4.48 vs 4.51 CW Decode

[Jim Rogers, W4ATK]

I am saving a copy of 4.48 so I can go back to it.  

IMHO 4.51 is:
        1) More difficult to use
        2) Does not perform as well as 4.48
        3) Lacks the weak signal performance of 4.48
        4) Requires constant adjustment of the RF Gain control

This was  the absolute best CW decoder I had ever used. I am still using 4.51 but have not yet hit the sweet spot with it. Perhaps others are having better results and will share AGC settings, threshold setting etc.

I am can copy call signs, signal reports at elevated speeds, but alas I am CW challenged and am using CW decode to help me get over the 18 to 20 wpm block.


Jim, W4ATK
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

I agree.  I just installed 4.51 yesterday, but it doesn't seem to work as well as 4.48, even using the same settings for slope and threshold.  I will experiment more before I give up on 4.51, but I will probably revert back to the previous version.
Art  WB8ENE

---------- Original Message ----------
From: W4ATK <[hidden email]>
To: Elecraft Reflector <[hidden email]>
Subject: [Elecraft] 4.48 vs 4.51 CW Decode
Date: Fri, 29 Jun 2012 10:52:26 -0500

I am saving a copy of 4.48 so I can go back to it.  

IMHO 4.51 is:
1) More difficult to use
2) Does not perform as well as 4.48
3) Lacks the weak signal performance of 4.48
4) Requires constant adjustment of the RF Gain control

This was  the absolute best CW decoder I had ever used. I am still using 4.51 but have not yet hit the sweet spot with it. Perhaps others are having better results and will share AGC settings, threshold setting etc.

I am can copy call signs, signal reports at elevated speeds, but alas I am CW challenged and am using CW decode to help me get over the 18 to 20 wpm block.


Jim, W4ATK
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html


____________________________________________________________
Fast, Secure, NetZero 4G Mobile Broadband. Try it.
http://www.netzero.net/?refcd=NZINTISP0512T4GOUT2
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[wayne burdick]

CW decode can and will be improved in the next release after 4.51. But  
we have to get 4.51 into production status because of the AGC  
improvements. CW decode requires its own AGC, in effect, so that the  
user can take advantage of the wider dynamic range of the IF/AF AGC.

73,
Wayne
N6KR

On Jun 29, 2012, at 8:52 AM, W4ATK wrote:

> I am saving a copy of 4.48 so I can go back to it.
>
> IMHO 4.51 is:
> 1) More difficult to use
> 2) Does not perform as well as 4.48
> 3) Lacks the weak signal performance of 4.48
> 4) Requires constant adjustment of the RF Gain control
>
> This was  the absolute best CW decoder I had ever used. I am still  
> using 4.51 but have not yet hit the sweet spot with it. Perhaps  
> others are having better results and will share AGC settings,  
> threshold setting etc.
>
> I am can copy call signs, signal reports at elevated speeds, but  
> alas I am CW challenged and am using CW decode to help me get over  
> the 18 to 20 wpm block.
>
>
> Jim, W4ATK
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Thomas Horsten]

Hi Wayne,

Since the effects of 4.51 have come up again, I'm wondering if you got
around to investigating the (to my ears negative) effect on APF that it
brought on? I seem to remember you were going to do some measurements, but
I can't remember having seen the results of those.

73, Thomas M0TRN

On 29 June 2012 17:14, Wayne Burdick <[hidden email]> wrote:

> CW decode can and will be improved in the next release after 4.51. But
> we have to get 4.51 into production status because of the AGC
> improvements. CW decode requires its own AGC, in effect, so that the
> user can take advantage of the wider dynamic range of the IF/AF AGC.
>
> 73,
> Wayne
> N6KR
>
> On Jun 29, 2012, at 8:52 AM, W4ATK wrote:
>
> > I am saving a copy of 4.48 so I can go back to it.
> >
> > IMHO 4.51 is:
> >       1) More difficult to use
> >       2) Does not perform as well as 4.48
> >       3) Lacks the weak signal performance of 4.48
> >       4) Requires constant adjustment of the RF Gain control
> >
> > This was  the absolute best CW decoder I had ever used. I am still
> > using 4.51 but have not yet hit the sweet spot with it. Perhaps
> > others are having better results and will share AGC settings,
> > threshold setting etc.
> >
> > I am can copy call signs, signal reports at elevated speeds, but
> > alas I am CW challenged and am using CW decode to help me get over
> > the 18 to 20 wpm block.
> >
> >
> > Jim, W4ATK
> > ______________________________________________________________
> > Elecraft mailing list
> > Home: http://mailman.qth.net/mailman/listinfo/elecraft
> > Help: http://mailman.qth.net/mmfaq.htm
> > Post: mailto:[hidden email]
> >
> > This list hosted by: http://www.qsl.net
> > Please help support this email list: http://www.qsl.net/donate.html
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Bill K9YEQ]

