Hi Stephane,

> I don't think it is catalog because when I use PRiSM, I have
> better results with UCAC4 than with USNO A2 (yes in 5'
> x 5', you have about only stars brighter than mag 10 ;-) ).

That may be true for Prism but read these comments from the PinPoint author:

http://forums.dc3.com/showthread.php?4694-Downloading-and-Using-PinPoint-Reference-Catalogs-(Updated-May-2018)

Then take a look here about someone having a problem with UCAC4 and PinPoint:

http://forum.mainsequencesoftware.com/t/unable-to-solve-using-pinpoint-and-ucac4-catalog/2228

For fields larger than 15'x15' the GSC can be used. For smaller fields, USNO A2.0 is recommended.

>But in my case, it is no : 5' x 5' : even with parameters at maximum (image >scale tolerance at 80%),

If you know the image scale you should not increase image scale tolerance. That will only make solving slower and possibly unreliable.

> the spiral search takes ... less than one second
> (I thought well it was not my Ryzen 5 2600 which slow down computing

Previously you said PinPoint was taking a long time. That happens when PinPoint is doing a spiral search internally. When that happens it will almost always be faster for PinPoint to do the All-Sky solve to get the initial coordinates which PinPoint will then use to do a more accurate plate solve.

> About "max solve time" with APPM (so PinPoint), I don't understand this
> parameter : I don't understand PinPoint stops researchs before the limit
> time (the last time, I have time limit at 10 sec and the search has stopped
> at 0.689 sec

If PinPoint failed early then there should be an error message. What error message does APPM indicate was returned from PinPoint?

If you are having APPM save your failed images, can you zip a few of them along with the corresponding APPM log file and post or email me a dropbox link (or similar).

> one hour to produce a pointing model of 80 fields.

You should be able to solve about 2 images per minute with APPM (120/hour) if the download time is not too long. This is when using the option to use the last solution offset to for the following image.

-Ray Gralak
Author of APCC (Astro-Physics Command Center): http://www.astro-physics.com/index.htm?products/accessories/software/apcc/apcc
Author of PEMPro V3: https://www.ccdware.com
Author of Astro-Physics V2 ASCOM Driver: https://www.siriusimaging.com/apdriver

