Astro Tortilla Help Please

[Jkulin]

Hi All,

I had Astro Tortilla working nicely but it has now stopped working and fails to solve, I have done a clean install and re-install yet it still fails.

I ran the log files and this was the clipping from it: -

The images was upload to : - http://nova.astrometry.net/user_images/1844926#annotated

Ignore the slight drift and out of focus, as until I had sorted the problem out then wasn't going to go to the effort until resolved.

Tried again this morning using the file, here's the link to the log file: -
https://drive.google.com/file/d/0B7BhttDpn--2R0ItWTVCYS1LMzA/view?usp=sharing

Here's the link to the cfg file: -

https://drive.google.com/file/d/0B7BhttDpn--2WFFIZnF1TWxBUDg/view?usp=sharing

Here's a link to the cfg image clipping: -

https://drive.google.com/file/d/0B7BhttDpn--2cThCUGNVc0prRFE/view?usp=sharing

And here is a link to the original fit file: -

https://drive.google.com/file/d/0B7BhttDpn--2RklEOUtUUjVHaWM/view?usp=sharing

Can anyone advise?

Many Thanks in anticipation of your help.

[Merlin66]

John,

When you go to that Astrometry link, the image appears to have been solved OK?

Am I missing something??

 

[geoflewis]

Hi John, I downloaded your FITS and ran it through my install of AT and it solve in 30.4 secs using index 4207-04.fits (see log extract below), so not sure why yours failed.

2017-10-31 00:36:23,086 - astrotortilla - INFO - Connecting to camera...
2017-10-31 00:36:23,102 - astrotortilla - INFO - Exposing: 5.00 seconds
2017-10-31 00:36:29,404 - astrotortilla - INFO - Reading image from camera
2017-10-31 00:36:29,404 - astrotortilla - INFO - Solving...
2017-10-31 00:36:30,184 - astrotortilla - INFO - Your group is currently "mkpasswd".  This indicates that your

2017-10-31 00:36:30,295 - astrotortilla - INFO - gid is not in /etc/group and your uid is not in /etc/passwd.

2017-10-31 00:36:30,513 - astrotortilla - INFO - The /etc/passwd (and possibly /etc/group) files should be rebuilt.

2017-10-31 00:36:30,621 - astrotortilla - INFO - See the man pages for mkpasswd and mkgroup then, for example, run

2017-10-31 00:36:30,841 - astrotortilla - INFO - mkpasswd -l [-d] > /etc/passwd

2017-10-31 00:36:30,950 - astrotortilla - INFO - mkgroup  -l [-d] > /etc/group

2017-10-31 00:36:31,167 - astrotortilla - INFO - Note that the -d switch is necessary for domain users.

2017-10-31 00:36:31,276 - astrotortilla - INFO - Reading input file 1 of 1: "/cygdrive/c/Users/Geoff/Downloads/L_Luminance_2017-10-29_23-18-34_Bin1x1_1200s__-15C.fit"...

2017-10-31 00:36:31,605 - astrotortilla - INFO - Computing image percentiles...

2017-10-31 00:36:32,368 - astrotortilla - INFO - Extracting sources...

2017-10-31 00:36:32,479 - astrotortilla - INFO - Downsampling by 2...

2017-10-31 00:36:48,976 - astrotortilla - INFO - simplexy: found 2372 sources.

2017-10-31 00:36:53,891 - astrotortilla - INFO - Solving...

2017-10-31 00:36:55,420 - astrotortilla - INFO - Reading file "/cygdrive/c/Users/Geoff/Documents/Astrophotography/AT Solve/2/L_Luminance_2017-10-29_23-18-34_Bin1x1_1200s__-15C.axy"...

2017-10-31 00:36:55,529 - astrotortilla - INFO - Field 1 did not solve (index index-4209.fits, field objects 1-10).

2017-10-31 00:36:55,638 - astrotortilla - INFO - Field 1 did not solve (index index-4208.fits, field objects 1-10).

2017-10-31 00:36:55,746 - astrotortilla - INFO - Field 1 did not solve (index index-4207-11.fits, field objects 1-10).

