[Seisunix] seg2segy

[Maarten Vanneste]

I am hereby forwarding an email from a student at ICG who experienced problems with the third-party code seg2segy distributed with SeismicUnix4.1. For one reason or the other, his email was not distributed to the list.

Dear all,

 

I have some seismic data in seg-2 format. I usually convert the data from seg-2 to segy using seg2segy an then I convert the segy data to su using segyread.

By doing this I do not have any error message, neither from seg2segy nor segyread and the headers are okay but I do not recover the signal properly.

I have 121 traces of 14s but after 7s the signal is set to zero and before the signal does not look good. Even the trace 121 which corresponds to the sweep

(10s) have not the good shape and is zeroed after 7s.

I read this data with an other software and things went fine, I correctly recovered the signal. Moreover, if I use this software to do the conversion from

seg-2 to segy and then use segyread, it works fine.

Any suggestions?

 

Best regards,

 

Guillaume Sauvin

ICG

With best regards / Vennlig hilsen
Maarten

Dr. Maarten Vanneste
Offshore Geotechnics Division
NGI, Oslo - Norway
www.ngi.no

See also: International Centre for Geohazards
www.geohazards.no

[Heigl, Werner]

Could the data set in question be made available?

[John Stockwell]

Hi

Simply telling us that something doesn't work doesn't tell us much.

-John

John Stockwell | jo...@dix.Mines.EDU
Center for Wave Phenomena (The Home of Seismic Un*x)
Colorado School of Mines
Golden, CO 80401 | http://www.cwp.mines.edu/cwpcodes
voice: (303) 273-3049

Our book:
Norman Bleistein, Jack K. Cohen, John W. Stockwell Jr., [2001],
Mathematics of multidimensional seismic imaging, migration, and inversion,
(Interdisciplinary Applied Mathematics, V. 13.), Springer-Verlag, New York.

_______________________________________________
seisunix mailing list
seis...@mailman.mines.edu
https://mailman.mines.edu/mailman/listinfo/seisunix
Unsubscribe: seisunix-u...@mailman.mines.edu

[Maarten Vanneste]

Hello again,

We are further looking into this issue and tested the conversion from seg2 to segy on different platforms. We noticed that the c-code compiled on a 64-bit linux platform gave erroneous results, whereas the same shot gather converted on cygwin 32-bit executable appeared to have worked just fine... No error messages appear however.

The shot gathers are about 6.5 MB size, so we will do some further testing before sending out the files.

With best regards / Vennlig hilsen
Maarten

[Heigl, Werner]

If the file size is an issue you could send a decimated set.

[Chand Shyam]

Hi Maarten

 

I had problems with SeisUnix when I installed on my 64 bit machine with Fedora Linux  but after tweeking with some compilation options while reinstalling SU it works fine. Probably you have the same kind of problem.

 

Regards

 

shyam

Dr. Shyam Chand
Geological Survey of Norway (NGU)
7491 Trondheim, NORWAY
Ph:(+47) 73904283 (Office);   73904000 (Reception);

     (+47) 41148186 (Mobile);    Fax: (+47) 73921620

http://www.ngu.no ;                http://www.mareano.no

http://www.uib.no/people/nglbh/GANS/index.html

http://www.uib.no/people/nglbh/GANS/people.html

[John Stockwell]

Yes. That came up a couple of years ago. SEG2SEGY does not apparently
work on 64 bit systems. The problem likely can be traced to the
use of "short" and "long" integers in the code, which do not translate
from 32 to 64 bit.

So, try putting in this line:

CFLAGS = -m32

above the LFLAGS line in the Makefile and recompile. See if
compiling it 32 bit fixes the problem.

> Hello again,
>
> We are further looking into this issue and tested the conversion from seg2
> to segy on different platforms. We noticed that the c-code compiled on a
> 64-bit linux platform gave erroneous results, whereas the same shot gather
> converted on cygwin 32-bit executable appeared to have worked just fine...
> No error messages appear however.
>
> The shot gathers are about 6.5 MB size, so we will do some further testing
> before sending out the files.
>
> With best regards / Vennlig hilsen
> Maarten
>
>
>
> On Thu, Jul 16, 2009 at 2:44 PM, Heigl, Werner <
> Werner...@usa.apachecorp.com> wrote:
>
>> Could the data set in question be made available?
>>
>>
>>

>> *From:* seisunix...@mailman.mines.edu [mailto:
>> seisunix...@mailman.mines.edu] *On Behalf Of *Maarten Vanneste
>> *Sent:* Thursday, July 16, 2009 4:48 AM
>> *To:* seisunix
>> *Cc:* guillaum...@free.fr
>> *Subject:* [Seisunix] seg2segy

John Stockwell | jo...@dix.Mines.EDU

[Maarten Vanneste]

Thanks for the response. That probably explains it as our linux system indeed runs on 64-bit. Will take a look at it after some holidays...

With best regards / Vennlig hilsen
Maarten

Dr. Maarten Vanneste
Offshore Geotechnics Division
NGI, Oslo - Norway
www.ngi.no

See also: International Centre for Geohazards
www.geohazards.no

[陈可洋]

Dear all:     I want to kown clearly the following codes' principles: sumigfx, sumigffd, sumigsplit, sumigpspi ...  please suggest me the papers they are refer to? Thank you very much for you help!     With best regards / keyang chen      2009-7-20

---

好玩贺卡等你发，邮箱贺卡全新上线！

[Jan Valenta]

Does the -m32 flag work for anybody? It works for me only if I make
everything 32 bit (the whole cwp stuff). If I recompile only the
seg2segy in 32-bit then the result is still wrong.

Is it possible to make only the seg2segy 32-bit while keeping the rest
64-bit?

Jan.

John Stockwell píše v Pá 17. 07. 2009 v 10:15 -0600: