Which OPTICs design program is best ?

[Joseph]

Hi:

I've been using ZMAX for 10 years of more and I need to upgrade my program.
Is there a better, and less-expensive program available now ?

If so which one and why ?

Thanks,

Joseph Labrum
Labrum Optical Engineering

[Dave Schaack]

Post by Joseph
I've been using ZMAX for 10 years of more and I need to upgrade my
program. Is there a better, and less-expensive program available now ?
If so which one and why ?

Joseph, do you just want to attract quick, snide comments about ZEMAX, or do
you really want to discuss the merits of different programs? If the latter,
you need to detail what you find lacking in ZEMAX.

SYNOPSYS is cheaper than ZEMAX, and can do things ZEMAX cannot. However,
ZEMAX can do things that SYNOPSYS cannot. I use both.

[Helpful person]

Post by Dave Schaack
SYNOPSYS is cheaper than ZEMAX, and can do things ZEMAX cannot. However,
ZEMAX can do things that SYNOPSYS cannot. I use both.

I used to use Synopsys way back when it was a DOS program. Then it was
so versatile that you could do almost anything with it. (It would not
cook breakfast.) It was not a program for amateurs but for real
designers it was wonderful.

I have not used the latest windows version. Has it lost some of its
versatility? Is it still a one man show (Don Dilworth)?

[Dave Schaack]

Post by Helpful person
I used to use Synopsys way back when it was a DOS program. Then it was
so versatile that you could do almost anything with it. (It would not
cook breakfast.) It was not a program for amateurs but for real
designers it was wonderful.
I have not used the latest windows version. Has it lost some of its
versatility? Is it still a one man show (Don Dilworth)?

It is still a one man show, but Don is very responsive to his users. I
don't think it has lost much, if anything, but I never used the DOS version,
so can't say for sure. I did use the time-shared version over a 300 baud
modem back in the early 80s. That was my introduction to commercial ray
tracing programs.

[Helpful person]

What can Zemax do that Synopsys cannot?

Please visit my web site at www.richardfisher.com

[Dave Schaack]

Post by Helpful person
What can Zemax do that Synopsys cannot?

ZEMAX has a _lot_ of features and more are constantly being added. I can't
give a comprehensive answer. But one big thing is that SYNOPSYS can
optimize only over 6 configurations, which is plenty for most work. I think
the number of available configurations in ZEMAX may be arbitrarily high; I
have used up to 16 configurations for special purposes.

[Helpful person]

Post by Dave Schaack

Post by Helpful person
What can Zemax do that Synopsys cannot?

ZEMAX has a _lot_ of features and more are constantly being added. I can't
give a comprehensive answer. But one big thing is that SYNOPSYS can
optimize only over 6 configurations, which is plenty for most work. I think
the number of available configurations in ZEMAX may be arbitrarily high; I
have used up to 16 configurations for special purposes.

That's interestng. In the old DOS days, if I remember correctly,
Synopsys did not have a specific limit. With the limited DOS resources
one could trade off number of surfaces versus number of colors etc.

Please visit my web site at www.richardfisher.com

[Dave Schaack]

Post by Helpful person
That's interestng. In the old DOS days, if I remember correctly,
Synopsys did not have a specific limit.

Hmmm. The limit was 6 configurations back in the mainframe/time sharing
days. The DOS version was, in part, supported and modified by Breault
Research, so that may be the source of the difference.

[Helpful person]

Post by Dave Schaack

Post by Helpful person
That's interestng. In the old DOS days, if I remember correctly,
Synopsys did not have a specific limit.

Hmmm. The limit was 6 configurations back in the mainframe/time sharing
days. The DOS version was, in part, supported and modified by Breault
Research, so that may be the source of the difference.

I was using Synopsys before Breault. Maybe I got it wrong.

[Jim Klein]

Post by Dave Schaack

Post by Helpful person
That's interestng. In the old DOS days, if I remember correctly,
Synopsys did not have a specific limit.