I have sent in my FT radio so I cannot give you my settings.  I find the
latest update for CW decode to be the best yet.  I do narrow my filter.  I
also use the decode to improve my copy speed and was easily in the 20WPM +
speed.  I send around 15+ with the KXPD3.  

73,
Bill
K9YEQ


-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of
[hidden email]
Sent: Friday, June 29, 2012 11:01 AM
To: [hidden email]
Cc: [hidden email]
Subject: Re: [Elecraft] 4.48 vs 4.51 CW Decode

I agree.  I just installed 4.51 yesterday, but it doesn't seem to work as
well as 4.48, even using the same settings for slope and threshold.  I will
experiment more before I give up on 4.51, but I will probably revert back to
the previous version.
Art  WB8ENE

---------- Original Message ----------
From: W4ATK <[hidden email]>
To: Elecraft Reflector <[hidden email]>
Subject: [Elecraft] 4.48 vs 4.51 CW Decode
Date: Fri, 29 Jun 2012 10:52:26 -0500

I am saving a copy of 4.48 so I can go back to it.  

IMHO 4.51 is:
1) More difficult to use
2) Does not perform as well as 4.48
3) Lacks the weak signal performance of 4.48
4) Requires constant adjustment of the RF Gain control

This was  the absolute best CW decoder I had ever used. I am still using
4.51 but have not yet hit the sweet spot with it. Perhaps others are having
better results and will share AGC settings, threshold setting etc.

I am can copy call signs, signal reports at elevated speeds, but alas I am
CW challenged and am using CW decode to help me get over the 18 to 20 wpm
block.


Jim, W4ATK
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html


____________________________________________________________
Fast, Secure, NetZero 4G Mobile Broadband. Try it.
http://www.netzero.net/?refcd=NZINTISP0512T4GOUT2
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[wayne burdick]

I listened hard to APF with 4.51 and can't tell any difference. OTOH  
I'm using fairly low threshold and slope settings (the defaults).

Wayne

On Jun 29, 2012, at 9:21 AM, Thomas Horsten wrote:

> Hi Wayne,
>
> Since the effects of 4.51 have come up again, I'm wondering if you  
> got around to investigating the (to my ears negative) effect on APF  
> that it brought on? I seem to remember you were going to do some  
> measurements, but I can't remember having seen the results of those.
>
> 73, Thomas M0TRN
>
> On 29 June 2012 17:14, Wayne Burdick <[hidden email]> wrote:
> CW decode can and will be improved in the next release after 4.51. But
> we have to get 4.51 into production status because of the AGC
> improvements. CW decode requires its own AGC, in effect, so that the
> user can take advantage of the wider dynamic range of the IF/AF AGC.
>
> 73,
> Wayne
> N6KR
>
> On Jun 29, 2012, at 8:52 AM, W4ATK wrote:
>
> > I am saving a copy of 4.48 so I can go back to it.
> >
> > IMHO 4.51 is:
> >       1) More difficult to use
> >       2) Does not perform as well as 4.48
> >       3) Lacks the weak signal performance of 4.48
> >       4) Requires constant adjustment of the RF Gain control
> >
> > This was  the absolute best CW decoder I had ever used. I am still
> > using 4.51 but have not yet hit the sweet spot with it. Perhaps
> > others are having better results and will share AGC settings,
> > threshold setting etc.
> >
> > I am can copy call signs, signal reports at elevated speeds, but
> > alas I am CW challenged and am using CW decode to help me get over
> > the 18 to 20 wpm block.
> >
> >
> > Jim, W4ATK
> > ______________________________________________________________
> > Elecraft mailing list
> > Home: http://mailman.qth.net/mailman/listinfo/elecraft
> > Help: http://mailman.qth.net/mmfaq.htm
> > Post: mailto:[hidden email]
> >
> > This list hosted by: http://www.qsl.net
> > Please help support this email list: http://www.qsl.net/donate.html
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[W5RDW]

When I first installed 4.51, I had to fiddle around with a few adjustments for the decode function to work as before, primarily adjusting THR lower. It seemed 4.48 liked a THR of 7, but now 4.51 seems to like a THR of 3. I have the AGC THC at 10 or so. Also, I always use the SPOT function to get the received signal in the proper place to decode.

