GoTo : wrong coordinates for targets

[soundfan]

I'm from Germany, so same time zone.

Indeed I was thinking too, that this behaviour started with the last change to daylight saving time, but I can't tell for sure.

[admin]

Hi folks, thanks for continuing to contribute to this - something very odd going on.

I just tested the GOTO of Regulus here and get this in my log

Code: Select all

Debug   10:17:17.454603 #1  Sequencer : Info      : 'Converting RA=10:08:22.3,Dec=+11:58:00 (J2000) position to RA=10:09:40.4,Dec=+11:50:53 (JNOW)' while running step Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00

The entry for Regulus in the default SharpCap bright star catalog is

Code: Select all

HIP 49669,α Leo,32 Leo|Regulus,Cor Leonis,Rex,Kabeleced|Star|10.139531|11.967207|1.36|||

And the RA=10.139531 and Dec=11.967207 should be read as 10:08:22.3 and 11:58:00 as I am seeing.

The values that @soundfan sesms to have recorded in his log are (J2000) RA=10.195277778 and Dec=11.96666667.

The two possibilities I am imagining are

1) A custom catalog is being used that has bad data in it (less likely)
2) Somehow SharpCap is reading the default catalog wrongly

But if it is option 2, how? A common error in this sort of thing is to forget about languages where the ',' character is used as a decimal point, but I dealt with that long ago, and anyway, the errors in the values don't really make sense for that mistake. Even so, it may be worth changing the computer language to English which uses the '.' and test to see if the error goes away. I will test with French/German later to see if that introduces an error.

Right....

now I think I see it - writing all that explanation and suddently it makes sense.

When the Goto tool is used, the catalog target is selected and the co-ordinates taken from that. This is then used to create a temporary sequence that actually does the work of going to the target - the temporary sequence is recorded in the log and looks fine

Code: Select all

SEQUENCE
    SET PLATESOLVE/FOCUS SETTINGS EXPOSURE 4
        MOUNT GOTO RA=10:08:22,3,Dec=+11:58:01
        TARGETNAME "HIP 49669 (α Leo, 32 Leo, Regulus, Cor Leonis, Rex, Kabeleced)"
    END SET
END SEQUENCE

But, note that the fractional arc-seconds in RA in the target has a comma as the decimal separator. That sequence is encoded as text in the sequence editing language, so to get the co-ordinates to move to, SharpCap has to re-interpret the text 'RA=10:08:22,3,Dec=+11:58:01'. That re-interpretation is going wrong because of the ',' decimal separator

Code: Select all

Debug  	19:33:41.424859	#1 	Sequencer : Starting  : Move the mount to (RA,Dec) RA=10:08:22,3,Dec=+11:58:01 (RA=10:11:43,0,Dec=+11:58:00)

Note that the co-ordinates in brackets which are the reinterpreted version does not match the original text version. The RA in particular is wrong and I will have to go look at the code to see why it happens, as it isn't obvious. The Dec being one second difference is just a rounding error.

The reason this started happening was that I changed the code to show the tenths of a second of RA a while back (previously only whole seconds were used). This is really needed for RA as 1 second of RA is equal to 15 seconds of arc (only 24 hours of RA vs 360 degrees in a full circle), but that meant that the ',' for decimal separator got added.

I will try to fix this today and include the fix in today's update. Thanks again for digging into this enough for me to spot the pattern.

Robin

[soundfan]

Just amazing, Robin!

Incredible how fast and successful your support and problem solving is.

Looking forward to seeing the next update. I'll test asap and report back!

CS Mike

[admin]

Oooh, interesting....

I had already got some code that handled this problem in the case of French number formats, so when I tested those, it works, but it breaks for German ones!

I wonder why...

Robin

[soundfan]

Uhh, didn't know that French and German have different number formats.

One learns new stuff every day...

CS Mike

[admin]

Hi,

no, nor did I before today!

Turns out that the 'digit grouping character' which is used to separate groups of 3 digits to make them more readable is a comma in German and a space in French. So, the number 'one billion' would be

English : 1,000,000,000
French : 1 000 000 000
German : 1.000.000.000

It's the last one that causes the trouble, since the co-ordinates were actually being converted to text using the english way, so the RA would have been '10:08:22.3'. When trying to interpret the text '22.3' as a number in German, you get 223, since the dot is ignored as a digit grouping character. In French and English you get 22.3 (even though French uses a comma for decimal separator, it accepts the dot).