> -----Original Message-----
> From: ap-gto@... [mailto:ap-gto@...]
> Sent: Thursday, June 13, 2019 12:55 AM
> To: ap-gto@...
> Subject: Re: [ap-gto] APPM and All-Sky solves
>
>
>
> Hi Ray,
>
>
> Because I spent a lot of time to understand parameters of PinPoint and APPM (This is the same), I think I have
> understood the failure of my solves.
>
> I don't think it is catalog because when I use PRiSM, I have better results with UCAC4 than with USNO A2 (yes in 5'
> x 5', you have about only stars brighter than mag 10 ;-) ).
>
>
>
> > (but I have already remarked such a solve is less accurate with these
> > two last solutions so I would like prefer 1/).
>
> Actually, that is incorrect. PinPoint will use the All-Sky coordinates as if APPM had originally given them to it,
> and then more accurately solve. Thus there is the same accuracy as if PinPoint had solved without having to use
> All-Sky coordinates. In effect this dramatically shortens what might have been a very long duration spiral search!
>
>
> My problem is here ! Your last remark is OK for classic instruments .... A medium - big field with refractor : in fact no
> problem, even with little search parameters on PinPoint, sucess is at each solving !
>
> But in my case, it is no : 5' x 5' : even with parameters at maximum (image scale tolerance at 80%), the spiral
> search takes ... less than one second (I thought well it was not my Ryzen 5 2600 which slow down computing ... and
> it is not a competitor SSD but a SSD). With a refractor, I let you imagine the failure when you slew from Zenith to
> 20° above horizon (As APPM makes at the beginning of session) or when you slew from East to West ... Yes, I have
> not the perfect optical tube, with 20' when I slew from West to East for example, but it is the only reason I use a
> pointing model and not to boast me the target must stay in a hole of 3 arcsec without guiding (I don't care this !) ...
>
>
> So when you write APPM uses a spiral search, it is the search of PinPoint and not APPM which calculates each
> time a new center of image in order to simulate a spiral. It works on PRiSM in about 2 seconds because I can define
> a limit of spiral search (time and often area) and it is software which calculates the spiral search ... About "max
> solve time" with APPM (so PinPoint), I don't understand this parameter : I don't understand PinPoint stops researchs
> before the limit time (the last time, I have time limit at 10 sec and the search has stopped at 0.689 sec ...).
>
> And why not to use all-Sky : because, even my 5'x5', the research time is about ... 100 to 150 sec ! Even with near
> coordinates center, with nova.astrometry.net (internet) or in local with ansvr. With 100 fields, I let you imagine time
> of model ... Sorry to compare, but with PRiSM, it takes only two seconds at each field and it works on 95% of case
> (without all-sky search type of astrometry.net) because this is the software which produce spiral search. But PRiSM
> produce a txt file that is wrote only for each field : theorcial coordinates (hour angle, dec) and the same in apparent
> coordinates so difficult to use at a APPM file ... Sorry to compare also with the simplicity of making model with my
> 10µ mount : in the mode "making pointing model", mount points a field and for each field pointed, if a sync telescope
> is send, it is the real center of image and so on ... moreover, pointing model is refined field after field so after 10
> fields, target is near the center of image and with the same little field for me (spectroscopical studies also), it takes
> one hour to produce a pointing model of 80 fields.
>
>
> But I'm going astray ... So, I am reduced to :
> 1/ use less than 20 fields for my pointing model, as now, I have produced in more one hour ... After all, as I auto-
> guide ...
> 2/ Spent a lot of time to try to make the long file of APPM with results sending by PRiSM (not sure of sucess).
> 3/ Ask you APPM makes spiral research instead PinPoint and respect max solve time.
>
> Please, don't forget : the pointing model is often here for "atypical" instruments and not for couple 16803 - FSQ 106
> !
>
> Regards
>
> Stephane
>
>
>
>
>
>
> -Ray Gralak
> Author of APCC (Astro-Physics Command Center): http://www.astro-
> physics.com/index.htm?products/accessories/software/apcc/apcc
> Author of PEMPro V3: https://www.ccdware.com
> Author of Astro-Physics V2 ASCOM Driver: https://www.siriusimaging.com/apdriver
>
> > -----Original Message-----
> > From: ap-gto@... [mailto:ap-gto@...]
> > Sent: Wednesday, June 12, 2019 8:39 AM
> > To: ap-gto@...
> > Subject: Re: [ap-gto] APPM and All-Sky solves
> >
> >
> >
> > Hi Ray,
> >
> > Thank you for your reply and your advice about USNO A2.0. I will change catalog.
> >
> > As I was a very old question for me, please, could you confirm I have fine understood :
