Discussion Forums

Now able to partition large standard SETI data files, and save hex content in text ASCII format...

14 replies [Last post]
ENGRPOD
Offline
Joined: 2011-06-29
Posts: 11

To jrseti and all other setiQuest members:
After considerable searching all day, I have finally worked out a means to convert the
standard binary-formatted SETI data files into text format, listing the contents in hex.
To do it, I used two freeware packages, one I had encountered previously, called HJ-Split,
to break up the files (there is also a version to rejoin them, when needed), and, once the
files are broken up, a hex editor called HxD.
As I have already said, both packages are freeware; both can be easily located using
Google, however, for sake of ease of reference, I will give the location information here.
The package HJ-Split, as well all of its variants, can be found through websites
www.hjsplit.org, and www.freebyte.com; likewise, the package HxD can be found through
website www.mh-nexus.de, as well.
Now, before going on, I want to be quite clear, I have very definitely NOT yet found a
means of converting the standard 1.9-2.0 Gbyte SETI data files into ASCII or text,
althouhg I will obviously be making every possible effort to do so. However, what I have
now provided all of you with, given some time, experimentation, and the need for some
relatively minor additional work, is a means whereby those impossibly large and extremely
cumbersome standard SETI data files may now, finally, be broken up relatively
conveniently, or so I think. Mind you, I am not entirely positive, nor have I as yet taken
my thoughts to their logical conclusion. However, I have used HJ-Split before, both to
break up files, as well as to join them; I know the package works, and I have now actually
used HJ-Split to physically brak up file 2010-11-06-dorothy_tauceti_1420_1-8bit-01.dat on
website http://184.73.186.167/download/2010-11-06-dorothy_tauceti_1420_1/ into 196 files,
195 of which are each 10 Mbytes in size. I had obviously been trying to break up that
large main data file on the premise that it was, in fact, exactly 2.0 Gbytes; however, it
actually turned out to be slight smaller. Thus, partition 196 oif the break up actually
turned out to be 3.2 Mbytes, pretty exactly; all of the other partitions, 1-195, were,
exactly, 10 Mbytes, as I have already said.
Of course, I have not yet determined whether HJ-Split, as it is presently designed, can,
in fact, allow the breakup to occur in a convenient fashion, in terms of data formatting.
All I AM sure about, is that the partition does work, that the file segments do then load
into the freeware hex editor HxD conveniently, and, best of all, that HxD, unlike many
freeware hex editors, does apparently allow a hex file to be copied to the stanard Windows
clipboard, the contents of which may then be pasted into virtually any word processor
desired. At present, I am trying to figure out how to re-format such a resulting text
file, so that it can then be loaded into Excel, which I do think I should then be able to
do, if need be, one piece at a time, using those manageable 10 Mbyte partitions. I grant
you that the size of the resulting spreadsheet is certainly trivial, and, for
signal-processing purposes, is by no means what I would need for the various
application-software packages I am presently envisioning using. However, the method for
converting the hex representation in the hex editor into an equivalent text format
representation of hex very definitely does work, and I have, in fact, been able to do it.
While getting that text version of the hex data into into an Excel spreadhsheet, with the
proper number of columns to allow for direct conversion to decimal complex-voltage values,
times, and, frequencies, will doubtless take some time, I by no means think such a task is
insurmountable, nor do I think it pointless. In fact, I think it entirely necessary, and
the first step toward allowing a spreadsheet approach to finally be able to be used for
SETI data analysis. Once I can show all of you the various signal-processing and
time-series packages I am envisioning using actually functioning, I think I may then be
able to convince many of you of the desirability of at least having such an approach