2017-10-31 00:36:55,855 - astrotortilla - INFO - Field 1 did not solve (index index-4207-10.fits, field objects 1-10).

2017-10-31 00:36:55,966 - astrotortilla - INFO - Field 1 did not solve (index index-4207-09.fits, field objects 1-10).

2017-10-31 00:36:56,075 - astrotortilla - INFO - Field 1 did not solve (index index-4207-08.fits, field objects 1-10).

2017-10-31 00:36:56,184 - astrotortilla - INFO - Field 1 did not solve (index index-4207-07.fits, field objects 1-10).

2017-10-31 00:36:56,292 - astrotortilla - INFO - Field 1 did not solve (index index-4207-06.fits, field objects 1-10).

2017-10-31 00:36:56,401 - astrotortilla - INFO - Field 1 did not solve (index index-4207-05.fits, field objects 1-10).

2017-10-31 00:36:56,512 - astrotortilla - INFO -   log-odds ratio 166.563 (2.1744e+72), 20 match, 0 conflict, 30 distractors, 24 index.

2017-10-31 00:36:56,621 - astrotortilla - INFO -   RA,Dec = (11.6556,28.1941), pixel scale 0.69849 arcsec/pix.

2017-10-31 00:36:56,730 - astrotortilla - INFO -   Hit/miss: ++++-++++--+-+-+-++-+--+---+-------+-+---------+-+(best)--------------------------------------------------

2017-10-31 00:36:56,838 - astrotortilla - INFO - Field 1: solved with index index-4207-04.fits.

2017-10-31 00:36:56,947 - astrotortilla - INFO - Field 1 solved: writing to file /cygdrive/c/Users/Geoff/Documents/Astrophotography/AT Solve/2/L_Luminance_2017-10-29_23-18-34_Bin1x1_1200s__-15C.solved to indicate this.

2017-10-31 00:36:57,058 - astrotortilla - INFO - Field: /cygdrive/c/Users/Geoff/Downloads/L_Luminance_2017-10-29_23-18-34_Bin1x1_1200s__-15C.fit

2017-10-31 00:36:57,167 - astrotortilla - INFO - Field center: (RA,Dec) = (11.66, 28.19) deg.

2017-10-31 00:36:57,276 - astrotortilla - INFO - Field center: (RA H:M:S, Dec D:M:S) = (00:46:37.350, +28:11:38.475).

2017-10-31 00:36:57,384 - astrotortilla - INFO - Field size: 39.0388 x 29.4355 arcminutes

2017-10-31 00:36:57,493 - astrotortilla - INFO - Creating new FITS file "/cygdrive/c/Users/Geoff/Documents/Astrophotography/AT Solve/2/L_Luminance_2017-10-29_23-18-34_Bin1x1_1200s__-15C.new"...

2017-10-31 00:36:57,604 - astrotortilla - INFO - Creating index object overlay plot...

2017-10-31 00:36:57,727 - astrotortilla - INFO - Creating annotation plot...

2017-10-31 00:36:58,773 - astrotortilla - INFO - Parsing results...
2017-10-31 00:36:59,802 - astrotortilla - INFO - Solved in 30.4s

 

Here is a screen shot of my AT screen showing Scale Min and Scale Max updated by AT from your solved image in case that helps.

Cheers, Geof

[Jkulin]

Thanks Geoff and Merlin, thank-you for the reply, that is exactly what I wanted to determine, so you can both solve but it won't solve with that configuration either on my laptop or my workstation, that's why I upload the config files, any ideas where I should look to resolve the problem?

[Merlin66]

John,

Don't honestly know..... I'd need to think about.

 

[John78]

1 hour ago, Jkulin said:

Thanks Geoff and Merlin, thank-you for the reply, that is exactly what I wanted to determine, so you can both solve but it won't solve with that configuration either on my laptop or my workstation, that's why I upload the config files, any ideas where I should look to resolve the problem?

Did you copy his customs options line that -downsample 2 is important when the image is huge like yours.

[geoflewis]

2 hours ago, Jkulin said:

any ideas where I should look to resolve the problem?

