$)C hi, Daniel, I have also spent quite some time on the random engine of GATE(Geant4), so maybe I can give my suggestions. 1.  To solve your problem about the root file output, don't disable the ROOT output, instead, set all the related root output option (single, hits, coincidences, etc.) as 0, thus there will only be an empty gate.root file with a fixed size of 11kB. 2.  I have double about your method of modifying one number in the endOfRun.rndm file. I remember that there's a talk about this issue in this email list some time ago. Someone who has more knowledeg about the random generator engine of Geant4 said you cannot get statistically independent randoms by just modify the *.rndm file. So I suggest you be careful about your method. You can search the user talk and you can also find a small program which can generate a *.rndm file. Best regards Yuxuan below is the email and program by Angela Foudray, July 12, 2006: Hey Seyjoon, We've got a little program that writes out a file in the format that GATE is looking for.  It's not exactly pretty, but it does what you are looking for.  Here's a few things about it: 1) It always writes the output to a file called "test.rndm" (so copy this file to something else before running the program again) 2) It always ends in a segmentation fault, but its really done it's job (been too lazy to change this since it does what you are really looking for it to do - i.e. make a file with GATE random seed syntax) 3) You run it with the seed as a flag passed at the command line, so to use it you run: ./testRandom 31245 (exchange 31245 with the random seed of your choice).  And voila, a file with a new seed. Send all your praise and kisses to Garry Chinn (<a href=mailto:gchinn@stanford.edu>gchinn@stanford.edu</a>). Let me know if you have any questions. Ange Seyjoon wrote: hi   'GATE-USERS Talk' webpage doesn't open in opengatecollaboration site. I hope to search for a method of generating random seed file (.rndm) but it just shows blank page.   Can anyone tell me how to make it?   Regards, Seyjoon <hr> 180!F 4^6sAx >_HD! 8^@O 5N5U! <a href=http://us.rd.yahoo.com/mail_kr/taglines/mail4/*http://kr.content.mail.yahoo.com/cgland>;u 7N?n >_HD! 8^@O</a>?!<-4B 8^=CAv8& ?)1b@z1b 2x>n4Y 3u@; <v @V4d4O4Y. <tt> </tt> <hr><tt> _______________________________________________ Gate-users mailing list </tt><a href="mailto:Gate-users@lists.healthgrid.org"><tt>Gate-users@lists.healthgrid.org</tt></a><tt> </tt><a href="http://lists.healthgrid.org/mailman/listinfo/gate-users"><tt>http://lists.healthgrid.org/mailman/listinfo/gate-users</tt></a><tt>  </tt> <tt>_______________________________________________ gate-users mailing list gate-users@lphe1pet1.epfl.ch http://lphe1pet1.epfl.ch/mailman/listinfo/gate-users </tt> <table width=100%> <tr valign=top> <td> <td>"Daniel Woodsworth" <danieljw@phas.ubc.ca> Sent by: gate-users-bounces@lists.healthgrid.org 07/16/2007 06:30 PM <td>                To:        gate-users@lists.healthgrid.org         cc:                 Subject:        [Gate-users] /gate/output/root/setSaveRndmFlag</table> <tt>Hi, I am trying to set up a system to run statistically independent runs/simulations of a single physical setup (phantom, activity, geometry, scanner, electronics etc.) The way I am doing this, is I have a little script that takes an old endOfRun.rndm file and edits one of the numbers, which (I believe), when the simulation is run from a directory containing an endOfRun.rndm file causes Gate to produce a new (independent) stream of random numbers. So the script makes multiple, different, endOfRun.rndm files and launches a simulation from distinct directories. This all seems to work fine and yield independent simulations. However, if I try and disable the ROOT output (/gate/output/root/disable), then I no longer get independent runs. I am guessing this is since the setSaveRndmFlag is part of the /root tree structure. Ie the command /gate/output/root/disable kills any commands which start with /gate/output/root, including setSaveRndmFlag. At any rate, somehow disabling the ROOT output disrupts the GATE procedure of using the endOfRun.rndm file as a starting point for the random number generator. I would like to be able to run simulations with just interfile output, as sometimes I only need projection data, and the root output slows down the simulation and tends to be very large data files. Has anyone come across this? Does anyone know how to force GATE to use endOfRun.rndm as the starting point for the random number stream, with the ROOT output disabled? Thanks for any help. Daniel _______________________________________________ Gate-users mailing list Gate-users@lists.healthgrid.org http://lists.healthgrid.org/mailman/listinfo/gate-users </tt>