Re: [wsjt-devel] WSJT-X vs JTDX sensitivity comparison

[Jim Brown via wsjt-devel]

On 9/18/2021 10:08 AM, Joe Taylor via wsjt-devel wrote:

It turns out that Mike, W3IP, was using WSJT-X for FT8 with the
"Decode | Fast" setting.

With nearly any modern computer one should use "Decode | Deep" for 
maximum FT8 sensitivity.


Was in the shack today, and checked WSJT-X setting and it was, indeed, 
set to Fast. I'm running the latest Beta, computer is pretty fast with 
lots of RAM. Don't know if it came that way, but I'm unlikely to have 
not set it to Deep.


I'll pass this along on the Slack VHF site.

73, Jim K9YC



___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX sensitivity comparison

[Joe Taylor via wsjt-devel]

Hi all,

In case you had wondered about the following post from K1HTV:

On 9/14/2021 1:55 PM, Rich - K1HTV via wsjt-devel wrote:
After this past weekend's ARRL VHF contest, one of the top SOHP scorers, 
Mike, W3IP, posted a note to the PVRC reflector along with the breakdown 
of his score. In his comments he compared the FT8 sensitivity of WSJT-X 
vs JTDX.  Mike is a technically sharp guy and I trust what he has 
reported. Here is what W3IP wrote:


*"When operating on FT8, I ran WSJT-X and JTDX in parallel. The JTDX 
decode capability on weak signals is significantly better - but JTDX 
doesn't recognize contest mode. I had several contacts that decoded only 
on JTDX so I had to manually advance the transmit side exchange on 
WSJT-X each time. I had only one occasion when I decoded a wanted signal 
in WSJT-X only. What a pain!"*


I wonder what might explain the better performance of JTDX over WSJT-X 
on weak signals and if something can be done so WSJT-X can match the 
weak signal performance of JTDX.


I used WSJT-X v2.5.0-rc5 in the VHF contest on 6M & 2M and it performed 
well for me on FT8. I haven't yet compared WSJT-X at my QTH with JTDX.


It turns out that Mike, W3IP, was using WSJT-X for FT8 with the
"Decode | Fast" setting.

With nearly any modern computer one should use "Decode | Deep" for 
maximum FT8 sensitivity.


-- 73, Joe, K1JT


___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX sensitivity comparison

[Roeland Jansen via wsjt-devel]

then, like always -- use the right tool for the right job?

On Wed, Sep 15, 2021 at 11:12 AM Jim Brown via wsjt-devel <
wsjt-devel@lists.sourceforge.net> wrote:

> On 9/14/2021 11:43 PM, Laurie, VK3AMA via wsjt-devel wrote:
> > That has been my experience as well. On the surface JTDX offers a
> > greater number of decodes, but many were false decodes.
>
> If I made the QSO and it shows up on LOTW, it wasn't a false decode!
> And, BTW, WSJT-X is not without false decodes. In my experience, a lot
> more than JTDX.
>
> 73, Jim K9YC
>
>
>
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX sensitivity comparison

[Jim Brown via wsjt-devel]

On 9/14/2021 11:43 PM, Laurie, VK3AMA via wsjt-devel wrote:
That has been my experience as well. On the surface JTDX offers a 
greater number of decodes, but many were false decodes.


If I made the QSO and it shows up on LOTW, it wasn't a false decode! 
And, BTW, WSJT-X is not without false decodes. In my experience, a lot 
more than JTDX.


73, Jim K9YC



___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX sensitivity comparison

[Jim Brown via wsjt-devel]

On 9/14/2021 10:55 AM, Rich - K1HTV via wsjt-devel wrote:
*"When operating on FT8, I ran WSJT-X and JTDX in parallel. The JTDX 
decode capability on weak signals is significantly better - but JTDX 
doesn't recognize contest mode. I had several contacts that decoded only 
on JTDX so I had to manually advance the transmit side exchange on 
WSJT-X each time. I had only one occasion when I decoded a wanted signal 
in WSJT-X only. What a pain!"*


More than a few fine 6M ops hanging out on the Slack VHF channel have 
done the same experiment, with the same result. I did it early in the 
2000 season on the advice of N6ML, and again in 2021. Both years, by 
late May I was using JTDX exclusively for FT8, and ran WSJT-X only for 
modes that JTDX does not support. JTDX has other virtues and faults, but 
if you want to work weak signals on FT8, it's the clear choice.


