similar pattern. The science code to compare reobservations is not quite ready yet.
____________


BOINC WIKI

hlr
Send message
Joined: 12 Aug 09
Posts: 5
Credit: 1,918
RAC: 0
Canada
Message 926668 - Posted: 17 Aug 2009, 2:06:25 UTC

Thank you. I realize that one Work Unit isn't meaningful in itself. What I'm asking is how many spikes, pulses, triplets and gaussians does it take for the SETI team to BOTHER going back to look at a section of sky based on a single, substantiated Work Unit? For example, 3 of anything in these results is more interesting than 0 to me, but what is the minimum number of these signals that make a result scientifically interesting enough to check it out over a longer period of time (considering how expensive telescope time is relative to the vastness of space)?

I hope this is clearer. Regards, hlr

David A. Lethe
Send message
Joined: 23 Dec 00
Posts: 2
Credit: 69,979
RAC: 0
United States
Message 927277 - Posted: 19 Aug 2009, 21:43:30 UTC - in response to Message 926668.

I feel the same way and there just isn't anything that tells us whether or not the X hours of CPU time was worth the effort.

Throw us a bone. Generalize the threshold for something "interesting", "exciting", and holy sh*t, drop everything and compare results with such-and such, cause we may have found ET.

If it takes a few reobservations because 99% of the work units have spikes, triplets and so on, then why not use your database to tell us how we did? I.e, when enough work units come in from different computers running the same thing, and those work units provide something "exciting", then send out an email or let the BOINC engine pop something up on a screensaver to all machines

You have to remember, a lot of us are running this stuff on home computers, you have to give us something to look forward to for the time we run the software .. if not, people will stop running it and move on.


____________

David A. Lethe
Send message
Joined: 23 Dec 00
Posts: 2
Credit: 69,979
RAC: 0
United States
Message 927278 - Posted: 19 Aug 2009, 21:43:30 UTC - in response to Message 926668.
Last modified: 19 Aug 2009, 21:47:37 UTC

... if work results A,B,C,and D all need to line up a certain way to have an exciting moment, then upload this in the background to computer D after A,B,C's work units have been checked in. Then the guy with computer D can have some fun watching the screen hoping this will mean something.

It is stupid and immature, but we Americans want instant gratification or we get bored.
____________

Profile Johnney Guinness
Volunteer tester
Avatar
Send message
Joined: 11 Sep 06
Posts: 3092
Credit: 2,614,621
RAC: 2,430
Ireland
Message 927335 - Posted: 20 Aug 2009, 2:25:58 UTC - in response to Message 926323.

Hello:
I'm new (3 tasks so far) and in looking at the results of the second task, I see 5 spikes; 1 pulse; 7 triplets. I've read the home page explanation of what these are, but there is no indication about the range of values these tasks are likely to produce. What constitutes an exciting result to the SETI team--or at least one that comes to their attention as worth another look by them?
Thank you,
hlr

Q:What constitutes an exciting result to the SETI team--

A: The brand new Near-Time Persistency Checker (NTPCkr)(a process which ranks the statistical interest of points on the sky).

Do read the NTPCkr FAQ to help you understand how it works.

John.
____________

Bill
Send message
Joined: 13 Aug 09
Posts: 1
Credit: 1,633
RAC: 0
United States
Message 927358 - Posted: 20 Aug 2009, 4:12:09 UTC

This is all very new to me. I think I know the goals, but I don't know how to get there. What am I looking for? And how will I know when I find it?
____________

Profile Johnney Guinness
Volunteer tester
Avatar
Send message
Joined: 11 Sep 06
Posts: 3092
Credit: 2,614,621
RAC: 2,430
Ireland
Message 927558 - Posted: 20 Aug 2009, 22:25:52 UTC - in response to Message 927358.

This is all very new to me. I think I know the goals, but I don't know how to get there. What am I looking for? And how will I know when I find it?

Bill,
There is a very steep learning curve with the new Near-Time Persistency Checker. There is a lot of science information, you need to do some reading if you want to understand how it works.

But in general, your just volunteering your PC's free time. So its the project scientists that will be doing the in depth analysis.

John.
____________

Profile Donegal_TDI
Avatar
Send message
Joined: 14 Nov 02
Posts: 153
Credit: 26,925,080
RAC: 7
Ireland
Message 927701 - Posted: 21 Aug 2009, 13:39:57 UTC - in response to Message 927558.
Last modified: 21 Aug 2009, 13:43:33 UTC

8 out of 10 of the top 10 results from the NTPCkr come from mostly the same region of sky, the bottom right hand single star of the great square of Pegasus, Markab.

From Wikipedia.....
"Alpha Pegasi (α Peg / α Pegasi) is the third brightest star in the constellation Pegasus (despite its "alpha" designation) and one of the four stars in the asterism known as the Great Square of Pegasus. It has the traditional name Markab (or Marchab).

