[Elecraft] Announcing the Elecraft K3 Transceiver: 10/100W,

[Lyle Johnson]

Hello John!

> An Achilles heel of some SDR's been software longevity and stability, 
> with some folks driven away from otherwise excellent radios due to 
> system crashes and other quirks. Since at least one (think 
> domestic)manufacturer has had significant user issues with their SDR, it 
> would be interesting to hear from someone on the design team who could 
> comment on the K3's software architecture and what longevity/stability 
> testing has been done to date.

Unlike PC based radios, which share the signal processing computer with 
all sorts of applications, the K3 uses embedded processors for all logic 
and processing.  As a result, we have control over what the DSP and 
control processors are doing in the radio.  This in itself promotes 
stability.

Wayne and I are not co-located.  He is in Silicon Valley; I'm north of 
Seattle.  He writes the code for the control processor while I write the 
code for the DSP.  As you can imagine, the fibers have been lit with 
code releases flying back and forth as we've been developing the radio 
over the last three years, and especially during the last two years as 
we've been running prototype hardware and refining the design.

Like most companies, we have developers and we have testing done by 
others who are not writing the code.  We identify a problem -- maybe a 
bug, maybe the next feature on the list that needs to be implemented -- 
go off into our corners and work out a solution, test it among 
ourselves, then expose it to the larger test community.  When we're all 
satisfied, it goes into the release area.

We want the K3 to be dynamic and exciting.  We hope to release not only 
incremental improvements to existing features, but added features from 
time time.  Just like the K3 itself was a surprise to many, if not all, 
of you, some of the new features will also be a (pleasant!) surprise.

73,

Lyle KK7P