Hmmm. The limit was 6 configurations back in the mainframe/time sharing
days. The DOS version was, in part, supported and modified by Breault
Research, so that may be the source of the difference.

KDP-2 has 75 configurations but can be re-compiled for any number.


James E. Klein
***@earthlink.net

Engineering Calculations
http://www.ecalculations.com
***@ecalculations.com
Engineering Calculations is the home of
the KDP-2 Optical Design Program
for Windows and (soon) MAC OSX
Free KDP-2 (DEMO) downloadable!
1-818-507-5706 (Voice and Fax)

[Helpful person]

Post by Jim Klein
KDP-2 has 75 configurations but can be re-compiled for any number.
James E. Klein

Didn't you write the program with dynamic arrays so that recompilation
would not be necessary?

Please visit my web site at www.richardfisher.com

[Jim Klein]

Post by Helpful person

Post by Jim Klein
KDP-2 has 75 configurations but can be re-compiled for any number.
James E. Klein

Didn't you write the program with dynamic arrays so that recompilation
would not be necessary?
Please visit my web site at www.richardfisher.com

Ever since dynamic arrays became available in Fortran 90, I have used
them. I simply have not had the time to rewrite that part.

When the lens database routines are updated, then the arrays will be
dynamic.

If someone needs more configs, I can recompile and expand the capacity
but so far 75 configs has been enough for my users.




James E. Klein
***@earthlink.net

Engineering Calculations
http://www.ecalculations.com
***@ecalculations.com
Engineering Calculations is the home of
the KDP-2 Optical Design Program
for Windows and (soon) MAC OSX
Free KDP-2 (DEMO) downloadable!
1-818-507-5706 (Voice and Fax)

[Helpful person]

Post by Jim Klein
Ever since dynamic arrays became available in Fortran 90, I have used
them. I simply have not had the time to rewrite that part.
When the lens database routines are updated, then the arrays will be
dynamic.
If someone needs more configs, I can recompile and expand the capacity
but so far 75 configs has been enough for my users.
James E. Klein

I didn't mean to be critical. I know how difficult it is to write and
support a large program.

Please visit my web site at www.richardfisher.com

[w***@gmail.com]

Joseph,
Let me add to your question after having browsed your website-
How do you model an acousto-optic deflector in an optical engineering
program (other than just as a grating)?
Wolfgang

Post by Joseph
I've been using ZMAX for 10 years of more and I need to upgrade my program.
Is there a better, and less-expensive program available now ?
If so which one and why ?
Thanks,
Joseph Labrum
Labrum Optical Engineering

[Dave Schaack]

Post by w***@gmail.com
Let me add to your question after having browsed your website-
How do you model an acousto-optic deflector in an optical engineering
program (other than just as a grating)?

Wolfgang, what is it that you need to model? If the acousto-optic deflector
is not messing up the beam, then the irradiance of the beam is not going to
change significantly over the distance it propagates through the deflector.
Thus a grating model should suffice for most purposes, it seems to me.

[w***@gmail.com]

Hi Dave,
Yes, that is a good first approximation- just put a grating between two
pieces of glass. One could model the deflector more exactly by
diffracting off of a volume refractive index grating.

A yet more complete model would frequency shift and deflect a laser
beam off of a travelling index modulation. If you want to model a
heterodyning interferometer, this would be appropriate. If you wanted
to use an optical design program to do this, you would probably have to
create multiple interferometer traces with incremental phase shifts on
the deflected beam.

I'm happy with a working index modulation scheme. Should some
'physical optics propagation' capable programs be able to handle this?
Wolfgang

Post by Dave Schaack

Post by w***@gmail.com
Let me add to your question after having browsed your website-
How do you model an acousto-optic deflector in an optical engineering
program (other than just as a grating)?

Wolfgang, what is it that you need to model? If the acousto-optic deflector
is not messing up the beam, then the irradiance of the beam is not going to
change significantly over the distance it propagates through the deflector.
Thus a grating model should suffice for most purposes, it seems to me.