Hi John,

As you will see from my screenshot I don't set any Sigma parameters, but anyway I modified mine to include your --Sigma 70 and it then failed to solve. I reduced the Sigma to 50 and 20 and it then solved on both with improving solve times at around 25 seconds, so you could try running with reduced --Sigma, which will increase the number of items that Astrotortilla idendifies as stars for its solve routine. I also find that --Downsample 2 helps, so you might want to try that also. Please let us know how you get on.

Regards, Geof

[Jkulin]

Geof, John & Merlin, just doing a bit of fire fighting with clients for the next few hours, but really really appreciate your testing and help.

[kens]

I can see something that for some reason is a common issue with AT. Namely repetition of the scale parameters "-H" and "-L" which seems to throw astrometry

I've only got the log to go by but I reckon if you look at your screen you'll have "Scale minimum" and "Scale maximum" set but also in "Custom options" you'll have the -H and -L parameters which are the same thing. So remove them from Custom options.

Your values (0.48 0.67 degwidth) tally with Geof's solve

 

 

[kens]

On closer inspection its even worse. One set of parameters is reversed with -H 0.48 and -L 0.67

[geoflewis]

8 minutes ago, kens said:

also in "Custom options" you'll have the -H and -L parameters which are the same thing. So remove them from Custom options.

Well spotted, I missed that, but totally agree that the dupication (-H and -L) should be removed from custom options. Also great catch that the values are also reversed. I check the deg values against my arcminwidth values to check they were the same, but completely missed that they are reversed. Removing them is the right way to go.

[Jkulin]

Thanks Gents, finally managed to have a look now after a manic day, I inputted exactly as you suggested and have attached a screen capture and unfortunately it failed to solve, I'm going to try on my laptop and see if there is anything stopping it locally, I'll revert shortly....I hope!

[geoflewis]

Hi John,

Remove the --plot-scale 0.25 as you don't need that, its just that I like to save an image of the solve data, so use --no-plots instead. Also the downsample should be --downsample 2 not -downsample 2 (note the double dash)

[Jkulin]

Thanks so much Geof, it solved that, brilliant, I imagine it will be even quicken once it knows the position of the mount from EQMod? So these parameters will hopefully sort things for the 8"RC, any ideas what the parameters should be for the new Esprit 100ED that arrives tomorrow?

I truly appreciate all of your help, ta muchly.

[geoflewis]

3 minutes ago, Jkulin said:

it solved that, brilliant, I imagine it will be even quicken once it know the position of the mount from EQMod? SO these parameters will hopefully sort things for the 8"RC, any ideas what the parameters shoudl be for the new Esprit 100ED that arrives tomorrow?

Great news John, I knew that we'd get there .

The only parameters that you'll need to change are the Scale Minimum and Scale Maximum settings as they need to reflect the image scale of your optical configuration. Please note however you will also need to download the relevant Cygwin Index files for AT to use to solve these images. I have several configurations so have a large range of Indexes permanently downloaded. They way I calculate my Scale Max/Min is to upload an image to astrometry.net and get the on-line solve engine to solve it, which will then tell you the image scale in arcmins. You can then set your Scale Max and Scale Min in AT to bracket say 50% either side of your image scale, e.g. for your image that we'd been solving, I think I set an initial range of 30 arcmin to 60 arcmin. Note I also use the Scale Refinement of 0.1 as what this does is prompt AT to update the Scale Max and Scale Min to the optimum solved image scale. I then save those settings as a new config file (Save-->Settings) giving it a meaningful file name e.g. 8"RC-CameraModel, Esprit 100ED-CameraModel, etc.

The specifications for your Esprit 100 and my TSAPO100Q are not that disimilar so FWIW when using it with my QSI583 CCD camera the Scale Max is 116 arcminwidth and and Scale Min is 96 arcminwidth. When I use my Canon 600D DSLR with the TSAPO100, then those values change to 144' (max) and 120' (min). I hope these help, but as a I said before its probably best just to set a fairly wide range based on astronmetry.net solve and then get AT to calculate and update the optimum settings for your imaging configurations.

Good luck, Geof

[Jkulin]

Thanks Geof, I had downloaded all 32GB of the indexes, but all that did was delay even further searches.

