--- "George, W5YR" <w5yr@att.net> wrote:
> Well, let's look at the other side of the story:
>
> If a system can be updated or re-defined by a single
> ROM flashing, then
> this suggests strongly that the system is fairly
> unsophisticated in that it
> can function from the contents of a single ROM.
Does it necessarily represent "less" sophistication?
On the contrary, I think it could be argued it
actually represents more, at a minimum a higher degree
of cleverness. As an example the unsophisticated
single conversion K2 can stand proud amongst the top
dollar triple and quadruple conversions receivers out
there. Sophistication alone does not assure good
performance, in many case it may actually hurt. One of
my global criticism of the imported radio makers is
their propensity to keep adding parts until the box
does what they think it should rather than standing
back a few steps to rethink about it for a fresh
restart.
<snip>
> Why isn't the PRO updatable the same way? Mainly
> because it is a much more
> complex system containing not only a dedicated DSP
> processor chipset but
> also some 5 or 6 other uprocessors in various parts
> of the radio, and each
> of them has burned-in ROM instructions. If all of
> these processors were
> operated in a distributed network, not only would
> the system complexity
> greatly increase - along with the cost - but
> operating speed could become
> compromised unless some pretty fast busses and
> memory devices were used.
Why would you even want to try and network an array a
disparate microprocessors in a radio? I'm not sure if
I can fully follow the logic of using 5 or 6
processors in a ham rig anyhow? What could they all
possibly be doing that can't be handled by other
simpler means? One thing to date that the import radio
makers haven't done is expand their concept of what a
radio really is, they are still just incrementally
expanding on the rather narrowly defined classic
transceiver concept that goes back more than 40 years.
My fear based on what I've seen so far is that the
Orion is trying to hard to do the same rather than
expanding the "outside of the box" thinking
(literally) - concepts pioneered in the Kachina 505
and later the Pegasus.
<snip>
> I think that this whole idea of "software-defined"
> radios has gotten way
> out of hand and divorced from reality. The fact that
> a radio may be largely
> *controlled* by software does not equate to its
> being a *software-defined*
> radio.
I think arguably radios like the Pegasus/Jupiter and
the upcoming Orion are indeed software defined in that
they are not particularly limited to what they can do
inside of small ~5KHz slices of RF spectrum. They are
merely bounded by their roofing filter bandwidths, as
such they will always be narrow bandwidth mode
devices. I'm also convinced that the Pegasus/Jupiter
could be significantly "redefined" from the way they
behave now by firmware alone. I'm thinking in terms of
something other than a just CW/SSB box. Granted they
will always be constrained by their synthesizer
traits, ADC choice, and DSP horsepower.
<snip>
>In order to remain current and to add many
> types of new features and
> capabilities, at least some of the existing hardware
> must be replaced,
> either by modification or substitution. This calls
> for a modular
> construction which can be much more expensive than
> more conventional
> layouts. For an example, compare the extensively
> shielded modules used with
> the complex backplane structure of the Kachina 505
> DSP with the
> implementation of the Pegasus. Yet, when a fairly
> minor change was made in
> the Kachina to upgrade the Speech Monitor among
> other things, one was
> required to purchase a new $400 module and firmware
> and a new software
> control program to enjoy those advantages.
I suspect part of the costliness of upgrading the 505
was due in part to being the first into the wilderness
of DSP ham gear, and the economical, ideal parts
didn't exist yet. It also may represent some poor
design choices of the overall architecture that didn't
allow for more future enhancements by firmware alone.
The 505's chassis design has always struck me as being
overly complex. I would also agree though that the
Pegasus/Jupiter could have used just a bit more
modularity, but not much. And yes they could stand
better shielding, especially from outside sources of
RF.
Duane
N9DG
__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/
|