[Elecraft] Software Development Goals

[Tom W8JI]

>     I am consummately pleased with my K3.  I am astounded 
> with the raw
> performance and dazzled with the features.


So am I.

The K3 is the first radio I have ever bought without some 
silly engineering shortfall. The only things I see people 
complaining about are specialized personal use or 
application issues. It's impossible without years of field 
refinement to get on the very top of the curve, but they are 
already close and respond fast.

> surpassing expectations masterfully.  The fact they have 
> incorporated in
> their development strategy a transparent and responsive 
> forum easily
> accessible to the unqualified and very well qualified is 
> revolutionary

I've been involved with other groups and it is often to get 
them to even consider major issues. It's so frustrating I 
vowed to never again work with a radio manufacturer. For 
example one manufacturer had an amplifier keying issue. The 
radio spit out RF before it told the amplifier to turn on. I 
knew it was happening because my homebrew amp has a 
"hot-switch" sensor that prevents keying (relay transfer) 
while RF is present, and I looked at it on a scope and could 
see the issue. It took almost a week of actual work to get 
them to look at the problem. The engineer responsible kept 
saying he checked it and it was fine, but it turned out he 
never did check. When they finally checked they fixed it, 
but historically everything they did worked that way. It 
worked that way with ALC issues that caused keyclicks, and 
it worked that way with a dozen other bugs some of which 
never were resolved.

All three major Japanese manufacturers are out of touch the 
same way.  It actually takes decades to address some very 
simple problems.

Thankfully Elecraft has chosen a different approach.

> and as yet unimplemented features.  Frequently, 
> operational flexibility or
> designer prerogative is misconstrued as a failure or a bug 
> and unimplemented
> features thus far have a history of being implemented.

That's the problem I see. While there are shortfalls in 
specialized areas, it is always an application specific and 
design critical area like a noise blanker or IF port use. 
What works well in one case might hurt other uses so they 
have to find a compromise, and that takes time. They have to 
learn all the different field applications, it can't really 
be planned because planning would take so much time the 
first radio would never leave the assembly line.

73 Tom



.