[Dave Schaack]

Post by w***@gmail.com
A yet more complete model would frequency shift and deflect a laser
beam off of a travelling index modulation. If you want to model a
heterodyning interferometer, this would be appropriate. If you wanted
to use an optical design program to do this, you would probably have to
create multiple interferometer traces with incremental phase shifts on
the deflected beam.
I'm happy with a working index modulation scheme. Should some
'physical optics propagation' capable programs be able to handle this?
Wolfgang

Wolfgang, I'm still unclear about what answers you want to get from the
model. You already know about the frequency shift, so what does the index
modulation model give you in addition? POP type codes, of course, can get
into this level of detail. You have to ask yourself whether you need a
time-dependent model or whether some sort of time average will suffice.
There are probably codes out there that can provide whatever you need, but
they may not be standard optical design codes.

[w***@gmail.com]

Hi Dave-
I think that you're asking the right question. For my immediate needs,
a grating will do just fine. If I do more exact modelling, it's just
for the sake of curiosity, not for a specific answer.

I should add that I'm trying to deflect at multiple angles
simultaneously (by mixing multiple (RF) drive frequencies), so my
deflector would have to be modeled by a grating with multiple periods.
I'm a bit new to optical design programs, but I'm assuming that this is
possible to do.
Wolfgang

Post by Dave Schaack

Post by w***@gmail.com
A yet more complete model would frequency shift and deflect a laser
beam off of a travelling index modulation. If you want to model a
heterodyning interferometer, this would be appropriate. If you wanted
to use an optical design program to do this, you would probably have to
create multiple interferometer traces with incremental phase shifts on
the deflected beam.
I'm happy with a working index modulation scheme. Should some
'physical optics propagation' capable programs be able to handle this?
Wolfgang

Wolfgang, I'm still unclear about what answers you want to get from the
model. You already know about the frequency shift, so what does the index
modulation model give you in addition? POP type codes, of course, can get
into this level of detail. You have to ask yourself whether you need a
time-dependent model or whether some sort of time average will suffice.
There are probably codes out there that can provide whatever you need, but
they may not be standard optical design codes.

[Dave Schaack]

Post by w***@gmail.com
I should add that I'm trying to deflect at multiple angles
simultaneously (by mixing multiple (RF) drive frequencies), so my
deflector would have to be modeled by a grating with multiple periods.
I'm a bit new to optical design programs, but I'm assuming that this is
possible to do.

Wolfgang, that's interesting. I'm not immediately sure that the optical
design programs would give the right answer in this case. (You would simply
stack gratings.) I think that in reality you will get not only the directly
diffracted beams, but also multiply diffracted beams. Usually, in the
optical design programs, you tell it which diffracted order you want.
Thinking out loud, it probably will work, but you'd have to use multiple
configurations to track all of the beams you may be interested in.

[w***@gmail.com]

Dave,
[in retrospect, I should have started another discussion string on this
subject, but let's finish this here]

Yes, stacking the gratings will give you complications (at least in
Optica, the program that I use). Separating them along the beam
propagation direction gives you combined diffractions, as if you're
diffracting off of the sum and difference frequencies of the different
modulations of your diffraction grating. These occur in acousto-optic
deflectors as well, but they are much less pronounced.
One way around this is to model each diffraction with a different
light source, but this gets computationally expensive, and sometimes
leads to complications later in the trace.
I don't have a POP package right now, but I should soon. The index
grating is high on my immediate list of things to try!
Wolfgang

Post by Dave Schaack

Post by w***@gmail.com
I should add that I'm trying to deflect at multiple angles
simultaneously (by mixing multiple (RF) drive frequencies), so my
deflector would have to be modeled by a grating with multiple periods.
I'm a bit new to optical design programs, but I'm assuming that this is
possible to do.

Wolfgang, that's interesting. I'm not immediately sure that the optical
design programs would give the right answer in this case. (You would simply
stack gratings.) I think that in reality you will get not only the directly
diffracted beams, but also multiply diffracted beams. Usually, in the
optical design programs, you tell it which diffracted order you want.
Thinking out loud, it probably will work, but you'd have to use multiple
configurations to track all of the beams you may be interested in.

