Hi all,

> The magenta problem appears once when the program is first started and
> on the first images loaded. Of a set of three images the first images
> appear magenta, the rest are ok. This happens regardless if I switch to
> another image set. Next time I load the images the magenta problem has
> gone away.
>
> Then I exit AFW and restart and load an image and again the first is
> colored magenta och the rest are ok.

I can confirm what Jorgen said - but only for the images made with our ST-9E!

When I open images made with our ST-6, all of them appear in magenta ...
although the the first image usually suffers the hardest from the problem. What
I mean is hard for me to describe with words (especially in English), but it can
be seen under:

    http://www.bph.ruhr-uni-bochum.de/~axelm/tso/temp/astrometrica.jpg  (131kB)

The images were opened in order of their Filenames: x13520_001.fit,
x13520_002.fit, etc...

Interesting is, that I can work several sessions with AFW withour problems and
then one day suddenly the magenta-problem appears. The only thing I can do is to
delete the exe-file reinstall it.

I don't know if it helps, but what I have noticed is, that the problem mostly
(not always) appears when I have downloaded a new mpcorb.dat-file.

-Axel
--
---------------------------------------------------
Turtle Star Observatory              Obs-Code: #628
c/o Axel Martin
Friedhofstr. 15
D-45478 Mülheim-Ruhr

Tel: ++49 - 208 - 5 51 51

http://www.turtlestar.de    Turtle Star Observatory
http://spica.turtlestar.de      Zeitschrift "SPICA"
---------------------------------------------------

Hi all! The magenta problem appears once when the program is
first started and on the first images loaded. Of a set of three images the first
images appear magenta, the rest
are ok. This happens regardless if I switch to another image set. Next time I
load the images the magenta problem has gone away.

Then I exit AFW and restart and load an image and again the first is colored
magenta och the rest are ok.

Best regards

Jorgen Danielsson
Sweden

----- Original Message -----
From: "Herbert Raab" <herbert.raab@...>
To: <Astrometrica@yahoogroups.com>
Sent: Friday, July 06, 2001 6:05 PM
Subject: Re: [Astrometrica] accuracy improvement


>
> Stefano Sposetti writes:
> > For pre-occultation purposes (measurements of relative bright stars
> > and asteroids ) I need an accuracy of 0.001s in RA and 0.01s in
> > declination in the MPC-format-string.
>
> I have to add several new settings in preperation of the proposed,
> new format for submitting observations to the MPC. If it is not
> very urgent, I will add that option with these changes (which will
> be made in one of the next versions, no matter when the new format
> will be acitvated). Hope that's OK...?
>
>   Best greetings,
>    Herbert
>
> ========================================================
> Astrometrica Beta Site: http://www.astrometrica.at/Beta
> Post: Astrometrica@yahoogroups.com
> Unsubscribe: Astrometrica-unsubscribe@yahoogroups.com
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>

Andrea Pelloni wrote:
> Herbert, now we have the need to blink big images (3600x3600
> pixels), and we found that Win Astrometrica gives as
> result "Uncorrected Parameter".
It would be a great help if you could specify the exact error
message, as displayed by the software. It should include a
hint in which procedure the error occured. (Ideally, send me
a screenshot of the message. But please send it privately to
me, as the mailing list will strip off any attachments.)

> We have found that the software is unable to read the .fts
> of NEAT, and this could be very interesting if corrected,
> as you know many of us make recovery of their MPs on old
> plates and NEAT CCD images, so ...
The problem is that the NEAT files contain values between
-32768 and +32767 (i.e., signed integers), but no value for
BZERO that would shift the values in the positive value range
(0..65535). In my humble optinion, this is not conform with
the description of the FITS standard. But anyway, there is an
undocumented feature in Astrometrica to overcome this problem:
Add the line "BZERO=32768" to the Settings section of the INI
file, and Astrometrica will add that offset. (If the FITS header
contains a value for BZERO, the sofware will, of course, use
that value.)

> The software has the option "Load Dark Frame" and " Load Flat
> Field", but there are no commands for performing the needed
> operations.
> Will be added later?
If you load the dark frame and/or the flat field image *before*
you load the images, the software will automatically apply the
image calibration when the images are loaded.

