Thanks, Claus. I'm up and running again. I didn't realize the the trace files had to be massaged into a new format so that iNSpect would understand them.
I had to take a second step after I ran the "traceFix.py" program on my wireless.tcl example. The trace file had lines that began with the character "M" that were being used, apparently, for two different purposes. I deleted the ones at the beginning of the trace file and then iNSpect was able to visualize the results. Hopefully, some day, the traces from NS will work with iNSpect without any manipulation. Thanks again, John On 9/27/07, Claus Christmann <[EMAIL PROTECTED]> wrote: > > > Hi John, > > iNSpect is a little picky on the trace files. That's at least what I > found.... > > Let me explain how I started my iNSpect experience (using wireless1.tcl > from the examples in ns) > > 1) make sure the you have > $ns_ node-config -macTrace ON > in your wireless1.tcl file. iNSpect needs the MAC trace. > > 2) run ns wireless1.tcl > > 3) FIX your trace file! > iNSpect requires the initial node position to be in the first few lines > of the trace file. Use a python program called traceFix.py > (http://toilers.mines.edu/Public/NsInspect) to do this. > > 4) create a matching .config file for iNSpect. The most important thing > seems to be to have the number of nodes right (and the size of your > terrain) > > 5) run "iNSpect -t wireless1.config wirelss1-outFix.tr" > > Another method is to create a viztrace file, using a perl script called > portInspect (http://www.geocities.com/vin_sdm/ns2-wireless.htm) > > This requires you to have a motion file for iNSpect. > If you have your own motion file you normally wouldn't use random > motion. In this case make sure that you use calcdest to create the GOD > information. Be aware that calcdest and iNSpect require the motion file > to be in different order: calcdest requires anti-chronological order, > iNSpect requires chronological order. > > I hope this create more insight than confusion... > > > Claus > > > > John Zavgren wrote: > > Greetings: > > > > I've been switching over to iNSPect from NAM. I built the tool and it > seems > > to work fine... it runs all the demos. But, whenever I try to visualize > a > > NAM trace file, it crashes and dumps core. > > It visualizes my mobility files without any problems... but it dies on > my > > trace files. NAM will display the trace files, but iNSect won't. (I > > activated the new-trace directive.) > > > > I'd like to use one of the very simple examples from the NS distribution > > (I'm using 2.31) to drive iNSpect. (I must be doing something > egregiously > > wrong.) > > > > If one of you out there in NS land, has done this simple experiment > > successfully, can you please cut and paste the command that you used in > an > > email and sent them to me? > > > > Thanks! > > > > -- > Claus Christmann > Graduate Research Assistant > > Georgia Institute of Technology > 270 Ferst Drive > Atlanta, GA 30332-0150 > > http://uav.ae.gatech.edu > > > -- John Zavgren 603-371-0513 (home) 603-801-2094 (cell) [EMAIL PROTECTED] Four out of three people have difficulty with fractions.