[Dave Schaack]

Post by w***@gmail.com
I don't have a POP package right now, but I should soon. The index
grating is high on my immediate list of things to try!

Which package are you going to get? Why that one?

[w***@gmail.com]

Dave,
Basically, I've been working on ultrafast technology for a few years,
and I'm looking to work on some projects that would benefit greatly
from software design. I'm OK with steep learning curves, and would
like flexibility.

I'm young in the field, so I need to get an 'industry standard'
package. This basically means that I need to get a package that has
been well marketed. Most folks on this list understandably scoff at
people who buy software for this reason, but I'm trying to make
decisions to launch a career. You can probably guess the direction I'm
headed by these statements. Go to the job search engine 'indeed.com'
and rank optics software according to job solicitations. (Code-v is
out for me because of the cost).

Any suggestions would be helpful. I'm glad to have returned to the
original discussion topic!
Wolfgang

Post by Dave Schaack

Post by w***@gmail.com
I don't have a POP package right now, but I should soon. The index
grating is high on my immediate list of things to try!

Which package are you going to get? Why that one?

[Dave Schaack]

<***@gmail.com> wrote in message news:<***@h48g2000cwc.googlegroups.com>...


I'm young in the field, so I need to get an 'industry standard'
package. This basically means that I need to get a package that has
been well marketed. Most folks on this list understandably scoff at
people who buy software for this reason, but I'm trying to make
decisions to launch a career. You can probably guess the direction I'm
headed by these statements. Go to the job search engine 'indeed.com'
and rank optics software according to job solicitations. (Code-v is
out for me because of the cost).

I can completely understand this motivation. However, I couldn't figure out
how to do that search in the 1 minute I was willing to spend. See below.


Any suggestions would be helpful. I'm glad to have returned to the
original discussion topic!

Wolfgang, I get the impression from the combination of these comments that
you are talking about ZEMAX. I hope that is not the case, because I believe
you will be sorely disappointed in its physical optics propagation
capability.

[Jamie]

Generally, one can separate the temporal (and more exact physical
optics) model from the design of systems using AOM. Remember that the
optical system is, for all intents and purposes, static to the signal
injected into the sonic field. Further, the work of the optical
designer, ultimately, is the design of real optics that give a linear
transfer function by balancing all (or as many as possible) of the
higher order geometrical terms (aberrations).

Typically I will model the AOM as an immersed grating but I will "zoom"
the grating frequency over its corresponding RF bandwidth using
multiple configurations. One may even weight these configs with the
AOM's RF spectral efficiency as measured or specified by the
design/vendor and even include spurious orders. In this model, all of
the spatial frequencies that support diffraction are included in the
model and their resulting (diffracted) optical response in optical
plane wave spectrums are accounted as well.

The optical system is then optimized for the application based on these
interactions. This allows the optical design to accommodate all signals
that may be used in the device based on its linear system response.
Arbitrary signals cases can be assessed on the basis of superposition.

If you are concerned about non-linear effects such as intermodulation
products, non-linear interaction (ie approaching 100% diffraction
efficiency) or non-linear propagation in the sonic field, then these
types of analyses are best left to Matlab or Mathcad.

James Carter
http://www.jacarter3.com
http://www.opticalconsulting.com

Post by Dave Schaack

Post by w***@gmail.com
I should add that I'm trying to deflect at multiple angles
simultaneously (by mixing multiple (RF) drive frequencies), so my
deflector would have to be modeled by a grating with multiple periods.
I'm a bit new to optical design programs, but I'm assuming that this is
possible to do.

Wolfgang, that's interesting. I'm not immediately sure that the optical
design programs would give the right answer in this case. (You would simply
stack gratings.) I think that in reality you will get not only the directly
diffracted beams, but also multiply diffracted beams. Usually, in the
optical design programs, you tell it which diffracted order you want.
Thinking out loud, it probably will work, but you'd have to use multiple
configurations to track all of the beams you may be interested in.

