On 15 Jun 2007 14:11:02 GMT, Allodoxaphobia
wrote:

On Fri, 15 Jun 2007 05:54:12 -0700, Danny Richardson wrote:
On Fri, 15 Jun 2007 01:18:12 -0700, 4nec2 wrote:


Unfortionally however, it uses the old MiniNec 3.13 engine. Those
interested in antenna modelling should know the drawbacks and pitfalls
as compared to then Nec2 or Nec4 engine used by EZnec and others...

The exception to that is Antenna Model

http://www.antennamodel.com/

"Best viewed at 6400x4800" :-/

Well, I don't know what you are using for a browser, but it works fine
for me using Firefox and my screen resolution is 1280 X 1024

Ian White GM3SEK wrote:

FYI, Roy will be at the Friedrichshafen hamfest in Germany next week, so
he may already be traveling.

Like most travelers, he may have access to e-mail, but is unlikely to
have a newsgroup feed.



Google groups can be used to send/receive newsgroup messages

On 15 jun, 14:54, Danny Richardson wrote:

The exception to that is Antenna Model

http://www.antennamodel.com/

Yes, you are correct. I did not mention 'old mininec 3.13 engine' by
accident. :-)

Arie.

On 15 Jun, 01:18, 4nec2 wrote:
With respect to EZNEC. MANA is much more ambedextrous than
Eznec. It does provide all the old stuff that eznec does
but it also supplies what is new from the last decade.
[...]
Ofcourse even optimisation is not new to antennas
which is why the program is offered for free to
amateurs.

Unfortionally however, it uses the old MiniNec 3.13 engine. Those
interested in antenna modelling should know the drawbacks and pitfalls
as compared to then Nec2 or Nec4 engine used by EZnec and others...

Do you know of a program that for optimisation can work with all
dimensions being variable
AT THE SAME TIME?
At the moment I am confined to using a number of variables in a
antenna array
until the objective is achieved and then by moving these variables to
the bottom of the list
it allows me to work on the next series of variables and so on until
acceptability
is obtained for the whole antenna array. Doing this takes a very long
time when each line
or wire comprises of seven (7) variables even with a high speed
computor.

On 16 jun, 21:45, art wrote:

Do you know of a program that for optimisation can work with all
dimensions being variable AT THE SAME TIME?

Yes, maybe. If I am correct you do have some AO experience. 4nec2 does
contain an optimiser based on the same principles, however it's not
Mininec based as OA was, but Nec2 based (or Nec4 based if you own a
Nec4 license).
You can directly read-in your *.ao files into 4nec2, they will be
automatically converted. Complicated ao files may need some minor
manual changes afterwards.

You can include as much variables as you like in your antenna model
and for each run, you can select the variables to optimise from the
list. When the run has completed and the selected quality of your
model has been optimized as good as possible, you can (partly)
deselect previous variables and select another set of variabels and
continue optimization based on the outcome of previous run.

Furthermore it also contains a GA (genetic algorithms) based optimizer
similar to the one included in the commercial SuperNec package.

When optimizing multiple variables at the same time, this type of
optimizer is prefered. It also avoids the model of being trapped in
local maxima. The drawback of this is the increase of optimization
time needed.

However be aware, it needs some 'learning curve'.
There seem to be two distinct groups of people, those who adore it and
those who think it's much to complicated. It seems not possible to
'sit' anywhere between those both extremes.

4nec2 wrote:
On 16 jun, 21:45, art wrote:


Do you know of a program that for optimisation can work with all
dimensions being variable AT THE SAME TIME?


Yes, maybe. If I am correct you do have some AO experience. 4nec2 does
contain an optimiser based on the same principles, however it's not
Mininec based as OA was, but Nec2 based (or Nec4 based if you own a
Nec4 license).
You can directly read-in your *.ao files into 4nec2, they will be
automatically converted. Complicated ao files may need some minor
manual changes afterwards.

