Glitch - Kenwood or Bill?

[John Ronan]

On 21 Apr 2009, at 01:58, Bill Vodall WA7NWP wrote:

> Am I glitching or does the D710 really need to be fixed:
>
I didn't think it had any difficulties. The one I tested (with about  
1.6MB of traffic in kiss mode) is now (fairly permanently in my car).

I've ordered another one for the shack, and I'll see if I can  
organise something between the two of them when I receive the radio.

John

> -----
>
>
> On Mon, Apr 20, 2009 at 5:26 PM, Rick Muething <rmuething at c> wrote:
>> It doesn't take a Kenwood fix..all it takes is using an escape  
>> sequence in
>> the driver to avoid the "TC 0<Cr>" sequence.  It turns out that  
>> sequence can
>> be avoided by a simple replacement of any 'C' with a two-byte  
>> sequence
>> [0xDB, 'C' ] during HDLC byte stuffing which is then automatically  
>> removed
>> by the standard KISS protocol once it gets into the TNC firmware.  
>> I am
>> afraid changing it in the firmware in the 710 serial processor could
>> negatively impact other programs that want to exit the KISS mode or
>> temporarily go to what I call "front panel" mode on the TNC's  
>> serial port.
>
> They will already be negatively impacted if they ever expect KISS to
> be a pure binary data string.
>
>> If one really wanted to do it correctly the ONLY way to exit the  
>> KISS mode
>> should be the standard KISS exit sequence. This would require all  
>> software
>> to use that sequence before sending a non KISS command like "TC  
>> 0<Cr>" to
>> the D710.
>
> That's is the way it needs to be.
>
> If Kenwood does exit from KISS mode when it sees a "TC 0" in the
> binary data stream, then it violates the KISS protocol, is unreliable
> and needs to be fixed.    At the very least they could do what some
> folks are doing with 3 control-C's and listening for some seconds of
> free time before and after the control codes.   That's a work around
> but it works.     Having to change 20 years of existing software
> drivers because one new radio has a glitch is not the right way to fix
> it.
>
> Bill
> _______________________________________________
> Seatcp mailing list
> Seatcp at wetnet.net
> http://wetnet.net/mailman/listinfo/seatcp
>

--
John Ronan <jronan at tssg.org>, +353-51-302938
Telecommunications Software &  Systems Group,  http://www.tssg.org