Being thick, where did you get your Scale Minimum and Scale Max from when you solved it with Astrometry.net, as I had previously been using 0.48 and 0.67, yet your scales and adjustments were totally different, I only ask for when I upload an image from the new Esprit, I need to know which index to download and where to obtain the figures from?

[kens]

Also, you can remove the "--downsample 2" parameter in Custom options and just put 2 in the Downscaling field. That then adds "--downsample 2" to the solver command line.

[geoflewis]

9 hours ago, Jkulin said:

where did you get your Scale Minimum and Scale Max from when you solved it with Astrometry.net, as I had previously been using 0.48 and 0.67, yet your scales and adjustments were totally different, I only ask for when I upload an image from the new Esprit, I need to know which index to download and where to obtain the figures from?

Hi John,

If you look at the astrometry.net solved details for your image (panel on the right) it privides some statistics including image size in arcmins - for the one you shared taken with your 8" RC those are Size 39 x 29.4 arc min., so I just set the Scale Min Max field so include a buffer range either side of the largest of those to values, typically half to 150%, so something like 20 arcmin (Scale min and 60 arcmin (Scale max). AT then solves the image (will take a bit longer, but I do this during the day so not wasting image time) and due to inclusion of Scale Refinement 0.1, it recalculates and updates the values for optimum solving. Once I've got these values I save the new config with meaningful name and then rerun with log open so that I can see which Index solved the image for this configuration. Its probably a long winded approach, but I found AT quite tricky to configure so like to take small steps to be sure I catch everything that I need.

Regarding the values that you were using (0.48 & 0.67) those were degrees, so you were using the default scale degwidth, where as I use scale arcminwidth mainly because I then don't need to convert astrometry.net values from arcmin to degrees. (NB 0.48 deg = 29 arcmin; 0.67 deg = 40 arcmin so broadly the same range as I was using).

Another way to get the correct indexes is to refer to the Index table provided by Astrometry.net via the following link.

http://astrometry.net/doc/readme.html

I'd recommend including a range of Indexes, so for your 8" RC maybe Indexes 4206-* to 4208-*, but you could just run with 4207-*

I hope this helps. Once you get AT tamed you'll find it one of the best tools for AP there is; it has saveed me literally hours compared with trying to get on faint targets visually.

Cheers, Geof

[geoflewis]

8 hours ago, kens said:

Also, you can remove the "--downsample 2" parameter in Custom options and just put 2 in the Downscaling field. That then adds "--downsample 2" to the solver command line.

Thnaks Ken, I did not know that these were the same - I had wondered about it, but once I got AT working I stopped messing with the parameters. Cheers, Geof

[Jkulin]

Geof and everyone, thanks for all your kind help.

My Esprit arrived today, so lets see what fun I have with it as there are some clear spells due, so fingers crossed.

[geoflewis]

1 hour ago, Jkulin said:

Geof and everyone, thanks for all your kind help.

My Esprit arrived today, so lets see what fun I have with it as there are some clear spells due, so fingers crossed.

Excellent, have fun. Looking forward to seeing some images from it.

[Jkulin]

Everything was going fine tonight and the Astrophotography tools crashed, so rebooted the systems and re did everything as I knocked the Dec handle :-(

After redoing everything I went to do Astro Tortilla again and it came up with this exception as the window below, has anyone any ideas?

 

[geoflewis]

12 minutes ago, Jkulin said:

Everything was going fine tonight and the Astrophotography tools crashed, so rebooted the systems and re did everything as I knocked the Dec handle :-(

After redoing everything I went to do Astro Tortilla again and it came up with this exception as the window below, has anyone any ideas?

 

Hi John,

I have never seen that error, but the clue maybe in the last line - it looks like you lost the ASCOM connection, so AT couldn't sync to your mount. Have you tried rebooting everything. You may need to shut down then fire up your laptop again rather than a soft restart, to be sure to clear the USB ports.

Hope you get it sorted.

Geof

[Jkulin]

Thanks Geof will try that again and let you know.

[Jkulin]

Thanks once again, a soft reboot did not indeed fix, but a hard reboot did, but in the meantime it clouded over :-( I only needed 2 more blue subs for a set.

Scopenights says it will clear by 12 so worth sitting up and just grabbing those last few subs.