You can include as much variables as you like in your antenna model
and for each run, you can select the variables to optimise from the
list. When the run has completed and the selected quality of your
model has been optimized as good as possible, you can (partly)
deselect previous variables and select another set of variabels and
continue optimization based on the outcome of previous run.

Furthermore it also contains a GA (genetic algorithms) based optimizer
similar to the one included in the commercial SuperNec package.

When optimizing multiple variables at the same time, this type of
optimizer is prefered. It also avoids the model of being trapped in
local maxima. The drawback of this is the increase of optimization
time needed.

However be aware, it needs some 'learning curve'.
There seem to be two distinct groups of people, those who adore it and
those who think it's much to complicated. It seems not possible to
'sit' anywhere between those both extremes.


I think this is true of all optimizers..
The more complex the thing being optimized, the more difficult it is to
have a generic interface that provides a way for the user to tell the
optimizer what the constraints and desires are. In the limit, you wind
up writing a program to implement the constraints, and another program
to implement the evaluation function, and yet another program to take
the optimized variables and turn it into a model which the modeling
engine can run.

On 18 Jun, 12:08, Jim Lux wrote:
4nec2 wrote:
On 16 jun, 21:45, art wrote:

Do you know of a program that for optimisation can work with all
dimensions being variable AT THE SAME TIME?

Yes, maybe. If I am correct you do have some AO experience. 4nec2 does
contain an optimiser based on the same principles, however it's not
Mininec based as OA was, but Nec2 based (or Nec4 based if you own a
Nec4 license).
You can directly read-in your *.ao files into 4nec2, they will be
automatically converted. Complicated ao files may need some minor
manual changes afterwards.

You can include as much variables as you like in your antenna model
and for each run, you can select the variables to optimise from the
list. When the run has completed and the selected quality of your
model has been optimized as good as possible, you can (partly)
deselect previous variables and select another set of variabels and
continue optimization based on the outcome of previous run.

Furthermore it also contains a GA (genetic algorithms) based optimizer
similar to the one included in the commercial SuperNec package.

When optimizing multiple variables at the same time, this type of
optimizer is prefered. It also avoids the model of being trapped in
local maxima. The drawback of this is the increase of optimization
time needed.

However be aware, it needs some 'learning curve'.
There seem to be two distinct groups of people, those who adore it and
those who think it's much to complicated. It seems not possible to
'sit' anywhere between those both extremes.

I think this is true of all optimizers..
The more complex the thing being optimized, the more difficult it is to
have a generic interface that provides a way for the user to tell the
optimizer what the constraints and desires are. In the limit, you wind
up writing a program to implement the constraints, and another program
to implement the evaluation function, and yet another program to take
the optimized variables and turn it into a model which the modeling
engine can run.- Hide quoted text -

- Show quoted text -

Yes I would agree thus the use of MINNINEC that is used within
the confines of its known limitations is just as good as any program
that use know applicable laws. If a program what you wan't to use
has a optimiser then the results should be similar tho one may have
an accurracy that is outside the field of amateur radio but not
so correct as to prove the MINNINEC inaccurate.
For myself I would like to model a dish antenna with multi elements
in Gaussian form but the movement of variables would be horrible
in terms of time consumed assuming some 60 odd elements each
consisting
of some 10 lines with all units variable.
Regards
Art

On Mon, 18 Jun 2007 12:08:09 -0700, Jim Lux
wrote:

I think this is true of all optimizers..
The more complex the thing being optimized, the more difficult it is to
have a generic interface that provides a way for the user to tell the
optimizer what the constraints and desires are. In the limit, you wind
up writing a program to implement the constraints, and another program
to implement the evaluation function, and yet another program to take
the optimized variables and turn it into a model which the modeling
engine can run.

Hi Jim, Arie,

It goes WAY beyond that (as if there weren't enough problems). In the
GA community, the introduction of unlimited variables leads to (very
quickly with even a few of them) what is called "combinatorial
explosion."

