Why are the reported astrometric positions (ra, dec) so low precision?

[augustfly]

I have added a file where the astrometric solution is apparently correct in the scale and rotation of the image, but where the astrometric center of the image (FITS WCS key, CRVAL1, CRVAL2) are shifted from the correct values.

The solution file contains the position to a very low degree of precision. I speculate that this is the source of the error.

https://github.com/adsass/astrometry/commit/16a9df59dae0fb7271bbb68903649c7cd3e6b2ae#diff-14da587b595bed0114ed9ec2bd8ff4f0R3

[augustfly]

So I took a look to figure out how the *.txt files were being written.

https://github.com/adsass/astrometry/blob/master/astrometry.net/astrometry.py#L38-L62

https://github.com/adsass/astrometry/commit/ab703ab0835e4d5990b9737d1956a2c1e8919603#diff-a186b04c98edbe3bbca41f2623684304R32

This looks to me like it is parsing the stdout from the solve-field command. Is that right?

Were any other "solved" products saved from these runs? It may have created a bunch of solved files according to the documentation. The solutions in those solved files may be much more precise than those stripped from the stdout.

For example, consider this thread:

https://groups.google.com/d/topic/astrometry/O_5ZdPOPQjI/discussion

It suggests that my assumption that the reference pixel is the center pixel is wrong because we did not "run solve-field with the "--crpix-center"".

So I need to know if we have any other saved outputs from the solve-field.

[apepe]

Hey Gus- yes we have other materilals and they can be found in the folder astrometry_results. However, I am having a hard time moving all these files from scratch to goodman_lab home (it times out). Can you access Odyssey and inspect them yourself? We should try to move these out of scratch asap

On Thu, Dec 19, 2013 at 6:02 PM, August Muench notifications@github.comwrote:

So I took a look to figure out how the *.txt files were being written.

https://github.com/adsass/astrometry/blob/master/astrometry.net/astrometry.py#L38-L62

ab703ab#diff-a186b04c98edbe3bbca41f2623684304R32https://github.com/adsass/astrometry/commit/ab703ab0835e4d5990b9737d1956a2c1e8919603#diff-a186b04c98edbe3bbca41f2623684304R32

This looks to me like it is parsing the stdout from the solve-field command. Is that right?

Were any other "solved" productshttp://astrometry.net/doc/readme.html#output-filessaved from these runs? It may have created a bunch of solved files according to the documentation. The solutions in those solved files may be much more precise than those stripped from the stdout.

For example, consider this thread:

https://groups.google.com/d/topic/astrometry/O_5ZdPOPQjI/discussion

It suggests that my assumption that the reference pixel is the center pixel is wrong because we did not "run solve-field with the "--crpix-center"".

So I need to know if we have any other saved outputs from the solve-field.

— Reply to this email directly or view it on GitHubhttps://github.com/adsass/astrometry/issues/3#issuecomment-30975224 .

[augustfly]

I do not have access to our Odyssey data, scratch, goodman_lab, or otherwise. Do you know what I need to do to get access? Could you write up the directory structure of the scratch/goodman_lab directories so I know where to look once I get in there?

[apepe]

ok- let's get you access to the directory...

On Fri, Dec 20, 2013 at 9:30 AM, August Muench notifications@github.comwrote:

I do not have access to our Odyssey data, scratch, goodman_lab, or otherwise. Do you know what I need to do to get access? Could you write up the directory structure of the scratch/goodman_lab directories so I know where to look once I get in there?

— Reply to this email directly or view it on GitHubhttps://github.com/adsass/astrometry/issues/3#issuecomment-31012785 .