> I and many other friends have found that in the .log file
> there are correctly shown in parenthesis many stars with such
> parameters bigger than the given ones, but at the same time
> there are other stars that have been accepted for astrometry/
> photometry even with errors bigger than the given ones
The software first does a data reduction including all stars.
It then rejects the stars with residuals larger than the limit
you have specified. A few stars that were just below the limit
in the first solution (containing all stars) might now have
residuals just above the limit. To avoid a deadlock situation,
where two subsets exclude each other with each new iteration,
the software stops after a limited number of iterations, and
thus it can include a few stars with residuals somewaht above
the limit specified by the user.

> Moreover, in order to make smaller the .log files, is it possible
> to have them with ONLY the accepted stars, discarding the "wrong"
> ones?
I think it's good when the LOG file is a complete documentation
of the measuerement, and it should therefore includes the rejected
stars also. (As an example: If you see a list of 100 reference
stars with the USNO-A2.0, you might think it's a good solution.
But if you also see that, at the same time, 200 stars have been
rejected, you may have a second look...)

> Many of my friends who are testing your software have found the
> same problem that I and another beta-tester previously informed
> to you, I mean the magenta color of the images.
Although some users have reported that kind of problem, I still
was not able to reproduce the problem on my PC, making it very
difficult to locate the error. If anyone is able to find a way
how this problem can be reproduced, please let me know immediately.

   Best Regards,
    Herbert Raab

Herbert, now we have the need to blink big images (3600x3600 pixels), and we
found that Win Astrometrica gives as result "Uncorrected Parameter". When we
try on images with a smaller number of pixels, there are no problems (up to
about 700x700 pixels).
I remembered that at the beginning of beta-testing, your software was unable
to read the files produced with KAF400, that were 768x512 pixels, because
the software was isuued for ST-6 images, so you modified very quickly the
software to work with the bigger CCD dimensions.
Can you make it again, arriving now at the numbers we need?
It could be very nice for many of us.

We have found that the software is unable to read the .fts of NEAT, and this
could be very interesting if corrected, as you know many of us make recovery
of their MPs on old plates and NEAT CCD images, so ...

The software has the option "Load Dark Frame" and " Load Flat Field", but
there are no commands for performing the needed operations.
Will be added later?

I have put in the settings the request to reject all the stars with
astrometric error (as per USNOA.2) bigger than  0.8 arcsec, and with a
photometric error bigger than 0.8 mag. I and many other friends have found
that in the .log file there are correctly shown in parenthesis many stars
with such parameters bigger than the given ones, but at the same time there
are other stars that have been accepted for astrometry/photometry even with
errors bigger than the given ones, as you can see in the enclosed files.

Moreover, in order to make smaller the .log files, is it possible to have
them with ONLY the accepted stars, discarding the "wrong" ones?

Many of my friends who are testing your software have found the same problem
that I and another beta-tester previously informed to you, I mean the
magenta color of the images. One of them told me, and I checked obtaining
the same results, that when we load an image that opens in magenta, if we
cancel it and then load it again, almost always the newly loaded image is
correctly in B&W. This for your information.

I inform you moreover that we are doing a good checking of the old and new
residuals (previously obtained with Astrometrica 325 and now reduced again
with WIN Astrometrica).
The first analysis says that there is a good implementation in astrometrical
accuracy with WIN Astrometrica, and we hope to have in a short time a
documented analysis to send to you.
Thank you in advance.
By soon

Andrea Pelloni
157 Frasso Sabino

----- Original Message -----
From: Herbert Raab <herbert.raab@...>
To: <Astrometrica@yahoogroups.com>
Sent: Friday, July 06, 2001 6:05 PM
Subject: Re: [Astrometrica] accuracy improvement


>
> Stefano Sposetti writes:
> > For pre-occultation purposes (measurements of relative bright stars
> > and asteroids ) I need an accuracy of 0.001s in RA and 0.01s in
> > declination in the MPC-format-string.
>
> I have to add several new settings in preperation of the proposed,
> new format for submitting observations to the MPC. If it is not
> very urgent, I will add that option with these changes (which will
> be made in one of the next versions, no matter when the new format
> will be acitvated). Hope that's OK...?
>
>   Best greetings,
>    Herbert
>
> ========================================================
> Astrometrica Beta Site: http://www.astrometrica.at/Beta
> Post: Astrometrica@yahoogroups.com
> Unsubscribe: Astrometrica-unsubscribe@yahoogroups.com
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>


[Non-text portions of this message have been removed]

Stefano Sposetti writes:
> For pre-occultation purposes (measurements of relative bright stars
> and asteroids ) I need an accuracy of 0.001s in RA and 0.01s in
> declination in the MPC-format-string.