A second issue (and it seems that Arie probably has at least one
algorithm to tackle this) is with an engine becoming stuck at a local
minima or maxima. This false solution ignores a better one nearby (or
further down the road) simply because it satisfied the criteria within
a restricted region of a curve.

Arthur's models quite obviously exhibit this last problem when his
designs can be bested with mediocre examples that have been drawn from
the dusty bookshelves. The science of GA has moved well beyond the
Paleolithic era of AO.

73's
Richard Clark, KB7QHC

Richard Clark wrote:
On Mon, 18 Jun 2007 12:08:09 -0700, Jim Lux
wrote:


I think this is true of all optimizers..
The more complex the thing being optimized, the more difficult it is to
have a generic interface that provides a way for the user to tell the
optimizer what the constraints and desires are. In the limit, you wind
up writing a program to implement the constraints, and another program
to implement the evaluation function, and yet another program to take
the optimized variables and turn it into a model which the modeling
engine can run.


Hi Jim, Arie,

It goes WAY beyond that (as if there weren't enough problems). In the
GA community, the introduction of unlimited variables leads to (very
quickly with even a few of them) what is called "combinatorial
explosion."

A second issue (and it seems that Arie probably has at least one
algorithm to tackle this) is with an engine becoming stuck at a local
minima or maxima. This false solution ignores a better one nearby (or
further down the road) simply because it satisfied the criteria within
a restricted region of a curve.

Arthur's models quite obviously exhibit this last problem when his
designs can be bested with mediocre examples that have been drawn from
the dusty bookshelves. The science of GA has moved well beyond the
Paleolithic era of AO.

73's
Richard Clark, KB7QHC
My comment was not so much about the details of the optimizers (for
which you've given some typical problems) but with the difficulty of
coming up with a generic optimizer that can manage generalized models,
generalized constraints, etc. Since the underlying optimizer algorithms
are available as canned packages/library routines in most cases, the
hard work is in formulating the stuff that goes to the optimizer.

On 18 Jun, 14:22, Jim Lux wrote:
Richard Clark wrote:
On Mon, 18 Jun 2007 12:08:09 -0700, Jim Lux
wrote:

I think this is true of all optimizers..
The more complex the thing being optimized, the more difficult it is to
have a generic interface that provides a way for the user to tell the
optimizer what the constraints and desires are. In the limit, you wind
up writing a program to implement the constraints, and another program
to implement the evaluation function, and yet another program to take
the optimized variables and turn it into a model which the modeling
engine can run.

Hi Jim, Arie,

It goes WAY beyond that (as if there weren't enough problems). In the
GA community, the introduction of unlimited variables leads to (very
quickly with even a few of them) what is called "combinatorial
explosion."

A second issue (and it seems that Arie probably has at least one
algorithm to tackle this) is with an engine becoming stuck at a local
minima or maxima. This false solution ignores a better one nearby (or
further down the road) simply because it satisfied the criteria within
a restricted region of a curve.

Arthur's models quite obviously exhibit this last problem when his
designs can be bested with mediocre examples that have been drawn from
the dusty bookshelves. The science of GA has moved well beyond the
Paleolithic era of AO.

73's
Richard Clark, KB7QHC

My comment was not so much about the details of the optimizers (for
which you've given some typical problems) but with the difficulty of
coming up with a generic optimizer that can manage generalized models,
generalized constraints, etc. Since the underlying optimizer algorithms
are available as canned packages/library routines in most cases, the
hard work is in formulating the stuff that goes to the optimizer.- Hide quoted text -

- Show quoted text -

The bottom line is that if they are both based around the same
theme then they should be similar. Buying a new shiny radio
does not change the news supplied by the old radio.
Ofcourse there are different types of optimisers, for instance
the latest antenna program using NEC has an optimiser that can only
be
used for Yagi's where other optimisers are not so limited.
Yes it may have the latest bells and whistles but if it is
designed only for yagis then it wouldn't meet my needs.
Regards
Art