On 27/06/2015 23:58, Steven Franke wrote: Hi Steve,
> Hi Joe - > Yes, I am comfortable with making wsprd_exp the official wsprd. It has been > working very well here. I had a couple of 16-decode cases again last night on > 20 meters. > > Say, I just happened to be sitting here working on tracking down the signal > dropouts. When you get a chance, would you please have a look at the images > linked below: > > https://dl.dropboxusercontent.com/u/33211132/375Hzdata.png > https://dl.dropboxusercontent.com/u/33211132/1500Hzdata.png > > The first one is absolute value of the complex 375 Hz data from a .c2 file > plotted as a 108x400 pixel image. The dropouts are clearly seen along to the > top of the image. > > The second one is the 1500 Hz c0 “common” data written from within the > writec2.f90 function. The second image is (108*4)x400 pixels - and shows that > the features have a 432-pt fundamental period (at 1500 Hz). > > I’m scratching my head over here trying to figure out how a pattern like this > gets produced. Right now I’m looking at the wspr_downsample function, and > specifically the lowpass filter function. Does this sound right to you? I’m > not clear on what the timf2 function is doing - do you think that the problem > could originate in there? That's some funky custom filtering going on there! One thing that looks wrong to me is that the variable 'nb' in wspr_downsample.f90 really ought to be initialized, I'd guess to '0'. Having said that a quick glance thought the code seems to imply that if 'nb' is zero then the whole divide weak and strong frequencies in timf2.f90 may not achieve anything. I may well be well of track here as my DSP knowledge is way short of this sort of custom filtering code :( > > Steve 73 Bill G4WJS. > > >> On Jun 27, 2015, at 5:46 PM, Joe Taylor <j...@princeton.edu> wrote: >> >> Hi Steve, >> >> Sorry to be slow in getting back to you. After my post about wsprd_exp >> I got involved in chasing a bug in the ISCAT decoder ... >> >> On 6/26/2015 12:31 PM, Steven Franke wrote: >>> I’m glad to see that you were able to confirm the improved performance >>> of the two-pass decoder. I’m guessing that your dataset includes a more >>> representative mixture of bands and conditions than the group of >>> 20m files that I used. Hence the smaller, but still significant, >>> increase in the number of decodes over the default wsprd. >> Probably so. I thought the increased number of decodes was very >> worthwhile, anyway. >> >>> I am surprised by your observation that the two-pass decoder is faster >>> than the default one. That’s not what I see here. Are you using your >>> wspr_timer.out times? Or some other measure of execution time? >>> The numbers that I reported were the “Total” times from wspr_timer.out. >> I ran both tests a couple of times, and I also used the “Total” times >> from wspr_timer.out. However, I was concentrating on decoder >> performance rather than timing, so my observation needs a more careful >> look before being taken very seriously. I hope to find time to look at >> it more thoroughly next week, and maybe see if any further optimizations >> are possible. >> >> Are you comfortable with making wsprd_exp the "official" wsprd now ? >> >> -- Joe ------------------------------------------------------------------------------ Monitor 25 network devices or servers for free with OpManager! OpManager is web-based network management software that monitors network devices and physical & virtual servers, alerts via email & sms for fault. Monitor 25 devices for free with no restriction. Download now http://ad.doubleclick.net/ddm/clk/292181274;119417398;o _______________________________________________ wsjt-devel mailing list wsjt-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wsjt-devel