available, if nothing else. Obviously, I could be entirely wrong about its practicality;
however, I certainly think the concept worth exploring, if nothing else.
Here, then, is the text depiction of the hex content of some of the first few lines from
hex editor HxD for the contents of that large SETI data file, as mentioned above:
FF 08 0D 1A 0D FB FE EE 06 05 08 01 03 04 ED 0B FF 10 10 11 04 04 0E 0C F4 18 0C 07 F5 04
F0 FF 07 01 04 11 1D 0B F9 EE 01 F5 14 FC FD F7 07 F9 01 F9 0A FB F7 FB FF 0A 09 00 FC F0
04 FB 04 00 FF FA 0E 0F E8 F9 F0 00 0B F9 08 EA 0A 0C 05 F5 06 F5 0A 04 F1 F5 03 FD 00 F7
FE 03 0F FA 06 06 0B 0A FB F9 0B FD F7 F0 07 02 07 04 FE 15 05 07 06 F1 06 02 F8 FA FB FF
05 01 F3 FD 06 02 FF FB FC F3 04 00 0B F9 FF 04 03 EF F8 00 ED 0F EF 05 F6 F7 F5 F5 06 F3
FD F8 03 03 FC 06 FF 10 07 07 0B 0B FE FF 0E 08 02 04 0A 04 12 14 F5 F3 07 EF 17 F3 F7 02
04 06 01 F7 00 FC 01 09 F1 01 EF F9 00 03 F9 FD FA 02 FF 05 00 19 00 09 07 FB FC FD F1 FA
0D FC 09 F5 00 FA F5 FD E8 09 FA 0C F7 04 F3 F8 06 FB 09 02 01 FE F1 00 FF FC F5 02 02 FA
07 0B 04 02 FA DA 01 FB 13 F7 F8 F9 02 FF FF EF FE FB 0F 03 01 0C 12 EF 07 FF EE 0B EF F7
17 FF 0D 10 06 F3 13 EF F4 F1 06 04 00 FB FC F6 F8 02 00 F7 0D F7 F0 F9 03 EC 14 FB 04 04
00 F4 F4 0E FA FA 05 FD 08 FB 11 EB 01 01 F6 03 06 FC 05 01 F3 F8 E3 03 F8 FE 03 FC ED 07
05 02 01 F3 F6 F0 11 FF 01 01 08 07 10 04 04 05 FC F5 F4 06 0B 11 F9 F7 EE F0 05 E4 07 FF
0B 0B EF 08 06 FA FB 03 F1 05 01 00 02 FA E8 FA F1 11 0D 02 FB F6 FC F6 0A F7 07 F7 13 F9
FD 04 03 06 02 F7 FE F1 08 08 FD 09 F3 F4 FF F4 EE F5 FB 00 FF 04 0A FE 11 08 F3 0D F9 03
07 FF 0C 06 F0 02 F8 EE 0E F6 0B F2 F1 0C 00 02 FB 0A 01 0B 01 F7 FC 04 FE FF 0B 00 FA 02
F6 08 0D 09 F8 01 F7 07 F5 00 FC F1 FB F9 FA F6 01 02 03 F8 05 F4 FE FF F1 F2 0D EC F2 F7
F5 FE 02 FB FC FC 0A FC FE F2 0B 01 F7 0B FD 03 FC 08 08 17 FE 0F 03 06 F9 F5 0C 03 06 17
09 F8 FB F8 EF F0 0D FB 01 E7 F5 00 02 04 04 F8 01 03 0B FC 02 0C FA F3 FF FA EF 05 F8 F6
01 FB FE 19 01 03 FC F2 FA 06 07 F7 FD 08 08 04 FC F4 EE 12 FC FD 0A F7 01 00 02 0B 15 F9
FE 05 05 1E 04 FE FC F8 0D 06 04 FB 05 FB FF 10 03 08 F4 F7 01 0B FF 0A F1 FD F8 FF 19 02
09 0A F7 F4 04 FB FE 00 0D 00 16 0D 02 09 02 F7 FE EA F0 00 F7 0C FC 05 FF FA 0D 0A 03 07
F1 FA 00 02 F7 09 FC FB 16 01 04 F3 0E 07 04 11 01 F5 09 01 FA 03 F4 FE FC F4 FD FA F4 0A
F2 0E 04 0A 00 FD FB 03 FB F9 02 F3 05 06 EB FB 03 F5 03 FE 1A 08 12 1B FC F3 06 FF 07 03
06 F6 F7 EF F5 FE FD EE 05 F8 F3 08 EF FA 0D 0C FE 04 F7 F1 04 00 FA FB FB F8 FE 07 F1 FF
EF E4 01 F3 04 FA 15 FF F6 02 08 05 04 04 F7 FD 16 08 07 FA FF F6 0C E9 FE 0A 05 1D F6 05
EC 10 00 03 08 F8 F9 02 FC FD F4 FF 06 04 05 03 04 01 FE 08 F4 EF 04 0D FB 03 02 05 03 06
F6 01 FE F7 FD F5 1B 0C 07 0C 0B F0 11 03 03 F2 FA F0 00 06 02 09 E8 F8 0F 10 18 14 03 02
F9 09 F4 F1 FC 04
Obviously, the remainder of the first 10 Mbyte file is similar. Now, clearly, that large
file must contain the complex voltage values; what I need to know, using the foregoing as
a case in point, is precisely how those complex-valued voltages can be sorted out from the
hex content depicted above. Also, how does one determine the times, and frequencies, for
each of those voltages. Once I know how to assign that information, and organize the hex
information depicted above, I can, with, I think, comparative ease, create an Excel
spreadsheet for each of those 10 Mbyte segments, which, using HJ-Split, I do think I can
realistically assemble into one large, massive multi-gigabye spreadsheet, which might then
be suitable for manipulation and processing using the various data-analysis software
packages I have in mind.
For those reasons, I would be most appreciative if someone, anyone, could please tell me
how to decipher that hex content, now that I finally have it in ASCII text format. I
realize that is not typically how all of you work with such SETI data; I do entirely
understand that. However, to me, at least, working with it in the above fashion at least
renders it reasonably tractable, and manipulable. If anyone would care for me to actually
send them the 196 partitioned segments of that large data file, I would of course be only
too glad to do so. I look forward to hearing any thoughts, views, and suggestions,
whenever might possibly be convenient.
Sincerely,
ENGRPOD.

