Experimental improvements to K3 CW text decode

[wayne burdick]

Hi all,

We have a new field-test K3 firmware release (MCU 4.58/DSP 2.80) that  
offers improved CW text decode. The improvement is due largely to the  
addition of an independent CW decode AGC system. The regular AGC  
threshold, decay, slope, and other settings now have virtually no  
effect on CW decode accuracy. Decode timing has also been changed  
slightly to improve copy of code sent with shortened letter spaces--
typical of operators who "rush" their code, running letters closer  
together.

(Of course there isn't much we can do about extreme QSB, though we did  
briefly consider adding natural language parsing and a predictive  
lexicographical and semantic engine.)

If you'd like to try this experimental release, please email me  
directly. You'll need to load the code using the instructions for beta  
code releases on our web site.

73,
Wayne
N6KR

______________________________________________________________
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]

Thanks for all the requests for the new field-test firmware. I have  
plenty of testers now :)  After testing is complete we'll get the  
firmware up as beta on the web site.

73,
Wayne
N6KR

______________________________________________________________
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

[Barry K3NDM]

Will this be the same algorithm that is used in the KX3? The one in the
KX3 seems to work pretty well.

73,
Barry
K3NDM

On 11/8/2012 12:00 AM, Wayne Burdick wrote:

> Hi all,
>
> We have a new field-test K3 firmware release (MCU 4.58/DSP 2.80) that
> offers improved CW text decode. The improvement is due largely to the
> addition of an independent CW decode AGC system. The regular AGC
> threshold, decay, slope, and other settings now have virtually no
> effect on CW decode accuracy. Decode timing has also been changed
> slightly to improve copy of code sent with shortened letter spaces--
> typical of operators who "rush" their code, running letters closer
> together.
>
> (Of course there isn't much we can do about extreme QSB, though we did
> briefly consider adding natural language parsing and a predictive
> lexicographical and semantic engine.)
>
> If you'd like to try this experimental release, please email me
> directly. You'll need to load the code using the instructions for beta
> code releases on our web site.
>
> 73,
> Wayne
> N6KR
>
> ______________________________________________________________
> 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