Re: [wsjt-devel] JT65 HF bands capacity, spectra allocation and software support

[Bill S. <g4...@cl...>]

Hi Igor,

some comments in line below.

On 18/02/2017 19:12, Игорь Ч wrote:
> There is simple answer why users prefer operate JT65 on the overloaded 
> HF bands while there is underused  JT9 subband.
I am not sure it is simple, there are many factors, here are some of them:

  * Many users persist with JT65-HF variants despite it not having JT9
    capability and a very old JT65 decoder tuned for EME that has had no
    improvements for many years. The driving forces are probably
    familiarity, inertia and to some extent stubbornness,
  * Many users have set goals based on available awards or simply self
    defined goals like HF DXCC in JT65, they will tend to shun JT9 until
    the JT65 based goals are achieved,
  * Large numbers of users do not set up rig control so dual mode
    JT9+JT65 is not practical for them, this is compounded by many older
    rigs with fixed crystal filters of less than 3kHz width,
  * The large amount of JT65 activity is self-propagating, many stations
    with JT9 capability will only call on JT9 when there is nothing
    un-worked from the JT65 callers,
  * Some users just prefer the musical tones rather than the somewhat
    annoying,  to the human ear, JT9 micro-warble,
  * ...

> .
> Anyone from DX location, let's say for instance from New Zealand, have 
> frequently met scenario where multiple signals coming back to his CQ call.
Agreed but until JT9 occupancy justifies, it doing the improvements for, 
multi-pass decoding in JT9 it is unlikely to be implemented, we must 
hope that until then operators are smart enough to spread out when a 
needed station is busy, after all it is not hard when there is plenty of 
space available.
> .
> Due to the nature of JT9 protocol&modulation neither one of two JT9 
> signals on the same frequency will be decoded in most cases, things 
> getting worse if there are more than two signals on the same 
> frequency. To make QSO possible experienced JT9 users call DX stations 
> with frequency split and DX station in response has to use split 
> operation or frequency hopping.
You can't argue your case both ways on this. It is a technical challenge 
to decoding overlapping signals but until band occupancy is much greater 
for JT9 simply spreading out solves this problem. For sure experience 
with both WSPR and JT65A show that multi-pass decoding can be 
implemented successfully for JT9 when necessary.
> .
> Now one could imagine if there is propagation open between Europe and 
> New Zealand and at least 10 operators calling one DX station using 
> frequency split, this way JT9 spectral benefit versus JT65 is almost 
> eliminated.
That doesn't scale, you assumption is only correct if every QSO involves 
a rare wanted station. The bulk of HF JT mode QSOs are relatively 
routine with a single station or maybe two replying to a single caller 
with a strong signal.
> .
> Multipass decoding will only work if at least one JT9 signal is 
> decoded on the frequency, otherway there is nothing to subtract.

Well obviously! If there are no overlapping signals then there is no 
need for the complexity and processing overhead of multi-pass decoding.

Given the fact that JT9 uses less bandwidth and is amenable to 
multi-pass decoding when band occupancy warrants it, there can be no 
argument that JT9 does not allow more QSOs per kHz then JT65A. This 
should not be a surprise, JT9 was designed to do a similar job to JT65 
without having to use the extra bandwidth required for EME above 144MHz, 
JT65 has to work with Doppler spreading and TRX instability which are 
not generally significant on HF.

The high occupancy of JT65A is more due to user factors than technical 
ones IMHO, making the pitch to play on bigger would just continue the 
illusion that JT65A is preferable. Have you considered that pressure on 
the space for JT65A on HF might be a good thing as it will entice more 
users to the superior JT9 mode or even to try other bands that are more 
challenging.

Having everyone following the herd to the band with the best propagation 
on HF where signals are rarely weak is almost defeating the point of 
weak signal modes. You predict race to higher powers above, that may 
happen and it will be very sad if it does. Let's hope that common sense 
prevails and operators get smarter and maybe even reduce power a few dB 
which is just as likely to fix the occupancy problems! There is surely 
plenty of resolution headroom available on HF JT modes judging by the 
bulk of signal reports I see exchanged.

73
Bill
G4WJS.