[w***@gmail.com]

Hi James,
Thanks for the detailed response!
Generally, I use a similar technique (with different grating periods)
to model an acousto-optic pulse shaper. The benefit of using a
ray-tracing package here is in modeling the temporal dispersion of an
ultrashort pulse (distance traveled along the propagation path vs.
wavelength).
Wolfgang

Post by Jamie
Generally, one can separate the temporal (and more exact physical
optics) model from the design of systems using AOM. Remember that the
optical system is, for all intents and purposes, static to the signal
injected into the sonic field. Further, the work of the optical
designer, ultimately, is the design of real optics that give a linear
transfer function by balancing all (or as many as possible) of the
higher order geometrical terms (aberrations).
Typically I will model the AOM as an immersed grating but I will "zoom"
the grating frequency over its corresponding RF bandwidth using
multiple configurations. One may even weight these configs with the
AOM's RF spectral efficiency as measured or specified by the
design/vendor and even include spurious orders. In this model, all of
the spatial frequencies that support diffraction are included in the
model and their resulting (diffracted) optical response in optical
plane wave spectrums are accounted as well.
The optical system is then optimized for the application based on these
interactions. This allows the optical design to accommodate all signals
that may be used in the device based on its linear system response.
Arbitrary signals cases can be assessed on the basis of superposition.
If you are concerned about non-linear effects such as intermodulation
products, non-linear interaction (ie approaching 100% diffraction
efficiency) or non-linear propagation in the sonic field, then these
types of analyses are best left to Matlab or Mathcad.
James Carter
http://www.jacarter3.com
http://www.opticalconsulting.com

Post by Dave Schaack

Post by w***@gmail.com
I should add that I'm trying to deflect at multiple angles
simultaneously (by mixing multiple (RF) drive frequencies), so my
deflector would have to be modeled by a grating with multiple periods.
I'm a bit new to optical design programs, but I'm assuming that this is
possible to do.

Wolfgang, that's interesting. I'm not immediately sure that the optical
design programs would give the right answer in this case. (You would simply
stack gratings.) I think that in reality you will get not only the directly
diffracted beams, but also multiply diffracted beams. Usually, in the
optical design programs, you tell it which diffracted order you want.
Thinking out loud, it probably will work, but you'd have to use multiple
configurations to track all of the beams you may be interested in.

[Jim Klein]

Post by Joseph
I've been using ZMAX for 10 years of more and I need to upgrade my program.
Is there a better, and less-expensive program available now ?
If so which one and why ?
Thanks,
Joseph Labrum
Labrum Optical Engineering

User support and price are conciderations.

CODE-V is the best but expensive.

KDP is my choice. It is low price, it comes with great user support
and it also comes with source code incase I (the autor) step in front
of a bus.


James E. Klein
***@earthlink.net

Engineering Calculations
http://www.ecalculations.com
***@ecalculations.com
Engineering Calculations is the home of
the KDP-2 Optical Design Program
for Windows and (soon) MAC OSX
Free KDP-2 (DEMO) downloadable!
1-818-507-5706 (Voice and Fax)

[w***@gmail.com]

Jim,
Have you ever considered open-sourcing KDP?
Wolfgang

Post by Jim Klein

Post by Joseph
I've been using ZMAX for 10 years of more and I need to upgrade my program.
Is there a better, and less-expensive program available now ?
If so which one and why ?
Thanks,
Joseph Labrum
Labrum Optical Engineering

User support and price are conciderations.
CODE-V is the best but expensive.
KDP is my choice. It is low price, it comes with great user support
and it also comes with source code incase I (the autor) step in front
of a bus.
James E. Klein
Engineering Calculations
http://www.ecalculations.com
Engineering Calculations is the home of
the KDP-2 Optical Design Program
for Windows and (soon) MAC OSX
Free KDP-2 (DEMO) downloadable!
1-818-507-5706 (Voice and Fax)

[Jim Klein]

Post by w***@gmail.com
Jim,
Have you ever considered open-sourcing KDP?
Wolfgang

Post by Jim Klein

Post by Joseph
I've been using ZMAX for 10 years of more and I need to upgrade my program.
Is there a better, and less-expensive program available now ?
If so which one and why ?
Thanks,
Joseph Labrum
Labrum Optical Engineering