> >
> > 1/ Even I don't check "Try ALL-Sky Plate solving if available", APPM configures PinPoint to do a spiral
> search until
> > "Max Solve Time" was reached ? (I didn't know that ... so if it is the case, please, add it in the already good
> help of
> > software).
> >
> > 2/ If I check "Try ALL-Sky Plate solving if available", APPM configures also PinPoint to do a spiral search ...
> But if it
> > fails after "Max Solve Time", APPM configures PinPoint in order to use the option one can find in Visual
> PinPoint for
> > using an ALL-Sky solution, as nova.astrometry.net by internet connection or local ansvr (but I have already
> > remarked such a solve is less accurate with these two last solutions so I would like prefer 1/).
> >
> > Regards
> > Stephane
> >
> >
> > Le mer. 12 juin 2019 à 05:06, 'Ray Gralak (Groups)' groups3@... [ap-gto] > gto@...> a
> > écrit :
> >
> >
> >
> >
> >
> > Hi Stephane,
> >
> > > I have some questions about the "all-sky" of APPM..
> > > I use Pinpoint V6 for solving plate.
> > >
> > > My field is little : 5' x 5'. I use UCAC4 catalog.
> >
> > I don't think that's the right catalog for you to use for telescope pointing calculations because it only goes
> > down to magnitude 16 and no brighter than mag 10. I think you would be better served by USNO A2.0
> because it
> > goes deeper *and* also contains brighter stars.
> >
> > > But most of the time, solving is bad even "All-Sky Max Solve Time" is at 300 sec !
> > > (I think it will be more simple if APPM uses a spiral research in case of failure)
> >
> > It would be better if you use USNO A2.0. Also, APPM configures PinPoint to do a spiral search so you are
> > limited by the speed of PinPoint. It helps if you have a CPU with a high single thread performance and a
> fast SSD to
> > store the catalog.
> >
> > Also, if APPM solves a single all-sky image, images that follow it will solve much quicker if you enable the
> > option "Use last plate solve's offset as hint for next plate solve" in APPM.
> >
> > >> For good solving in all-Sky mode, I would like to know if I must check some
> > > boxes in (Visual) PinPoint ?
> >
> > Nope, that won't make any difference. I think you just need a more appropriate sky catalog.
> >
> > > > Why must I to have an internet connection ?
> >
> > Nope, you don't need an internet connection.
> >
> > > Pinpoint doesn't work on All-sky without internet connection ?
> >
> > Sure, PinPoint can work all-sky without and internet connection. You can define the IP address and port of
> > your local machine, but you must setup your own all-sky solution (there are a couple available). Just do an
> internet
> > search for PinPoint+AllSky.
> >
> > -Ray Gralak
> > Author of APCC (Astro-Physics Command Center): http://www.astro-
> > physics.com/index.htm?products/accessories/software/apcc/apcc
> > Author of PEMPro V3: https://www.ccdware.com
> > Author of Astro-Physics V2 ASCOM Driver: https://www.siriusimaging.com/apdriver
> >
> > > -----Original Message-----
> > > From: ap-gto@... [mailto:ap-gto@...]
> > > Sent: Tuesday, June 11, 2019 9:14 AM
> > > To: ap-gto@...
> > > Subject: [ap-gto] APPM and All-Sky solves
> > >
> > >
> > >
> > > Hello Ray,
> > >
> > >
> > > I have some questions about the "all-sky" of APPM..
> > > I use Pinpoint V6 for solving plate.
> > >
> > > My field is little : 5' x 5'. I use UCAC4 catalog.
> > >
> > > So, for some (a lot of) extremal fields, solving is not directly possible with such a little field. With my
> > planetarium
> > > software (PRiSM), solve is quick from a depart point and a spiral resolution with internal resolution but
> text
> > file of
> > > APPM's measures is too technical and so too long to reproduce ...
> > > So, for couple APPM / PinPoint v6 : I must use "Try All-Sky Plate Solving if available" if I want hope to
> > solve about
> > > half of fields. But most of the time, solving is bad even "All-Sky Max Solve Time" is at 300 sec ! (I think it
> > will be
> > > more simple if APPM uses a spiral research in case of failure)
> > > In fact, I try to understand why with APPM/PinPoint V6, all-sky solving is very long and most of cases
> bad,
> > and why
> > > with the same image and with PRiSM, all-sky solving is fast and good in more 90% of cases ...
> > >
> > > For good solving in all-Sky mode, I would like to know if I must check some boxes in (Visual) PinPoint ? I
> > don't think
> > > because it is independant ...
> > > Why must I to have an internet connection ? Pinpoint doesn't work on All-sky without internet connection
> ?
> > So
> > > Pinpoint always uses Astrometry.net in all-sky mode or all-sky mode can be used with PinPoint and
> without
> > > Astrometry.net ?
> > >
> > > Regards
> > > Stephane
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>
>