I have to add several new settings in preperation of the proposed,
new format for submitting observations to the MPC. If it is not
very urgent, I will add that option with these changes (which will
be made in one of the next versions, no matter when the new format
will be acitvated). Hope that's OK...?

   Best greetings,
    Herbert

Hi Herbert, hi all,

For pre-occultation purposes (measurements of relative bright stars and
asteroids ) I need an accuracy of 0.001s in RA and 0.01s in declination in the
MPC-format-string.
Do you feel is there a big job to implement this click-able option in
Win-Astrometrica?

With my best regards,
Stefano Sposetti
143-Gnosca

Chris Spratt wrote:
> Any plans to add additional star catalogs to the final release?
> Perhaps the Tycho 2 catalog could be added.

Not Tycho 2, but rather the UCAC. While Tycho 2 has simply not
enough reference stars for most purposes, the UCAC has a density
compareable to the GSC, yet it has a precision similar to Tycho 2.
In fact, the UCAC positions are even better than Tycho for the
fainter stars (11mag and below) found in both cataloges.

The USNO people say that a preliminary version of the UCAC, that
covers the southern hemisphere and the aequatorial/ecliptical
region, will be available in early 2002. Astrometrica should be
able to acess the data by that time (given that the USNO does
not change the file format shortly before the release).

Astrometry of bright objects, where the error of the reference
stars dominates over the centroiding error, should really benefit
from the UCAC. However, for faint targets, where the centroiding
dominates the error, the situation will not change much...

   Greetings,
    Herbert Raab

Any plans to add additional star catalogs to the final release?
Perhaps the Tycho 2 catalog could be added.

Hallo,

als Anfänger benutze ich zunächst das Tutorial und erlebe einen Misserfolg. Ich
habe keine Einstellungen geändert und bekomme bei der Identifizierung der Sterne
auf der Beispielaufnahme nur 8 Identifizierungen,obwohl es ja eigentlich mehr
sein müssten. Ich benutze den USNO-A2.0.
Was ist zu tun?

Beste Grüße
	 von
	 Jörg!


Homepage für Amateurastronomie und Roboter

http://home.t-online.de/home/joerg.schirmer/index.htm

Herbert,

          The image I am talking about is one of the ones you supplied
for the tutorials.  It is the third image in the sequence.

          I'm thinking that perhaps the "ini" file picked up some of our
values for our telescopes, rather than using the default ones that
came with Astrometrica.  But I reloaded the entire set of files and
told the computer to overwrite any old files.  So I think that would have
gone back to your basic "ini" file.

          Let me try it again on a different computer and see if I get any
different results.

                  Paul

At 09:20 PM 6/25/2001 +0200, you wrote:

>Paul Weissmann writes:
> > We manually corrected the scale and pointing, and when we then told
> > the program to remeasure the field, it blew up with either a floating
> > point divide or the Microsoft "blue screen of death."
>Does not sound very good... Can you put the image on some ftp server
>where I can download it? If you give me the settings you used, I
>will try to find what happens here...
>
> > Also, my student commented that the tutorial referred to an older
> > version of Astrometrica, so he had some problems when things did
> > not appear as predicted by the tutorial.
>I changed some forms, but did not yet update the screenshots in
>the tutorial. But it's not much difference, so I thought it will
>still be easy to follow the tutorials. I will update the images in
>the documentation when I find some time...
>
>   Regards,
>    Herbert
>
>
>
>========================================================
>Astrometrica Beta Site: http://www.astrometrica.at/Beta
>Post: Astrometrica@yahoogroups.com
>Unsubscribe: Astrometrica-unsubscribe@yahoogroups.com
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Bob writes:
> I note that in reduction of some images, the reduction is successful
> (no warning) using "Reference Star Matching" set at 80 stars, whereas
> some images require at least 160.

