I have been examining the WSJT9.7 and WSJT-X 1.8 waterfalls running JT65b to try and pin down why some find one preferable to another. At the moment they are fed with the same 44.1kHz 16bit signals of 144Mhz sky (well, moon actually ) noise

The noise (a '-26' floor) in WSJT-X appears 'stretched' in the x-axis compared to the pointillist appearance of the noise ('-33') in WSJT. This may be the difference in display real estate; it is difficult to match the time and frequency display dimensions as a result of the differences in window sizing methods. A ruler shows WSJT-X to have maybe ~25% greater x dimension for the same bandwidth, and there is ~10% greater y dimension for the same time span. Even allowing for these differences, the WSJT-X x-axis appears to have additional 'smearing', which I initially thought to be possibly the product of some difference in averaging methods.

Looking deeper shows that the palette positioning in each is different: WSJT-X appeared to me to have a 3-colour range (dark blue, light blue and yellow); only a line-of-sight local EME signal eventually revealed red. WSJT appeared to have 4 colours (blue, green, red, white) but positioned at different levels . I do not know the relative amplitudes of palette colour positioning but it is possible that the increased colour granularity of WSJT is improving the subjective appearance, with WSJT-X using yellow across a wider range - thus joining and thus 'stretching' noise spikes of dissimilar amplitude.

In practice, after a little tweaking, I find the WSJT-X 3-colour waterfall reveals signals as well as WSJT, just not as 'precisely'. In addition, if an observer's judgement is skewed by report differences between the two, it can reinforce the impression that WSJT "shows weak signals better".

Finally, I have turned off DS (and renamed all the CALL3.txt files) and so far operation leads me to believe that WSJT-X in my system is sometimes decoding slightly better than WSJT. This anecdotal evidence is obviously not as compelling as Charlie's heroic efforts.

73
Pete G4PLZ

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
wsjt-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Reply via email to