BTW -- I also reported that result here last year.

73, Jim K9YC



___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX sensitivity comparison

[Laurie, VK3AMA via wsjt-devel]

On 15/09/2021 4:27 pm, Saku via wsjt-devel wrote:
f wsjt-x decode is set to "Normal" or "Deep" I would rather say 
"slightly better, in some cases".


That has been my experience as well. On the surface JTDX offers a 
greater number of decodes, but many were false decodes.


If JTDX is truly offering superior decoding have any patches been 
submitted to the wsjtx dev-team by the jtdx people?


de Laurie VK3AMA



___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX sensitivity comparison

[Saku via wsjt-devel]

HI !

My opinion:

I have also tested this some time ago running both programs in parallel 
with same linux PC, same rig IC7300, OCF dipole 80-10m.
If wsjt-x decode is set to "Normal" or "Deep" I would rather say 
"slightly better, in some cases".

But difference is so small that it did not cause any need to move to JTDX.

Rich - K1HTV via wsjt-devel kirjoitti 14.9.2021 klo 20.55:

*The JTDX decode capability on weak signals is significantly better *


--
Saku
OH1KH

___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

[wsjt-devel] WSJT-X vs JTDX sensitivity comparison

[Rich - K1HTV via wsjt-devel]

After this past weekend's ARRL VHF contest, one of the top SOHP scorers,
Mike, W3IP, posted a note to the PVRC reflector along with the breakdown of
his score. In his comments he compared the FT8 sensitivity of WSJT-X vs
JTDX.  Mike is a technically sharp guy and I trust what he has reported.
Here is what W3IP wrote:

*"When operating on FT8, I ran WSJT-X and JTDX in parallel. The JTDX decode
capability on weak signals is significantly better - but JTDX doesn't
recognize contest mode. I had several contacts that decoded only on JTDX so
I had to manually advance the transmit side exchange on WSJT-X each time. I
had only one occasion when I decoded a wanted signal in WSJT-X only. What a
pain!"*

I wonder what might explain the better performance of JTDX over WSJT-X on
weak signals and if something can be done so WSJT-X can match the weak
signal performance of JTDX.

I used WSJT-X v2.5.0-rc5 in the VHF contest on 6M & 2M and it performed
well for me on FT8. I haven't yet compared WSJT-X at my QTH with JTDX.

73,
Rich - K1HTV
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX

[Black Michael via wsjt-devel]

My CALL3.TXT was 0-size so no hinting was done.But I do want to get the test 
set that was used B4 for determining the traedeoff for speed and false decode 
rate and picking the current settings used in WSJT-X.
de Mike W9MDB
  From: Neil Zampella <ne...@techie.com>
 To: wsjt-devel@lists.sourceforge.net 
 Sent: Sunday, June 11, 2017 11:47 PM
 Subject: Re: [wsjt-devel] WSJT-X vs JTDX
   
 Michael, have you looked at the hint code that was added ... its a bunch of  
'if then' decision trees, which can explain the 'extra processing'.  
  As a user, I'd rather trust the 13K WSJT-X decodes to be REAL, rather than 
the 15K of which 2 or 3K may be spurious.
 