Dave Robinson
Dave Robinson's picture
Offline
Joined: 2010-04-29
Posts: 196
Glad you finally made it. It

Glad you finally made it. It was a long haul, but I hope you will enjoy processing the data that you have extracted.

Like you I found that the 1.9 GByte files provided by the SETIQuest team were too big to be tractable, and like you I decided to split the data up into usable block sizes. My first thought was simply to split it up into 1 second blocks, then any frequency domain processing I carried out would have a nice 1Hz resolution. However  giving it further thought I found that the 'strange' sample frequency used on the ATA which is a non integer number of Hz means that you can't get an integer number of samples into a 1 Second block. Thinking laterally; I decided that the Aliens wouldn't know what a second was; and so I really need not feel too constrained to do my work in periods of a second. As you know most FFT algorithms work best if fed with a data array which contain 2^n (where n is an integer) number of samples. So I arbitrarily decided that the unit of time I would use was the the largest 2^n number that was as close to 1 Second as possible. This turns out to be 2^23. So defining this new unit of time as a SETISecond (it actually about 0.96 seconds) It now provides a nice integer sample rate of 2^23 samples per SETISecond, no longer requiring you to think of something to do with the 1/3 sample you need to cope with if working in seconds.
 

In practice this turns out to be about the limit that many Maths packets can cope with before hitting the dreaded 'Out of Memory' error. It works for me - I commend this unit to the house;-)

Regards

Dave Robinson

ENGRPOD
Offline
Joined: 2011-06-29
Posts: 11
Some further elaboration on the hex data, with other thoughts...

Dear Dave:

I understand what you sent, and appreciate the thought; however, what I really need now is help to relate the hex I have managed to get into text ASCII format to the actual complex voltage values. Further, I have been unable to find out the interval between samples, as well as where the frequency information is located. I have virtually zero clue how the data is recorded; is one large 1.9Gbyte file a single frequency, or is it recorded at multiple frequencies? Also, where are the recording times located, and how does one relate those to the complex voltages? Further, how does one actually interpret the hex data?