I fear I have to go into the details of the reference star
identification:

The software first finds the stars in an image and stores them in a
list, and then reads the reference star data from the catalogue and
stores this in another list. To find the match between the stars on
the image and the reference stars in the catalogue, it takes the n
brightest stars from one list, buils all possible pairs of stars,
and tries to find that pair in among the pairs of the n brightest
stars in the other list. (Where n ist the number you set.)

Now consider a crowded field, and assume you are using the (incomplete)
USNO-SA2.0 reference stars catalog. The n brightest (and unsatured)
stars from the CCD image will probably be brighter stars in the range
14mag to 16mag. The USNO-SA2.0 contains prfereably faint reference
stars in the 16mag to 19mag range. For obvious reasons, the software
won't be able to find matching pairs.

The situation is better if you use a (more or less) complete catalogue,
like the full USNO-A2.0, or sometime in the (not too distant future) the
UCAC. The match usually works with a smaller number of stars here. But
again, there are pitfalls. Consider a crowded field were you did take
a deep image. Say that there are 20 stars so bright that they are
satured on your deep image. If you try a match using 20 stars, the
program will not find the 20 brightest stars from the catalogue on
the image, as they are satured (and therefore not detected) here.

It apparently helps a lot if you can limit the ranges in which the focal
length, the image orientation, and the pointing vary to small values
(though still as large as necessary). The smaller these ranges, the
more relieable the match will be.

In fact, the reference star identification is probably the most complex
routine in the software. It's the weak link, but if it fails, you could
still do a manual match.

Hope this helps...

   Herbert

Paul Weissmann writes:
> We manually corrected the scale and pointing, and when we then told
> the program to remeasure the field, it blew up with either a floating
> point divide or the Microsoft "blue screen of death."
Does not sound very good... Can you put the image on some ftp server
where I can download it? If you give me the settings you used, I
will try to find what happens here...

> Also, my student commented that the tutorial referred to an older
> version of Astrometrica, so he had some problems when things did
> not appear as predicted by the tutorial.
I changed some forms, but did not yet update the screenshots in
the tutorial. But it's not much difference, so I thought it will
still be easy to follow the tutorials. I will update the images in
the documentation when I find some time...

   Regards,
    Herbert

Herbert,

          I have a new summer student who just arrived and I introduced
him to the latest Beta version.  He found, and I checked this, that when
doing the moving asteroid detection, the program found lots of stars but
only identified a few.  We then tried a manual fit and found that both the
plate scale and center point were way off.  We manually corrected the
scale and pointing, and when we then told the program to remeasure the
field, it blew up with either a floating point divide or the Microsoft "blue
screen of death."  By the way, to fit the field we had to use a focal length
of ~3600 and a base angle of ~358 degrees.  We still didn't have it quite
right because the pixels do not seem to be of equal size.  All of this was
using the third of the moving object detection images.

          Also, my student commented that the tutorial referred to an older
version of Astrometrica, so he had some problems when things did not
appear as predicted by the tutorial.

          This all kind of surprises me because when I ran the tutorial about
4 months ago, it all worked very smoothly.

          Hope this information is useful to you.
                  Paul Weissman

At 04:12 PM 6/23/2001 -0500, Len Casady wrote:
>Can it be forced to find it by tweaking the Settings?  If you zoom in on the
>first image, you'll notice that the star that it is by has a grey circle
>partially around it to the left.  The asteroid is much more defined than the
>thing it locked onto.  Another example also in the first image, just above
>the asteroid and to the right, another object it locked on to that is
>envolved with a star.
>
>When the opposition point is not near the heart of the Milky Way it will be
>less of a concern  :)
>
>
>Thanks
>Len
>
>----- Original Message -----
>From: "Herbert Raab" <herbert.raab@...>
>To: <Astrometrica@yahoogroups.com>
>Sent: Saturday, June 23, 2001 8:07 AM
>Subject: Re: [Astrometrica] auto id
>
>
> >
> > Len,
> >
> > I just had a quick look at your images. When I used an aperture
> > radius of 4, the asteroid is only detected (marked by a grey cricle)
> > on the third image. It's too close to the star in the other two
> > images. When I tried an aperture radius of 3, the atsroid is detected
> > (marked by a grey cricle) in images two and three, but not on image 1.
> >
> > Of course, if the asteroid is not detected in one of the images,
> > the software is not able to identify it as a moving object.
> >
> > To recognice a moving object, the object, it must be present on all
> > images, it must be of similar brightness on all images, and it
> > must show a consistent, linear motion.
> >
> > Hope this helps...
> >
> >   Herbert
> >
> >
> > ========================================================
> > Astrometrica Beta Site: http://www.astrometrica.at/Beta
> > Post: Astrometrica@yahoogroups.com
> > Unsubscribe: Astrometrica-unsubscribe@yahoogroups.com
> >
> >
> > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
> >
> >
>
>
>========================================================
>Astrometrica Beta Site: http://www.astrometrica.at/Beta
>Post: Astrometrica@yahoogroups.com
>Unsubscribe: Astrometrica-unsubscribe@yahoogroups.com
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