Markab is a relatively average star nearing the end of its stellar evolution on the main sequence. Markab will soon enter the helium burning phase of its development, during which it will likely expand into a red giant. Like the Sun, it will probably end its life quietly as a white dwarf."

From the above, this star is old.
It's 140ly away from us, 5.8 solar radii in size,
has a spectral class of A0V and temperature of 8919 Kelvin.
Likely canditate to potentially have a very evolved civilisation?
____________


*** Those who know, don't speak,
those who speak, don't know ***

Profile Johnney Guinness
Volunteer tester
Avatar
Send message
Joined: 11 Sep 06
Posts: 3092
Credit: 2,614,621
RAC: 2,430
Ireland
Message 927715 - Posted: 21 Aug 2009, 14:53:22 UTC
Last modified: 21 Aug 2009, 15:12:07 UTC

Interesting Donegal_TDI,
There are 2 candidates almost around the same spot, they are just a little bit above, and to the right of Markab in Pegasus.

Candidates 20582818 and 20582801. So i used Microsoft WorldWide Telescope to zoom in on the area. This is a screen-shot and i marked the two candidates;



Using WWT (WorldWide Telescope)to zoom in on the area of sky, i see nothing that might stand out from the norm, but that would be taken for granted. We won't see Aliens waving back at us...LOL. WWT does show several well documents Galaxy's in the same area but they are not near our 2 candidates, and either way, the Galaxy's would be way off in the distance, not in our MilkyWay galaxy.

Markab is much closer to home, and if we are detecting something transmitted from a planet that orbits Markab, we won't see the planet in the image either. And another thing, if the planet we see is orbiting Markab, the planet will be moving so the spikes we detect will also be moving around on the sky map depending on the planets orbital position and the planets orbital year and depending on when Arecibo recorded each spike, Gaussian, pulse or triplet.

This will require further investigation, i will spend some time on this one. Its what i describe as promising :)

John.
____________

Profile Donegal_TDI
Avatar
Send message
Joined: 14 Nov 02
Posts: 153
Credit: 26,925,080
RAC: 7
Ireland
Message 927717 - Posted: 21 Aug 2009, 15:01:07 UTC - in response to Message 927715.
Last modified: 21 Aug 2009, 15:16:21 UTC

Haven't been able to find any details of discovered Exoplanets orbiting Markab.
Just because no one has found them, doesn't mean they are not there.
Just checked Markab on StarryNight 6,
no other stars of particular relevance near it.

Let us know what you come up with.
____________


*** Those who know, don't speak,
those who speak, don't know ***

Profile Johnney Guinness
Volunteer tester
Avatar
Send message
Joined: 11 Sep 06
Posts: 3092
Credit: 2,614,621
RAC: 2,430
Ireland
Message 927786 - Posted: 21 Aug 2009, 19:06:36 UTC
Last modified: 21 Aug 2009, 20:05:59 UTC

Ok, so i plot 8 of the candidates in the current top 10 list using the RA and DEC for each candidate inside WWT(Microsoft WorldWide Telescope) and i take a screen-shot of each one as I'm going along.
candidate numbers;
20582818
20582801
20504996
20493683
20499128
20504975
20496034
20504984

This is where the candidates are located in the sky;



Here is a plot of those candidates;



So then looking at the positions of the candidates, the three on the right hand side don't seem to have an obvious pattern or orbit. In fact none of the candidates seem to orbit Alpha Pegasi (Markab) at all! They look like they orbit something else!


But the 5 candidates on the left of Alpha Pegasi (Markab), do appear to have a pattern. They look like they are orbiting something that i cannot see in the image? So i roughly plot what might be an orbital plain for these 5 candidates, its marked as a green ellipse in this next image;



So far this seems to make sense. But in WWT, i am viewing the sky using the Digital Sky Survey (Optical). So i start searching for other telescopes that have surveyed the same part of the sky but in different wavelengths. So then i discover the "2MASS" sky survey (Two Micron All Sky Survey). Now the 2MASS survey shows many more objects that are not visible with other telescopes. So i take a screen shot of the same part of the sky around Alpha Pegasi (Markab), and i superimpose the image on top of the candidate plot i made.

This is the result;



The 2MASS survey does show some type of star located almost in the middle of the orbital signal plot i made.

Now look, i can't actually find any decent image of this unnamed star sitting to the upper left of Alpha Pegasi (Markab), so i really don't know what it is or what the name of the star is? but what i can say is that our 5 candidates appear to orbit something, and this unnamed star is in a position that looks quite close to the centre of the orbit.

This is the best i can do to show a closer image of the star that our candidates look like they are orbiting;



Here is a close-up screen shot of the unnamed star;



In my opinion, i do find this apparent orbital pattern for our candidates, and the unnamed star quite interesting. We could hypothesise and say that the candidates could be a radio source of some kind that might be on a planet orbiting the unnamed star. If it is a radio source on a planet, we are not going to be able to see the planet either way, we can only see the star.