Fine, I gather the hex data consists of signed complex voltage values; however, how many hex characters make up one sample, and given that I have actually provided the content of the start of one of the large 1.9 Gbyte files, how do I actually work backward to reconstruct those complex voltages in decimal format? That is the specific information I need, and which I am beseeching anyone, anywhere, within setiQuest, to please explain to me, or show me precisely where I can find that information? Mind you, I am obviously not asking, or expecing, anyone to literally lead me by the hand in the matter; I merely need to know, explicitly, where in the entire morass of setiQuest information, on the website, can I find such details, and actually relate those details to the specific content of that one file I partitioned, if only as a start?

Further, I have had some other thoughts as to the intention(s) of anyone actually sending information as an RF transmission. The presumption stereotypically depicted is that a species will infer that a logical approach would be to merely send the first few prime numbers, merely to announce their presence. Also, that such alien-generated information would necessarily be capable of being discerned using a purely Fourier approach, which, while I obviously understand the reason, strikes me as hopelessly naive at times, for a whole host of reasons.

In all of the literature "I have seen devoted to SETI, I have seem virtually no reference made to the existence, and/or use, of Walsh, Chrestenson, and/or Radamacher functions, all of which are related to the Hadamard matrices. Further, I quite honestly think that one of the most elemental methods whereby information could be sent could also be inherently mathematical, in terms of the virtually uniformly common fallacies all technologically sophisticated specieis would have had to wade through, in the course of their evolution. For example, I quite honestly think that an intelligent specieis might also possibly be inclined to send Euclidean geometric data; to include diagrams of simple points, lines, rays, intersecting lines, polygons, and the basic aspects of trigonometry.

Pierre Boulle, in his novel Planet of the Apes (which actually went far deeper than the film of the same name) actually brought up such a possibility, in his depiction of discussion between the human, and one of the apes; the discussion was begun using the universal fallacy of Euclidean geometry. Even taking the Ozma transmission, the famous one done sometime during the 1970s as a case in point, I entirely understand why the Aercibo dish would be depicted, then also a depiction of DNA, among other details of human life. However, I think, while well-intentioned, such an approach relies far too heavily on human thought contructs, hoping that a species intercepting the transmission would necessarily be able to infer the intended meaning of the symbology being transmitted.

I would likewise be most interested in any efforts to apply a purely information-theoretic approach to the entire problem, specifically, the concept of a point entropy source, and whether an RF transmission could, in fact, be directly analyzed to determine, on a purely bit level, whether it contains any information at all. I fully understand why an RF carrier is being searched for in SETI, from what jrseti had explained to me, and as I have seen pictorial examples of on the setiQuest website, using an ambiguity-function approach, to see if Doppler shift can, in fact, be discerned, so as to then be able to go back, and examine a given astronomical location more thoroughly. However, there has also been virtually no thought given to the concept of the potential use of nonsinusoidal transmissions, along the lines of the work of Prof. Henning Harmuth of Catholic Univ., who developed the sequency approach to analyze and characterize such signals. Then, too, from a quantum computing standpoint, I think it entirely possible that a qubit-like concept should at least be discussed, as well, which I have never yet seen mentioned anywhere in any serious SETI literature.

One other thought occurred to me; I realize that this might seem hopeflessly overly optimistic, however, consider one other thing, if you will: If you were an intelligent lifeform, as depicted in the film Contact, clearly inspired by the work and thoughts of Prof. Carl Sagan, it certainly strikes me as quite logical to expect that an alien race would not merely be sending out nonrandom transmissions in the hope of locating other life, although I obviously do not entirely discount such a possibility. What I do think an advanced alien race might well do would be to quite literally try to send specific technical instructions as to how to reach them, and on how to create the explicit technology needed to do so. I do not necessarily say that such a concept, which the film clearly sought ot depict, is necessarily right; however, I do think of it as yet another scneario certainly worth considering, at the very least.