Hi Ray,

Because I spent a lot of time to understand parameters of PinPoint and APPM (This is the same), I think I have understood the failure of my solves.

I don't think it is catalog because when I use PRiSM, I have better results with UCAC4 than with USNO A2 (yes in 5' x 5', you have about only stars brighter than mag 10 ;-) ).

> (but I have already remarked such a solve is less accurate with these
> two last solutions so I would like prefer 1/).

Actually, that is incorrect. PinPoint will use the All-Sky coordinates as if APPM had originally given them to it, and then more accurately solve. Thus there is the same accuracy as if PinPoint had solved without having to use All-Sky coordinates. In effect this dramatically shortens what might have been a very long duration spiral search!

My problem is here ! Your last remark is OK for classic instruments ... A medium - big field with refractor : in fact no problem, even with little search parameters on PinPoint, sucess is at each solving !

But in my case, it is no : 5' x 5' : even with parameters at maximum (image scale tolerance at 80%), the spiral search takes ... less than one second (I thought well it was not my Ryzen 5 2600 which slow down computing ... and it is not a competitor SSD but a SSD). With a refractor, I let you imagine the failure when you slew from Zenith to 20° above horizon (As APPM makes at the beginning of session) or when you slew from East to West ... Yes, I have not the perfect optical tube, with 20' when I slew from West to East for example, but it is the only reason I use a pointing model and not to boast me the target must stay in a hole of 3 arcsec without guiding (I don't care this !) ...

So when you write APPM uses a spiral search, it is the search of PinPoint and not APPM which calculates each time a new center of image in order to simulate a spiral. It works on PRiSM in about 2 seconds because I can define a limit of spiral search (time and often area) and it is software which calculates the spiral search ... About "max solve time" with APPM (so PinPoint), I don't understand this parameter : I don't understand PinPoint stops researchs before the limit time (the last time, I have time limit at 10 sec and the search has stopped at 0.689 sec ...).

And why not to use all-Sky : because, even my 5'x5', the research time is about ... 100 to 150 sec ! Even with near coordinates center, with nova.astrometry.net (internet) or in local with ansvr. With 100 fields, I let you imagine time of model ... Sorry to compare, but with PRiSM, it takes only two seconds at each field and it works on 95% of case (without all-sky search type of astrometry.net) because this is the software which produce spiral search. But PRiSM produce a txt file that is wrote only for each field : theorcial coordinates (hour angle, dec) and the same in apparent coordinates so difficult to use at a APPM file ... Sorry to compare also with the simplicity of making model with my 10µ mount : in the mode "making pointing model", mount points a field and for each field pointed, if a sync telescope is send, it is the real center of image and so on ... moreover, pointing model is refined field after field so after 10 fields, target is near the center of image and with the same little field for me (spectroscopical studies also), it takes one hour to produce a pointing model of 80 fields.

But I'm going astray ... So, I am reduced to :

1/ use less than 20 fields for my pointing model, as now, I have produced in more one hour ... After all, as I auto-guide ...

2/ Spent a lot of time to try to make the long file of APPM with results sending by PRiSM (not sure of sucess).

3/ Ask you APPM makes spiral research instead PinPoint and respect max solve time.

Please, don't forget : the pointing model is often here for "atypical" instruments and not for couple 16803 - FSQ 106 !

Regards

Stephane

 

-Ray Gralak
Author of APCC (Astro-Physics Command Center): http://www.astro-physics.com/index.htm?products/accessories/software/apcc/apcc
Author of PEMPro V3: https://www.ccdware.com
Author of Astro-Physics V2 ASCOM Driver: https://www.siriusimaging.com/apdriver