Anyway, I have this fixed now and tested on a machine set to German UI and it seemed to work. Please give the new version 4.1.12151 a go when you have a chance.

cheers,

Robin

[soundfan]

Anyway, I have this fixed now and tested on a machine set to German UI and it seemed to work. Please give the new version 4.1.12151 a go when you have a chance.

cheers,

Robin

Will do, right this afternoon!

Thank you for your lighting fast help in advance!

CS Mike

[soundfan]

Done testing.

Everything is working perfectly again!

Thanks again for your perfect support, Robin!

CS Mike

[admin]

Hi Mike,

glad to hear that it worked! Now all you need is some clear skies!

Robin

[Cepadues]

I just tested the GOTO of Regulus here and get this in my log

Code: Select all

Debug   10:17:17.454603 #1  Sequencer : Info      : 'Converting RA=10:08:22.3,Dec=+11:58:00 (J2000) position to RA=10:09:40.4,Dec=+11:50:53 (JNOW)' while running step Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00

The entry for Regulus in the default SharpCap bright star catalog is

Code: Select all

HIP 49669,α Leo,32 Leo|Regulus,Cor Leonis,Rex,Kabeleced|Star|10.139531|11.967207|1.36|||

And the RA=10.139531 and Dec=11.967207 should be read as 10:08:22.3 and 11:58:00 as I am seeing.

The values that @soundfan sesms to have recorded in his log are (J2000) RA=10.195277778 and Dec=11.96666667.

I can confirm the same behavior, with regulus :

Code: Select all

Sequencer : Info      : 'Converting RA=10:11:43,0,Dec=+11:58:00 (J2000) position to RA=10:13:01,1,Dec=+11:50:50 (JNOW)' while running step Move the mount to (RA,Dec) RA=10:08:22,3,Dec=+11:58:01 (RA=10:11:43,0,Dec=+11:58:00)

Which gives RA = 10.1952778 and DEC = 11.9666667

I'm not sure to understand the issue correctly, how does 'RA=10:08:22,3,Dec=..." gets interpreted as 'RA=10:11:43' ?

Incredible how fast and successful your support and problem solving is.

You don't get such support from almost any company, your dedication and problem solving skills are admiring!

Just for a baseline, I switched to english format in windows regional settings and repeated the test with regulus (still on the old version) :

Code: Select all

18:24:34 Starting  : 
18:24:34 Starting  : Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00)
18:24:35 Info      : 'Converting RA=10:08:22.3,Dec=+11:58:00 (J2000) position to RA=10:09:40.4,Dec=+11:50:53 (JNOW)' while running step Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00)
18:24:38 Info      : 'Mount slew complete, waiting for mount to settle.' while running step Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00)
18:24:48 Completed : Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00)
18:24:48 Starting  : Set imaging target name to 'HIP 49669 (α Leo, 32 Leo, Regulus, Cor Leonis, Rex, Kabeleced)'
18:24:48 Completed : Set imaging target name to 'HIP 49669 (α Leo, 32 Leo, Regulus, Cor Leonis, Rex, Kabeleced)'
18:24:48 Completed : 

This time it was spot on! As was your diagnostic, I'll make further testing with the new version and report back.

Thanks again for the help and support

Edit :

The update worked like a charm :

Code: Select all

18:49:14 Starting  : 
18:49:14 Starting  : Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00)
18:49:14 Info      : 'Converting RA=10:08:22,3,Dec=+11:58:00 (J2000) position to RA=10:09:40,4,Dec=+11:50:53 (JNOW)' while running step Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00)
18:49:26 Info      : 'Mount slew complete, waiting for mount to settle.' while running step Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00)
18:49:36 Completed : Move the mount to (RA,Dec) RA=10:08:22.3,Dec=+11:58:01 (RA=10:08:22.3,Dec=+11:58:00)
18:49:36 Starting  : Set imaging target name to 'HIP 49669 (α Leo, 32 Leo, Regulus, Cor Leonis, Rex, Kabeleced)'
18:49:36 Completed : Set imaging target name to 'HIP 49669 (α Leo, 32 Leo, Regulus, Cor Leonis, Rex, Kabeleced)'
18:49:36 Starting  : Wait for next live view frame
18:49:40 Completed : Wait for next live view frame
18:49:40 Starting  : Plate solve a single frame, sync the mount and re-goto the target co-ordinates
18:49:58 Info      : 'Mount movement complete, waiting for mount to settle.' while running step Plate solve a single frame, sync the mount and re-goto the target co-ordinates
18:49:58 Completed : Plate solve a single frame, sync the mount and re-goto the target co-ordinates
18:49:58 Completed : 

I've also tried it with Greek, Russian and Spannish locals and no issues to report!