I am also constrained to mention the novel Fountains of Paradise, by Arthur C. Clarke, due to the posing of the concept of a probe being specifically launched to seek out other life, and engage in deliberate trasmission to do so; I merely felt that such a novel concept, as expressed by Clarke, whom I found uniquely perceptive in his views as discussed in the novel 2001 (not the film version), to at least be seemingly worth a possible discussion here, as well. Your point about aliens not perceiving the concept of a second, however, I did find quite perceptive, as well as quite thought provoking; it did make me think a good deal, and was, in no small part, what prompted me to explain my foregoing thoughts, for whatever they might be worth as discussion points. I would look forward to hearing any thoughts and/or reactions, as well as (hopefully) any information anyone can provide as to how to interpret the SETI hex data I have now been able to finally get into text ASCII format.

Sincerely,

ENGRPOD.

Dave Robinson
Dave Robinson's picture
Offline
Joined: 2010-04-29
Posts: 196
Sorry for the delay in

Sorry for the delay in replying - I wasn't around yesterday. OK here is my understanding of the data format - maybe JRSETI or SIGBLIPS can put me right if I have misunderstood it.

Most sets of data are from a scan of the object lasting about 10 minutes; which are then broken into about 5 x 1.9GByte files, each of these files continues at the next sample from where the previous file left off, so that put together these 1.9 GByte files represent the complete record, without any missing data.

The data is sampled with an 8 bit ADC, so each hex character represents 1 sample. However note that the data acquisition is complex, so a complex pair is formed by the two adjacent hex values, thus in your stream you have
Sample 1     0xFF + 1i*0x08
Sample 2     0x0D + 1i*0x1A

etc.

The data stream has been taken with the ATA 'tuned' to one frequency (e.g. 1420MHz), and thus the average or DC value of the data stream you have represents the amplitude of the signal at this frequency. The actual frequency is given in the meta.txt file which is also provided along side the actual data files.- along with other data.which may or may not be of interest for your analysis.

The data sampling rate is weird - but no doubt there is very good reason it is set up this way, and is 8738133 and 1/3 s^-1 . Now you will know why I redefined it to be 2^23 samples per SETISecond.

Regarding your thoughts on processing the data streams; I think you have a blank white board to work with. You have arrived right at the begininning of the project, so virtually any thing you try will be original research. However be prepared for the shock you will get when you see the raw data - it is virtually nothing but pure Gaussian noise, so you will need to device some method of extracting the signal from the noise, else any Entropy based methodology will tell you thats all you have.

I am not ignoring your other fascinating points, but hope to respond later, when I have time to give them the thought that they deserve.

Regards

Dave Robinson

jrseti
jrseti's picture
Offline
Joined: 2010-07-22
Posts: 250
Dave, What you say is

Dave,

What you say is correct. You seem to understand the data well.

ENGRPOD's questions and confusions sparked an interest at SETI to better explain the setiData and make it more accessible to more people. So We have a new intern Kyle, whose screen name is "ksnodgrass". I'll ask his to become active on this Forum thread.

Kyle is a college student with the summer off and is eager to help. He knows nothing about the data, so i pointed him to everything we have so far, including documentation and this forum.

His project is to create a tutorial that will lead a person that known nothing about the data to understanding the data, and show how to load and use it in Octave. If he gets that done, we'll ask him to try to get the data loaded into several other programs such as MatLab and maybe the program from NIST ENGRPOD mentioned.

Kyle will be targeting the Windows operating system first.

-jrseti

-jrseti

herpy
Offline
Joined: 2010-08-27
Posts: 65
I made a program that does an

I made a program that does an 8738132 point fft on the data sets and then saves them to another file which you can then use to pull out sections of data to do whatever. I usually just make pictures but after reading some of your posts I made it output the power levels to a txt file.  It seems to work, and it might be what you are looking for.  There is a catch though.  After it converts the file they become 4 times bigger so if you just use one of the 1.9 gig chunks it ends up being around 8 gigs.  So if you combine multiple datafiles into one you get huge files.   The reason that it ends up making the files bigger is that after it does the fft on the complex 2byte pairs they get turned into float complex pairs which are 8 bytes.  After you convert the file you can then pull out small chunks which then get normalized and converted to power.