I note that in reduction of some images, the reduction is successful
(no warning) using "Reference Star Matching" set at 80 stars, whereas
some images require at least 160.  In the images tested, the number of
stars available seem to be fairly equal, and the images were taken the
same evening (ST7E F3.5, 28 x 18 minutes).  The images requiring a
minimum of 160 reference stars were at -20 DEC, images requiring 80 at
+6dec)- using USNO SA2 with mag lower limit 25 and upper limit 8.
Search radius tested at both 4 and 5 with the same results.

I am using a 350MhZ board w 128m ram, 100 bus, and when ref is set at
160, the reduction time is significant.  Any comments would be
appreciated.

Thanks

Bob

I've had this happen when to asteroid is close to or almost involved
with a star. I resorted to manually doing each image.

--- In Astrometrica@y..., ice@n... wrote:
> I have been playing with the Windows beta.  Very nice!  Using the
> Moving Object Detection is proving quite perplexing.  I can't seem
to
> get it to id what would seem bright easy targets.  I've played with
> the settings and just can't seem to get it to "see" them.  Faint
junk
> on the other hand easy to get  :)  I have a zip of one such
> asteroid.  It has three images with a text file with the scope focal
> length and ccd chip data.  The file size is 300K
>
> www.neosoft.com/~ice/autoid.zip
>
> Anyone want to take a look at the images and see if they can find
the
> right settings or understand why it's being detected.  The asteroid
> is about right in the center of the field.
>
> I have two other sets of asteroids similar to the one above that
also
> evade auto detection if interested in those as well.
>
> Thanks
> Len Casdy

> Can it be forced to find it by tweaking the Settings?
Decreasing the aperture radius helps a little bit. As I noted before,
with an aperture radius of 3, the asteroid is found on image 2, while
it is not found when setting the radius to 4.  However, setting the
radius too small will influence the centroiding and flux calculation,
so don't set it too small.

Crowded stars fields are a problem for moving object detections.
The sofware will miss real objects and find false movers. It's
still the human brain that has the last word here...

   Herbert

Can it be forced to find it by tweaking the Settings?  If you zoom in on the
first image, you'll notice that the star that it is by has a grey circle
partially around it to the left.  The asteroid is much more defined than the
thing it locked onto.  Another example also in the first image, just above
the asteroid and to the right, another object it locked on to that is
envolved with a star.

When the opposition point is not near the heart of the Milky Way it will be
less of a concern  :)


Thanks
Len

----- Original Message -----
From: "Herbert Raab" <herbert.raab@...>
To: <Astrometrica@yahoogroups.com>
Sent: Saturday, June 23, 2001 8:07 AM
Subject: Re: [Astrometrica] auto id


>
> Len,
>
> I just had a quick look at your images. When I used an aperture
> radius of 4, the asteroid is only detected (marked by a grey cricle)
> on the third image. It's too close to the star in the other two
> images. When I tried an aperture radius of 3, the atsroid is detected
> (marked by a grey cricle) in images two and three, but not on image 1.
>
> Of course, if the asteroid is not detected in one of the images,
> the software is not able to identify it as a moving object.
>
> To recognice a moving object, the object, it must be present on all
> images, it must be of similar brightness on all images, and it
> must show a consistent, linear motion.
>
> Hope this helps...
>
>   Herbert
>
>
> ========================================================
> Astrometrica Beta Site: http://www.astrometrica.at/Beta
> Post: Astrometrica@yahoogroups.com
> Unsubscribe: Astrometrica-unsubscribe@yahoogroups.com
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>

Len,

I just had a quick look at your images. When I used an aperture
radius of 4, the asteroid is only detected (marked by a grey cricle)
on the third image. It's too close to the star in the other two
images. When I tried an aperture radius of 3, the atsroid is detected
(marked by a grey cricle) in images two and three, but not on image 1.