User support and price are conciderations.
CODE-V is the best but expensive.
KDP is my choice. It is low price, it comes with great user support
and it also comes with source code incase I (the autor) step in front
of a bus.
James E. Klein
Engineering Calculations
http://www.ecalculations.com
Engineering Calculations is the home of
the KDP-2 Optical Design Program
for Windows and (soon) MAC OSX
Free KDP-2 (DEMO) downloadable!
1-818-507-5706 (Voice and Fax)

Yes but it never got any interest in it.

Besides, if someone buys KDP-2 for the less than $600.00 price, they
get the sounce code and since I have neither the inclination or the
means to stop them, they could distribute the code to the whole world.

Anyone who wants KDP-2 for the upgrade price of $129.00. just needs to
tell me they have an earlier version and since I'm easy, I'd believe
them. :-)

I do KDP-2 as an exercise in intellectual curiosity. I have a day job
that pays the bills.

Also, KDP-2 is free to schools and available for $50.00 to students if
their instructors don't already have it.

Giving an optical design program away for free to teachers is not
easy.




James E. Klein
***@earthlink.net

Engineering Calculations
http://www.ecalculations.com
***@ecalculations.com
Engineering Calculations is the home of
the KDP-2 Optical Design Program
for Windows and (soon) MAC OSX
Free KDP-2 (DEMO) downloadable!
1-818-507-5706 (Voice and Fax)

[Etienne Labaume]

Post by Jim Klein

Post by w***@gmail.com
Jim,
Have you ever considered open-sourcing KDP?

Yes but it never got any interest in it.

[...]

Post by Jim Klein
Also, KDP-2 is free to schools and available for $50.00 to students if
their instructors don't already have it.
Giving an optical design program away for free to teachers is not
easy.

Opensourcing a software is not only giving it away for free, it's
also getting users testing more, having other developpers correcting
bugs for you, porting the software to others systems, adding features,
and companies like OSDN mirroring the build and source code on their
website, saving your bandwith. And it doesn't keep you from selling
services about this software if you still want to earn money with it.

[Etienne Labaume]

Post by Joseph
I've been using ZMAX for 10 years of more and I need to upgrade my program.
Is there a better, and less-expensive program available now ?
If so which one and why ?

I'm new to optical design, but isn't this rule true:
"The best tool is the one you know the best"

How much time will you spend learning how to apply to
a new software all the tips you know about Zemax after
10 years of use ?

[Jamie]

My brother gives me a hard time when he says that "barbeque" is process
not a sauce. Well, it's been my experience that optical design is a
process, not a software program. If you know the process, then you use
the software to accomplish that process.

Having used Genii PC, OSLO, SynOpSys, KDP, ACCOS V and Code V, I have
found that all of the tips I know are implemented fairly well on all of
these with some doing better than others in certain areas. But the
process of finding a good solution regarding performance, complexity
and manufacturing yield is the same for all.

My question to you is: how many new tips will you learn if you leave
the familiar and venture out of your comfort zone? My guess is that
such a move will stimulate your creativity and re-organize your
thinking in new ways. This may make the transition much more profitable
and enjoyable than you think.

Just my thoughts regarding optical design...

James A Carter III
http://www.jacarter3.com
http://www.opticalconsulting.com

Post by Etienne Labaume

Post by Joseph
I've been using ZMAX for 10 years of more and I need to upgrade my program.
Is there a better, and less-expensive program available now ?

"The best tool is the one you know the best"
How much time will you spend learning how to apply to
a new software all the tips you know about Zemax after
10 years of use ?
--
Etienne

[Etienne Labaume]

Post by Jamie
the familiar and venture out of your comfort zone? My guess is that
such a move will stimulate your creativity and re-organize your
thinking in new ways. This may make the transition much more profitable
and enjoyable than you think.
Just my thoughts regarding optical design...

Challenging thoughts !