Here is a plot of some data with the columns averaged in a spreadsheet with the picture of the data pasted ontop.  
http://imageshack.us/f/809/ex1r.png/

I compiled it on windows xp using codeblocks since it seems like most people are using windows.
I just need a place to upload it if anyone wants to try it or I can post the messy source code on gist.

jrseti
jrseti's picture
Offline
Joined: 2010-07-22
Posts: 250
Awesome, I like the

Awesome, I like the picture.

Would you be willing to share the code and write a little tutorial on how you did this? If you are willing I could make a Wiki page for you to edit and for all of us to view.

 

Let me know.

-jrseti

-jrseti

jrseti
jrseti's picture
Offline
Joined: 2010-07-22
Posts: 250
ENGRPOD,   You

ENGRPOD,

 

You asked:

 

"what I need to know, using the foregoing as a case in point, is precisely how those complex-valued voltages can be sorted out from the hex content depicted above."

Maybe you already figured this out, but I'll try to answer it.

The voltages are complex (real, imaginary)  values, so the first byte in the file is the first real value, the second byte is the first imaginary value, and so on.

 

Hope that makes sense

-jrseti

-jrseti

jrseti
jrseti's picture
Offline
Joined: 2010-07-22
Posts: 250
setiData on Windows. Have you

setiData on Windows. Have you seen http://setiquest.org/wiki/index.php/Windows_setiData_tutorial? This tutorial was recently written and may be something we want to grow and improve.

Please have a look and ask questions about t so we can improve it for all.

-jrseti

Don_G
Don_G's picture
Offline
Joined: 2011-12-15
Posts: 26
setiData on Windows

> setiData on Windows.  Have you seen http://setiquest.org/wiki/index.php/Windows_setiData_tutorial?  This tutorial was recently written and may be something we want to grow and improve.

It seems that the download link provided in the "Windows setiQuest Data Tutorial" post ( http://sharesend.com/6a7vv ) is no longer available.  Are the target files available somewhere else?

Thank you,

-Don

ENGRPOD
Offline
Joined: 2011-06-29
Posts: 11
Some thoughts about the difficulties Don_G mentioned having...

December 20, 2011

To jrseti, Don_G, and all other SETI website users:

I saw the material Don_G mentioned about the dataset links apparently no longer working, as he described the problem. I have bene unable to be as involved with SETI as I had hoped initially on a purely amateur/hobbyist level, for a whole host of personal reasons; I merely wanted to also let everyone know I am very definitely still most eager to participate, and have very definitely not forgotten about trying to get mor einvolved. However, I will try to jump back into the fray of how to try to help make the SETI data more accessible, as rapidly as I am able, and am obviously curious whether Kyle was able to create the tutorial jrseti had mentioned Kyle would try to work on this past summer, I believe.

In any event, I will be trying as soon as I am able, to get back to trying to work with the SETI download data, as soon as I have some further time. I hope the concept I brought up as to how to partition the software, using the suggestions I had made in the past, were of some use, and would of course be most eager to know if anyone had actually tried the approach I had suggested, with any good effect, and/or positive results. As one further thought when such problems arise, perhaps jrseti and/or the ogher SETI staff could possibly arrange for some kind of phone bank, if you will, and/or possibly try to work with such open-source software groups as www.sourforge.net, in an effort to try to provide some means where would-be hobbyists and/or amateurs could possibly call more experienced users, whether at SETI itself, and/or elsewhere, to be able to help others like Don_G, myself, or still more amateurs and/or hobbyists, when such problems would arise; the ability to have such verbal (and/or webcam) discussions by amateurs and/or hobbyists, with those engaged in such efforts full-time, or others more experienced in using SETI data, could certainly, in my view, be exceedingly helpful, that is all I am saying.