Of course, if the asteroid is not detected in one of the images,
the software is not able to identify it as a moving object.

To recognice a moving object, the object, it must be present on all
images, it must be of similar brightness on all images, and it
must show a consistent, linear motion.

Hope this helps...

   Herbert

I have been playing with the Windows beta.  Very nice!  Using the
Moving Object Detection is proving quite perplexing.  I can't seem to
get it to id what would seem bright easy targets.  I've played with
the settings and just can't seem to get it to "see" them.  Faint junk
on the other hand easy to get  :)  I have a zip of one such
asteroid.  It has three images with a text file with the scope focal
length and ccd chip data.  The file size is 300K

www.neosoft.com/~ice/autoid.zip

Anyone want to take a look at the images and see if they can find the
right settings or understand why it's being detected.  The asteroid
is about right in the center of the field.

I have two other sets of asteroids similar to the one above that also
evade auto detection if interested in those as well.

Thanks
Len Casdy

Dear Beta-Tester,

the first public Beta version, V 4.0.6.171, is now available
for download. As compared to the previously available Version
4.0.6.162, a bug (occasional access violation after clicking
on an image for a manual measurement) has been corrected.

As usual, the new version is available for download from
this site:
             http://www.astrometrica.at/Beta/

At the same time, the Astrometrica mailing list, which has,
until now, only been used to announce new versions, is now
open for all users to post messages. I hope that it will be
used to exchange information and expiriences with the new
software. I will continue to announce new versions trough
this list.

Those who do not want to remain on the mailing list should
send a message to: <Astrometrica-unsubscribe@yahoogroups.com>

    Thank you for your attention!
      Herbert Raab

An error (invalid array index) in the calculation of the
uncertainty of the measured position can cause the program
to crash under certain circumstances. I have just put a
new, corrected version on the Astrometrica web page for
download (V4.0.6.161).

Beta Testers who have already downloaded the recent
version 4.0.6.159 may want to get the EXE file only:

   http://www.astrometrica.at/Beta/V406.zip

Those who have not yet downloaded Version 4.0.6 will find
the new version under the usual URL:

   http://www.astrometrica.at/Beta/

Thank you for your attention,

   Herbert Raab

Dear Beta-Tester,

I am sorry for the delay, but I have been quite busly lately,
both in my job an during the preperations for the ScienceWeek
here in Austria. Anyway, a new Beta-Version of Astrometrica
(V 4.0.6.159) is finally available for download.

A number of new features suggested by Beta-Testers are included.
A description of the new features is availabe here:

    http://www.astrometrica.at/Beta/News.html

The complete package (including demo images and documentation),
as well as the updated EXE file only, are available for download
from this site:

    http://www.astrometrica.at/Beta/

If no major problem will be reported, this version will porbably
be released for public beta testing in June.

Thank you for your attention!
   Herbert Raab

Dear Beta-Tester,

a new Beta-Version of Astrometrica (V 4.0.5.118) is available
for download. The two problems reported earlier this week have
been fixed, and a number of new features are included. A detailed
description of the new features is availabe here:

   http://www.astrometrica.at/Beta/News.html

The complete package (including demo images and documentation),
as well as the updated EXE file only, are available for download
from this site:

   http://www.astrometrica.at/Beta/

    Thank you for your attention!

      Herbert Raab

Dear Beta-Testers,

unfortunately, there are two bugs in V4.0.4.108, both concerning
the reference star identification.

First, the reference star identification may fail if the CCD images
are shifted by more than a few pixels against each other. While the
program will be able to match the reference stars in the first image,
it may fail to find the reference stars in further images. This error
will also affect the manual reference star identification, so there
is no work around at this time.

Second, the dialog for manual reference star alignment will not
appear when a text-window (for example, the LOG-file) is open.
The program will go idle will not respond to any user action.

I am currently working to fix both problems. I hope that a corrected
version will be available later this week.

I am sorry for any inconveniences.

