Hi Jean Louis, > On Jul 9, 2015, at 2:33 PM, f5djl <[email protected]> wrote: > > Hi all , > > I believe these questions/remarks are more for Steve and Bill. As part of > testing/implementation of WSPR on VHF with a group of french hams > (1HDI,2MM,6ABJ,6BYJ) we currently test v5700 with a variety of TRX and > interfaces , always in non band-hopping mode . > > My questions are related to how is supposed to work the randomization of TX > slot process and how/when exactly is the TX table recalculated ? > > 1/ For example if you start the application on two machines carefully time > synchronized ( ntp sync < one second) with a TX% at 20 % at a few seconds > interval but within same 2 mn slot, is it expected that TX sequence will > be the same , ie the TX slots will be synchronous on the two machines ?
I think that you could get the same pseudo-random sequence if all machines all use the same OS, compiler, libraries, etc. > > I had a quickly look at the code which I believe is involved ie > create_tx_schedule (int pctx) and I saw srand(time(0)) being used , I > thought wrongly most probably that the intent was to randomize the table by > generating a seed based on delta in seconds to reference 1970 ? I must > admit , I did not fully understood the purpose of the series of if statement > on pctx and the process of filling of the tx[] array and I have some serious > doubt on the effect of the band Hoppng . The if statements assign the minimum number of transmissions for each coordinated band and the maximum number of successive transmissions. These were considered to be the important constraints for the band-hopping application. > > 2/ I understand that TX next or disable/enable TX have no impact on the > schedule, is my understanding correct ? Correct. > > 3/ The table is calculated in next_tx_state function I believe at start of > application , then at the top of the hour ? BTW I saw a double TX > transmission at xx:58 and XX+1:00 today) Correct. > > Thanks for your patience to explain and educate, understanding this > behavior is important to us in a small network of VHF stations (<10) as we > are trying to minimize the chance to get two stations TX simultaneously and > then long period of simultaneous silence. So far I must admit we got some > rather unexpected results as the machines are much more than expected > simultaneously transmitting/receiving when at same Tx percentage . I think that that simply change the srand argument so that each user gets a different seed should fix the problem. srand(time(0)) -> srand(time(0) + getpid()) or something like this. As I was finishing this, Mike’s response just came in. It appears that we agree on what should be done, and he’s way ahead of me in terms of how to do it… Steve k9an > > Best regards > > Jean Louis > > > ------------------------------------------------------------------------------ > Don't Limit Your Business. Reach for the Cloud. > GigeNET's Cloud Solutions provide you with the tools and support that > you need to offload your IT needs and focus on growing your business. > Configured For All Businesses. Start Your Cloud Today. > https://www.gigenetcloud.com/ > _______________________________________________ > wsjt-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/wsjt-devel ------------------------------------------------------------------------------ Don't Limit Your Business. Reach for the Cloud. GigeNET's Cloud Solutions provide you with the tools and support that you need to offload your IT needs and focus on growing your business. Configured For All Businesses. Start Your Cloud Today. https://www.gigenetcloud.com/ _______________________________________________ wsjt-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/wsjt-devel