Clearly, such a phone bank would have to be done on a purely volunteer basis, with specific hours such calls could be allowed cleary specified. Also, the possibility of doing such discussions periodically using, for example, Skype, for a webinar-based approach, if you will, was yet another possibility that had crossed my mind, as I read what Don_G had submitted. In any event,, have a happy holiday, everyone, and I look forwardot trying to jump back into the SETI amateur/hobbyist fray, if you will, as soon as I might possibly be able. Oh, and also, I was recently doing some experimentation on signal-processing approaches to analyzing a goodly number of digits of pi and e, in an effort to look for patterns, that I will likewise also try to present, as soon as I am able.

Nice holiday, one and all,

ENGRPOD.

ENGRPOD
Offline
Joined: 2011-06-29
Posts: 11
Some thoughts about the difficulties Don_G mentioned having...

December 20, 2011

To jrseti, Don_G, and all other SETI website users:

I saw the material Don_G mentioned about the dataset links apparently no longer working, as he described the problem. I have bene unable to be as involved with SETI as I had hoped initially on a purely amateur/hobbyist level, for a whole host of personal reasons; I merely wanted to also let everyone know I am very definitely still most eager to participate, and have very definitely not forgotten about trying to get mor einvolved. However, I will try to jump back into the fray of how to try to help make the SETI data more accessible, as rapidly as I am able, and am obviously curious whether Kyle was able to create the tutorial jrseti had mentioned Kyle would try to work on this past summer, I believe.

In any event, I will be trying as soon as I am able, to get back to trying to work with the SETI download data, as soon as I have some further time. I hope the concept I brought up as to how to partition the software, using the suggestions I had made in the past, were of some use, and would of course be most eager to know if anyone had actually tried the approach I had suggested, with any good effect, and/or positive results. As one further thought when such problems arise, perhaps jrseti and/or the ogher SETI staff could possibly arrange for some kind of phone bank, if you will, and/or possibly try to work with such open-source software groups as www.sourforge.net, in an effort to try to provide some means where would-be hobbyists and/or amateurs could possibly call more experienced users, whether at SETI itself, and/or elsewhere, to be able to help others like Don_G, myself, or still more amateurs and/or hobbyists, when such problems would arise; the ability to have such verbal (and/or webcam) discussions by amateurs and/or hobbyists, with those engaged in such efforts full-time, or others more experienced in using SETI data, could certainly, in my view, be exceedingly helpful, that is all I am saying.

Clearly, such a phone bank would have to be done on a purely volunteer basis, with specific hours such calls could be allowed cleary specified. Also, the possibility of doing such discussions periodically using, for example, Skype, for a webinar-based approach, if you will, was yet another possibility that had crossed my mind, as I read what Don_G had submitted. In any event,, have a happy holiday, everyone, and I look forwardot trying to jump back into the SETI amateur/hobbyist fray, if you will, as soon as I might possibly be able. Oh, and also, I was recently doing some experimentation on signal-processing approaches to analyzing a goodly number of digits of pi and e, in an effort to look for patterns, that I will likewise also try to present, as soon as I am able.

Nice holiday, one and all,

ENGRPOD.

Don_G
Don_G's picture
Offline
Joined: 2011-12-15
Posts: 26
The download link for the

The download link for the source files, provided in the "Windows setiQuest Data Tutorial" post ( http://sharesend.com/6a7vv ), is no longer available.  Are these files available somewhere else?

herpy
Offline
Joined: 2010-08-27
Posts: 65
http://sharesend.com/t4p0r I

http://sharesend.com/t4p0r

I put it up there again.  

I only ever used it on the full 1.9 gig data files and the combined data files.  It should work on less data but you get 1 pixel of height per around 20 megs of data I think.  If you use a 1.9 gig file it will generate a file around 8 gigs and then you use that 8 gig file to generate the pictures.  One of the full data sets is around 32 gigs so you would end up with files bigger than 100 gigs if you used the whole thing.

I haven't changed anything since I got it working even though it would seem that you could save the output from the fft as a 16 bit float instead of a 32 bit float but I am not sure about that.

Don_G
Don_G's picture
Offline
Joined: 2011-12-15
Posts: 26
Thank you very much

Thank you very much herpy.

Merry Christmas,

-Don