Topband: FT8 Observations

[K4SAV]

I'm not sure how many people have actually compared the new digital 
modes with CW as far as low signal level decoding.  I did that for a 
long time when JT-65 first became available.  Back then it was much 
easier to separate one station and compare the reported S/N to what I 
see on my receiver when using a very narrow passband.  There was always 
a huge disparity, usually 30 to 40 dB.  Most of those numbers were taken 
right off the main screen of a TS-990s, so the accuracy may not be a lot 
but it's a pretty good indication of how close the signal is to the 
noise floor.

Then I found this:
http://www.arrl.org/forum/categories/view/31/page:2
You have to sign in to the ARRL site and look for the article titled 
"JT65, JT9, FT8, SNR explained".

That article says that the S/N reported by JT-65 is actually 29.7 dB 
more than it should be and JT-9 is 31.6 dB more than it should be. That 
agrees pretty closely with what I have been observing although my 
measurement show a slightly larger difference than that, but that could 
be because my measurements didn't have enough accuracy.

I used to try to identify a signal at that was close to the noise floor 
and see if JT-9 would decode it.  It never did.  At a level where JT-9 
does decode the signal, it would have been easy copy on CW.  So for me, 
I see no low level signal advantage to these digital modes.  I continue 
to wonder why other people say there is.  I wonder if others are using a 
wide passband when making comparisons (if they really do make 
comparisons).  For low level CW I usually use 150 Hz, sometimes a little 
less if there is QRM.

Jerry, K4SAV