It will decode any speed that I have encountered if the sending CW is computer generated (it gobbles up the high speed W1AW code practice!) or the CW Op has a very good fist. As the sending CW Op gets more sloppy, the decode function starts to suffer, as I would imagine most decoders would do. Some Ops do not leave a space between characters, so no wonder a decoder has trouble. It loves the contests, where I imagine many are using a keyboard or memory functions to send calls, exchanges, etc.

[Hisashi T Fujinaka]

On Fri, 29 Jun 2012, W5RDW wrote:

> When I first installed 4.51, I had to fiddle around with a few adjustments
> for the decode function to work as before, primarily adjusting THR lower. It
> seemed 4.48 liked a THR of 7, but now 4.51 seems to like a THR of 3. I have
> the AGC THC at 10 or so. Also, I always use the SPOT function to get the
> received signal in the proper place to decode.

Does anyone have any other pointers? I never have been able to get the
decoder to do much for me, even during a contest.

--
Hisashi T Fujinaka - [hidden email]
BSEE(6/86) + BSChem(3/95) + BAEnglish(8/95) + MSCS(8/03) + $2.50 = latte
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Jim Rogers, W4ATK]

Thanks Wayne, that is good news.

73s, Jim
On Jun 29, 2012, at 11:14 AM, Wayne Burdick wrote:

> CW decode can and will be improved in the next release after 4.51. But we have to get 4.51 into production status because of the AGC improvements. CW decode requires its own AGC, in effect, so that the user can take advantage of the wider dynamic range of the IF/AF AGC.
>
> 73,
> Wayne
> N6KR
>
> On Jun 29, 2012, at 8:52 AM, W4ATK wrote:
>
>> I am saving a copy of 4.48 so I can go back to it.
>>
>> IMHO 4.51 is:
>> 1) More difficult to use
>> 2) Does not perform as well as 4.48
>> 3) Lacks the weak signal performance of 4.48
>> 4) Requires constant adjustment of the RF Gain control
>>
>> This was  the absolute best CW decoder I had ever used. I am still using 4.51 but have not yet hit the sweet spot with it. Perhaps others are having better results and will share AGC settings, threshold setting etc.
>>
>> I am can copy call signs, signal reports at elevated speeds, but alas I am CW challenged and am using CW decode to help me get over the 18 to 20 wpm block.
>>
>>
>> Jim, W4ATK
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Guy, K2AV]

Perhaps if you are only talking about the CW decoder, but in general my
experience with 4.51 is utterly dead opposite of yours.  It *DOES* change a
lot of sweet spots, but once adjusted for those, the improvement in all
areas is rather extraordinary.  Again, the sweet spots for a lot of things
seemed to have moved, but given the changes in AGC, that's not really
surprising.  Using the second receiver in diversity, 4.51 improves signals
near or at the AGC intercept, allowing a truer separation of those.  We
tested this thoroughly at Field Day, and all who used the K3 with 4.51
remarked on how good the diversity was.

Weak signal performance with 4.51 was clearly much better by ear if one was
using an AGC with lower numbers in the settings.

73, Guy.

On Fri, Jun 29, 2012 at 11:52 AM, W4ATK <[hidden email]> wrote:

> I am saving a copy of 4.48 so I can go back to it.
>
> IMHO 4.51 is:
>        1) More difficult to use
>        2) Does not perform as well as 4.48
>        3) Lacks the weak signal performance of 4.48
>        4) Requires constant adjustment of the RF Gain control
>
> This was  the absolute best CW decoder I had ever used. I am still using
> 4.51 but have not yet hit the sweet spot with it. Perhaps others are having
> better results and will share AGC settings, threshold setting etc.
>
> I am can copy call signs, signal reports at elevated speeds, but alas I am
> CW challenged and am using CW decode to help me get over the 18 to 20 wpm
> block.
>
>
> Jim, W4ATK
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[mzilmer]

Contests aren't the best setting for CW decode, imho.  However, if you tune the CW slightly higher than the CWT indicates it seems to improve decoding when there is band noise present.  If you minimize the passband to 50 or 100 Hz, this also helps reduce noise in the decode channel.  With the passband minimized, the decoder gets good results with the CW and CWT matching.