John.
____________

hlr
Send message
Joined: 12 Aug 09
Posts: 5
Credit: 1,918
RAC: 0
Canada
Message 927875 - Posted: 22 Aug 2009, 0:39:37 UTC - in response to Message 927335.

Thanks John, both for your original reply referring me to the Ntpkr site and also for all the work you did this morning with the skyshots. Very interesting.

Assuming you're right about the orbits, that would be very exciting indeed--breaking-news, in fact, at least once proved in many months/years from now, whatever is causing the signals. Thanks for the insight into how the data can be collated and used.

hlr

Profile Johnney Guinness
Volunteer tester
Avatar
Send message
Joined: 11 Sep 06
Posts: 3092
Credit: 2,614,621
RAC: 2,430
Ireland
Message 927881 - Posted: 22 Aug 2009, 1:15:42 UTC - in response to Message 927875.
Last modified: 22 Aug 2009, 1:18:35 UTC

Thanks John, both for your original reply referring me to the Ntpkr site and also for all the work you did this morning with the skyshots. Very interesting.

hlr

Your welcome hlr! I'm glad i was able to help you. I'm probably way off the mark with my plots, its most likely RFI or radio interference. But you never know, lets keep our hopes up!

John.

Profile Johnney Guinness
Volunteer tester
Avatar
Send message
Joined: 11 Sep 06
Posts: 3092
Credit: 2,614,621
RAC: 2,430
Ireland
Message 928042 - Posted: 22 Aug 2009, 16:43:00 UTC

***Important Notice!***

Today, i went back and rechecked my plots, and i think its important that i admit that i made a mistake in my original plot! And i apologise for this!

When i rechecked the RA and DEC position for the 5 candidates i plotted, it seems that i must have plotted one of the positions wrong, it was my mistake for rushing as i plotted the positions, i was just excited when i saw the candidates were all appearing in the same area of sky.

The candidate 20504996 was plotted in the wrong position in the original images. When i rechecked the plot, this candidate had the wrong declination coordinates.

I have corrected the mistake in the following image. This next image shows candidate 20504996 in the correct position. With this correct position, now there is NO Obvious Elliptical Orbit anymore. I do apologise for being hasty in posting misleading information and images. I think its better that i admit my mistake rather than mislead people further.



I am sorry for making this mistake,
John.
____________

hlr
Send message
Joined: 12 Aug 09
Posts: 5
Credit: 1,918
RAC: 0
Canada
Message 928048 - Posted: 22 Aug 2009, 17:50:21 UTC - in response to Message 928042.

No harm done--at least from my end of this thread. There is still the slight possibility that you could be right, and still the exponential probability that, alas, the pattern of signals meant nothing even without a plotting error. So nothing has changed, really. A life without enthusiasm, or daring to be wrong, is excrutiatingly dull and fundamentally pointless when you get down to it.

hlr

Profile Donegal_TDI
Avatar
Send message
Joined: 14 Nov 02
Posts: 153
Credit: 26,925,080
RAC: 7
Ireland
Message 928084 - Posted: 22 Aug 2009, 22:16:55 UTC - in response to Message 928048.

Don't apologise John.
I'm sort of responsible for sending you off on this quest.
Your exercise was very informative and reminds us that this data we are crunching refers to real areas of the night sky.
I have recently taken up astronomy, very amateur I must add, but have a good 8" Celestron scope. When I look through it at the night sky, I'm in total jaw dropping awe of what is out there.
I know in my heart and soul, that we are not alone, the Galaxy is full of life, as to whether we will ever detect intelligence is solely down to technology and luck, as to where they are located in the Galaxy in relation to us.
They may be less than 10ly away, or they could be 1000ly away.

Something wonderful could be discovered at any time now.
The ATA is our big hope and as more dishes come on line in the array, it vastly increases the chances of detection.
____________


*** Those who know, don't speak,
those who speak, don't know ***

Profile ML1
Volunteer tester
Send message
Joined: 25 Nov 01
Posts: 8499
Credit: 4,194,507
RAC: 1,703
United Kingdom
Message 928185 - Posted: 23 Aug 2009, 14:25:57 UTC - in response to Message 928042.

Some good fun research there! And still a nice result. Full kudos for so quickly clearing up the small mistake in the first analysis.

Thanks for the search and efforts.

Now... To add to various possible hypothesis...

What's the error bounds on the coordinates that are reported for the s@h data? Are any "fudge factors" included?


If indeed there are any clusters or orbit-like paths of multiple signals, then that may well be a new discovery!

Keep searchin'!

Regards,
Martin


____________
See new freedom: Mageia4
Linux Voice See & try out your OS Freedom!
The Future is what We make IT (GPLv3)

Message boards : SETI@home Science : Spikes, Pulses & Triplets: what constitutes a meaningful value?

Copyright © 2014 University of California