> -----Original Message-----
> From: ap-gto@... [mailto:ap-gto@...]
> Sent: Wednesday, June 12, 2019 8:39 AM
> To: ap-gto@...
> Subject: Re: [ap-gto] APPM and All-Sky solves
>
>
>
> Hi Ray,
>
> Thank you for your reply and your advice about USNO A2.0. I will change catalog.
>
> As I was a very old question for me, please, could you confirm I have fine understood :
>
> 1/ Even I don't check "Try ALL-Sky Plate solving if available", APPM configures PinPoint to do a spiral search until
> "Max Solve Time" was reached ? (I didn't know that ... so if it is the case, please, add it in the already good help of
> software).
>
> 2/ If I check "Try ALL-Sky Plate solving if available", APPM configures also PinPoint to do a spiral search ... But if it
> fails after "Max Solve Time", APPM configures PinPoint in order to use the option one can find in Visual PinPoint for
> using an ALL-Sky solution, as nova.astrometry.net by internet connection or local ansvr (but I have already
> remarked such a solve is less accurate with these two last solutions so I would like prefer 1/).
>
> Regards
> Stephane
>
>
> Le mer. 12 juin 2019 à 05:06, 'Ray Gralak (Groups)' groups3@... [ap-gto] <ap-gto@...> a
> écrit :
>
>
>
>
>
> Hi Stephane,
>
> > I have some questions about the "all-sky" of APPM.
> > I use Pinpoint V6 for solving plate.
> >
> > My field is little : 5' x 5'. I use UCAC4 catalog.
>
> I don't think that's the right catalog for you to use for telescope pointing calculations because it only goes
> down to magnitude 16 and no brighter than mag 10. I think you would be better served by USNO A2.0 because it
> goes deeper *and* also contains brighter stars.
>
> > But most of the time, solving is bad even "All-Sky Max Solve Time" is at 300 sec !
> > (I think it will be more simple if APPM uses a spiral research in case of failure)
>
> It would be better if you use USNO A2.0. Also, APPM configures PinPoint to do a spiral search so you are
> limited by the speed of PinPoint. It helps if you have a CPU with a high single thread performance and a fast SSD to
> store the catalog.
>
> Also, if APPM solves a single all-sky image, images that follow it will solve much quicker if you enable the
> option "Use last plate solve's offset as hint for next plate solve" in APPM.
>
> >> For good solving in all-Sky mode, I would like to know if I must check some
> > boxes in (Visual) PinPoint ?
>
> Nope, that won't make any difference. I think you just need a more appropriate sky catalog.
>
> > > Why must I to have an internet connection ?
>
> Nope, you don't need an internet connection.
>
> > Pinpoint doesn't work on All-sky without internet connection ?
>
> Sure, PinPoint can work all-sky without and internet connection. You can define the IP address and port of
> your local machine, but you must setup your own all-sky solution (there are a couple available). Just do an internet
> search for PinPoint+AllSky.
>
> -Ray Gralak
> Author of APCC (Astro-Physics Command Center): http://www.astro-
> physics.com/index.htm?products/accessories/software/apcc/apcc
> Author of PEMPro V3: https://www.ccdware.com
> Author of Astro-Physics V2 ASCOM Driver: https://www.siriusimaging.com/apdriver
>
> > -----Original Message-----
> > From: ap-gto@... [mailto:ap-gto@...]
> > Sent: Tuesday, June 11, 2019 9:14 AM
> > To: ap-gto@...
> > Subject: [ap-gto] APPM and All-Sky solves
> >
> >
> >
> > Hello Ray,
> >
> >
> > I have some questions about the "all-sky" of APPM.
> > I use Pinpoint V6 for solving plate.
> >
> > My field is little : 5' x 5'. I use UCAC4 catalog.
> >
> > So, for some (a lot of) extremal fields, solving is not directly possible with such a little field. With my
> planetarium
> > software (PRiSM), solve is quick from a depart point and a spiral resolution with internal resolution but text
> file of
> > APPM's measures is too technical and so too long to reproduce ...
> > So, for couple APPM / PinPoint v6 : I must use "Try All-Sky Plate Solving if available" if I want hope to
> solve about
> > half of fields. But most of the time, solving is bad even "All-Sky Max Solve Time" is at 300 sec ! (I think it
> will be
> > more simple if APPM uses a spiral research in case of failure)
> > In fact, I try to understand why with APPM/PinPoint V6, all-sky solving is very long and most of cases bad,
> and why
> > with the same image and with PRiSM, all-sky solving is fast and good in more 90% of cases ...
> >
> > For good solving in all-Sky mode, I would like to know if I must check some boxes in (Visual) PinPoint ? I
> don't think
> > because it is independant ...
> > Why must I to have an internet connection ? Pinpoint doesn't work on All-sky without internet connection ?
> So
> > Pinpoint always uses Astrometry.net in all-sky mode or all-sky mode can be used with PinPoint and without
> > Astrometry.net ?
> >
> > Regards
> > Stephane
> >
> >
> >
>
>
>
>