Thank you for your patience,

   Herbert Raab

Dear Beta Tester,

if you have downloaded Version 4.0.4.108 of Astrometrica, please
call up the 'Settings' from in the 'File' menu. On the 'Settings'
page, please set the fit order for the Plate Constants. (The default
setting is 'Linear Fit').

The internal meaning of this setting has been changed, and if you
do not reset this value with the new version, you may get the
following error message:

   PlateConst: Invalid Fit Order
   (D:\Programme\Borland\Delphi5\Projects\Source\Astrometry.pas, line 87)

Thank you for your attention!

   Herbert Raab

Dear Beta-Tester,

unfortunately, the links for downloading the new version 4.0.4.108
were set incorrectly on the Astrometrica Beta Site yesterday.
Downloading resulted either in an old version (4.0.3.90), or in
an error message that the requested file is not available.

I have now fixed this, and version 4.0.4.108 should now be available
for download at the Astrometrica Beta site:

   http://www.astrometrica.at/Beta/

I am very sorry for any inconvenience.

    Herbert Raab

Dear Beta-Tester,

a new Beta-Version of Astrometrica (V 4.0.4.108) is available
for download. Several new features are included. Most notably,
situations in which the software was not able to match the
reference stars with the stars found in the CCD frames are
handeled much better now. Also, some minor bugs have been
corrected. A detailed description of the new features is
availabe here:

   http://www.astrometrica.at/Beta/News.html

The complete package (including demo images and documentation),
as well as the updated EXE file only, are available for download
from this site:

   http://www.astrometrica.at/Beta/

Thank you for your attention!

   Herbert Raab

Dear Beta-Tester,

a new Beta-Version of Astrometrica (V 4.0.3.90) is available for
download. This version corrects an error which caused the alignment
of larger images (larger than about 512x512 in size) to fail,
both for blinking and during the data reduction process.

Furthermore, the relevant program settings for the data reduction
are now written to the log-file. Some additional minor corrections
are also included in this version.

The complete package (including demo images and documentation),
as well as the updated EXE file only, are available for download
form this site:

   http://www.astrometrica.at/Beta/

Thank you for your attention!

   Herbert Raab

I received a message from a Beta-tester today whith several
qestions. I though that the answers are possibly of general
interest:

> I get the error message "Identification of reference stars probably
> in error!", and the coordinates are also way way off.
This error message means that the software could not find a perfect
match between the stars on your image and the reference stars, so it
did take the best match it could find - in most cases, the solution is
not valid.
Please note that the images from the tutorials were mirrored, as they
were taken at prime focus of a 0.6m f/3.3 telescope, i.e a telescope
system with an odd number of mirrors. If your own images are not mirror
reversed, make sure that the check box for flipping the images is not
checked in the program settings.
If that alone will not cure your problems, I'd suggest that you measure
the image with "Astrometrica classic". Note the central coordinates of
the image (displayed after the data reduction, together with the reference
star residuals), and the focal length and orientation of the images (using
the command 'Information' in the 'Measure' menu), and use these values.

> I have nice images of a know asteroid, but the software can't
> detect as a moving object.
Has the asteroid been detected as a star, i.e, is there a blue circle
around it? In that case, the software thinks it is a star because it
was moving less than the value specified for the 'stationary limit'
between each image.
If the minor planet is not circled at all in one of the images, the
software has not detected the asteroid there as any kind of object
all. Click on the asteroid after measuring the image, and check the
information displayed. Is the flux less than the value specified in
the 'minimal flux' setting? Is the RMS larger than the maxiumum allowed
in the settings? Is the signal/noise ratio less (or close) to the limit?

For further information on the various program settings, please read the
'Settings' section in the documentation. I hope that you will be able to
figure out what these settings mean. You may also want to try different
settings on the tutorials first, so see what happens when you change one
parameter or another.

Thank you for your attention!

   Herbert

Dear Beta Testers,

a new pre-release version of Astrometrica is now available for
download. Please visit this site for further instructions:

http://www.astrometrica.at/Beta

The ZIP-file (1.17 MB) that is available for download at this
site includes the software, as well as a set of images and part
of the MPCOrb databse. In addition, you will need the USNO-SA2.0
or the USNO-A2.0 star catalogue on CD-ROM.

   Thank you for your attention!

    Herbert Raab