Matt Zilmer
Consultant - Product Management Dept.
Magellan Navigation / MiTAC Digital Corp.
Tel: [phone]
Cell: [phone]
Si fractum non sit, noli id reficere


-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf Of Hisashi T Fujinaka
Sent: Friday, June 29, 2012 10:47 AM
To: W5RDW
Cc: [hidden email]
Subject: Re: [Elecraft] 4.48 vs 4.51 CW Decode

On Fri, 29 Jun 2012, W5RDW wrote:

> When I first installed 4.51, I had to fiddle around with a few
> adjustments for the decode function to work as before, primarily
> adjusting THR lower. It seemed 4.48 liked a THR of 7, but now 4.51
> seems to like a THR of 3. I have the AGC THC at 10 or so. Also, I
> always use the SPOT function to get the received signal in the proper place to decode.

Does anyone have any other pointers? I never have been able to get the decoder to do much for me, even during a contest.

--
Hisashi T Fujinaka - [hidden email]
BSEE(6/86) + BSChem(3/95) + BAEnglish(8/95) + MSCS(8/03) + $2.50 = latte ______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[wayne burdick]

Lyle and I have discussed this, and we're certain that we can have the  
best of both AGC and CW decoding with some further changes to  
firmware. We'll be looking into it when I get back from vacation.

Wayne

On Jun 29, 2012, at 11:15 AM, Guy Olinger K2AV wrote:

> Perhaps if you are only talking about the CW decoder, but in general  
> my
> experience with 4.51 is utterly dead opposite of yours.  It *DOES*  
> change a
> lot of sweet spots, but once adjusted for those, the improvement in  
> all
> areas is rather extraordinary.  Again, the sweet spots for a lot of  
> things
> seemed to have moved, but given the changes in AGC, that's not really
> surprising.  Using the second receiver in diversity, 4.51 improves  
> signals
> near or at the AGC intercept, allowing a truer separation of those.  
> We
> tested this thoroughly at Field Day, and all who used the K3 with 4.51
> remarked on how good the diversity was.
>
> Weak signal performance with 4.51 was clearly much better by ear if  
> one was
> using an AGC with lower numbers in the settings.
>
> 73, Guy.
>
> On Fri, Jun 29, 2012 at 11:52 AM, W4ATK <[hidden email]> wrote:
>
>> I am saving a copy of 4.48 so I can go back to it.
>>
>> IMHO 4.51 is:
>>       1) More difficult to use
>>       2) Does not perform as well as 4.48
>>       3) Lacks the weak signal performance of 4.48
>>       4) Requires constant adjustment of the RF Gain control
>>
>> This was  the absolute best CW decoder I had ever used. I am still  
>> using
>> 4.51 but have not yet hit the sweet spot with it. Perhaps others  
>> are having
>> better results and will share AGC settings, threshold setting etc.
>>
>> I am can copy call signs, signal reports at elevated speeds, but  
>> alas I am
>> CW challenged and am using CW decode to help me get over the 18 to  
>> 20 wpm
>> block.
>>
>>
>> Jim, W4ATK
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[W5RDW]

>Does anyone have any other pointers? I never have been able to get the
>decoder to do much for me, even during a contest.

Just pay close attention to centering the signal using the CWT decode meter. I usually fine tune a little around the center of the range and find a sweet spot. I have had good results with the last few CW Contests. The last contest I showed my CW challenged brother how well the decoder works. He was truly impressed and is leaning toward a K3 for his next rig.

Lately, I have set my AGC THR at 10-11 as a good point to work for all situations I encounter. Oh, also, my pitch is set either at 520 or 530 Hz usually.

If you can find a station sending code practice on the air, that would be a good place to see what  adjustments are needed for decoding computer generated CW and then you could go from there for not so perfect CW.