Surprising sensitivity of solver to field dimension parameters

[mlco...@gmail.com]

I have an image that solves locally with this command line:

$ solve-field -O -L 5 -H 10 -u aw --parity neg --ra 353 --dec 19.9372 --radius 10 --downsample 2 eqpeg_Light_012.fits

with Size: 8.13 x 4.6 arcmin (see Job 9423007 at nova.astrometry.net)

But if I change "-H 10" to "-H 8" the image does not solve. So even if "-L" parameter is greater than the lower limit, setting "-H" parameter less than the upper limit blinds the solver.

Is this possibly true in general?

[Eric SIBERT]

I have faced a similar issue with large field views (19°). I also
thought that using a narrow range would reduce search time. So I tried
18°-20°. Indeed, it was not solving. So, I have been back to 10-20°
range. Something to do with index selection? i.e. you need to use
footprints that are significantly smaller than the size of your picture
so that you have a chance that the footprint fit in your picture.

Eric

> --
> You received this message because you are subscribed to the Google
> Groups "astrometry" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to astrometry+...@googlegroups.com
> <mailto:astrometry+...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/astrometry/c6394b1c-c202-4457-8cdf-f4f39e07e5b8n%40googlegroups.com <https://groups.google.com/d/msgid/astrometry/c6394b1c-c202-4457-8cdf-f4f39e07e5b8n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Dustin Lang]

Sorry, I'm confused.  You're saying

-L 5 -H 10 works

but

-L 5 -H 8 does not work

And the real size is 8.13 wide.

That looks like the correct behavior to me :)

cheers,

dustin

--

You received this message because you are subscribed to the Google Groups "astrometry" group.

To unsubscribe from this group and stop receiving emails from it, send an email to astrometry+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/c6394b1c-c202-4457-8cdf-f4f39e07e5b8n%40googlegroups.com.

[Mark Copper]

I was working with a new configuration and didn't know the focal length.  Maybe a rule of thumb: when in doubt, use a wider scale.

[Bryan]

Dustin

I think his query is that why does it solve an image 4.6 x 8.13 with -L=5 (i.e.  greater than the smallest dimension), yet will not solve the same image when -H=8 (less than the largest dimension).

Bryan

[Dustin Lang]

Hi,

Ahh, okay, so the "-u aw" aka "--scale-units arcminwidth" literally refers to the WIDTH of the image, ie, the size of the X coordinate.  In this case, that dimension of the image is ~8.13 arcminutes.  So the "-L" and "-H" arguments must bracket that value.  The height of the image is totally ignored.

cheers,

dustin

To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/d7a3f466-42d4-47c2-be36-8fa2bf2a8036n%40googlegroups.com.