 Neil 
 KN3ILZ
 
  
 On 6/11/2017 3:49 PM, Black Michael wrote:
  
  I see why people are claiming JTDX decodes better. 
  I ran a 24-hour test side-by-side on the same audio stream and got the 
following numbers. 
  JTDX average dB on matched QSOs is ~1dB better (not a big deal) 
  JTDX got 15,690  decodes WSJTX got 13,692 decodes 
  JTDX got 2,144 decodes that  WSJT-X did not WSJTX got 146 decodes that JTDX 
did not. 
  So JTDX got ~15% more decodes...nothing to sneeze at.  Can't say I really see 
many false decodes but I guess really can't know for sure. 
  Here's a good example of a noticeable difference in processing this wav file. 
  
  https://www.dropbox.com/s/zzddsaqop94bfc9/170504_1159.wav?dl=1
  
  I minimized the values in the Advanced tab for JTDX but it still gets 12 
decodes and WSJT-X gets 7. It takes a bit over 5 seconds for JTDX to do its 
thing and a bit over 2 seconds for WSJTX. So there is some extra processing 
going on for sure.  Either in JTDX's "subpass 1" or "subpass 2"? Are sure we 
don't want to borrow something back from JTDX? We had a set of test files 
before that were used for timing and getting some params for false alarms and 
such. Perhaps we should run those again and compare? 
  
 
  
 
  
  
  
  --- 
 Michael D. Black  
 
 --
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
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel


   --
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
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX

[ea2ekh]

Could you send me a randomly picked up subset of the JTDX exclusive decodes 
together with a time stamp?  I have a live data feed from pskreporter for a 
visualization project and I can verify wether other stations received it and 
what they were running.

 I've got mostly  solid data for the last 30 days (except for a couple of 
problems with my data collection script)

For the data inclined, data is being fed to a Elasticsesrch cluster.


73,


Borja - EA2EKH




--
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
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX

[Wolfgang]

Title: Re: [wsjt-devel] WSJT-X vs JTDX



and there is another issue with those decoded 'very weak signals' on the HF bands!

In Europe we have different countries, different mentality, some like it soft, 
some like it 'hot'...

The huge 'CALL3.TXT' file is an essential part of the hinted decode of JTDX.
 
Given that there is a weak signal on the band, as JTDX claims to decode it by
hinted decode, maybe it even looks like a rare DX, some stations want to make 
sure and trying to work those. 

So, what's happening: hearing a weak signal = let's try with more power! 
And this is not the idea behind JT65 & JT9.


73 de Wolfgang
OE1MWW

--
Amateur radio is the most expensive type of free-of-charge communication!
Amateurfunk ist die teuerste Art der kostenlosen Kommunikation!


--
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
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] WSJT-X vs JTDX

[Neil Zampella]

Michael,

have you looked at the hint code that was added ... its a bunch of  'if 
then' decision trees, which can explain the 'extra processing'.


As a user, I'd rather trust the 13K WSJT-X decodes to be REAL, rather 
than the 15K of which 2 or 3K may be spurious.


Neil
KN3ILZ


On 6/11/2017 3:49 PM, Black Michael wrote:

I see why people are claiming JTDX decodes better.

I ran a 24-hour test side-by-side on the same audio stream and got the 
following numbers.


JTDX average dB on matched QSOs is ~1dB better (not a big deal)

JTDX got 15,690  decodes
WSJTX got 13,692 decodes

JTDX got 2,144 decodes that  WSJT-X did not
WSJTX got 146 decodes that JTDX did not.

So JTDX got ~15% more decodes...nothing to sneeze at.  Can't say I 
really see many false decodes but I guess really can't know for sure.


Here's a good example of a noticeable difference in processing this 
wav file.


https://www.dropbox.com/s/zzddsaqop94bfc9/170504_1159.wav?dl=1 



I minimized the values in the Advanced tab for JTDX but it still gets 
12 decodes and WSJT-X gets 7.
It takes a bit over 5 seconds for JTDX to do its thing and a bit over 
2 seconds for WSJTX.
So there is some extra processing going on for sure.  Either in JTDX's 
"subpass 1" or "subpass 2"?

Are sure we don't want to borrow something back from JTDX?
We had a set of test files before that were used for timing and 
getting some params for false alarms and such.

Perhaps we should run those again and compare?

Inline image

Inline image




---
Michael D. Black


--
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
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel