Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

Recommended Posts

Not good progress.

Found registry entry (Name=CygwinRoot Data=C:\Cygwin). Entered C:\cygwin\bin\solve-field into sharpcap settings and started test camera 1, which showed a pic of M42. However, platesolve on tools menu is still greyed-out. Currently waiting for windoze 7 machine (on which I installed AT normally) to fire up so that I can check the contents of the c:\cygwin folder to compare with the contents on the windoze 10 machine. Have a load of screenprints which I will attach if no discrepencies found. This could take a while ...

Link to comment
Share on other sites

  • Replies 189
  • Created
  • Last Reply

This is really weird.

When I installed AT on the windoze 10 machine, the only thing I did differently was not to download the 31.7GB of index files, I already had them so there was no point [even ignoring my desire that this machine never gets near the internet]. The only folder within cygwin that AT created was 'var'. I added the 'astrometry' folder to cygwin (creating the path \usr\share\ in which to hold it.

Checking the windoze 7 installation, there are a whole host of folders as well: bin, dev[+], etc[+], home[+], lib[+] & tmp, and usr contains a lot of other subfolders too.

Currently in the process of copying all this extra data from w7 to add to the w10 machine. Only things I am not copying is 'var' (which was created at installation) and 'astrometry'.

Will see how it goes then.

Link to comment
Share on other sites

Actually, I think I might try first uninstalling and reinstalling AT completely (without the index files) and see if that produces a more complete cygwin folder. If that works, it might be better than just copying things across.

I cannot believe that it would be set up to produce an unworkable copy unless you actually download 31.7GB of files at the time of installation that you may already have as a full and identical copy on another machine. For a total package of 32.6GB that would be crazy!

Link to comment
Share on other sites

OK,  got all the non-data folders installed and also the 41xx data files before going out. Started to run a plate solve on the M42 pic and it was running. Stopped it as I did not expect it to resolve with the 41xx data set but at least we have moved forward. Will copy across the 42xx data files later and give it a proper test. 

Thanks. 

Link to comment
Share on other sites

The bin/etc/var/usr and so on folders are installed by the cygwin setup that runs when AT is installed (They are standard unix/Linux folders - cygwin is just a Linux port under win).

Unless you manually download the install files (which is tedious and hard to get the right ones!) I don't think the cygwin install will be guaranteed to work properly (but you might get v lucky...!!). Manually copying them over from another installation won't guarantee a working install due to PATHs etc.

I think you are best off connecting the PC briefly to do the AT install at a minimum - you don't need to re download index files as these can be copied over manually. 

Also, for the index files, make sure you're not grabbing too many - you don't need all the small field ones unless you are imaging at long FL/small FoV. May reduce space...

Link to comment
Share on other sites

1 hour ago, coatesg said:

The bin/etc/var/usr and so on folders are installed by the cygwin setup that runs when AT is installed

But it didn't ... twice!

Looking at this page, it would appear that, if necessary, I can download an installer on my desktop which would collect all the necessary files, then transfer that to the laptop to perform the actual installation. But I would agree that this would be tedious, if what I have done does work. I am currently copying the 42xx files across and will start to perform some more extensive testing then. Just have to see how it goes. If it works great. If not, I can either do this or revert to PS2, which was seemed to work quite well, although the "approximate" values entered did need to be within a couple of degrees, which is closer than I would have liked.

1 hour ago, coatesg said:

for the index files, make sure you're not grabbing too many

I have a complete set of both. However, once in action I will keep the complete set on the external usb drive and produce subsets applicable for each individual setup (scope+camera), and I can copy the correct one into the "data" folder before setting up each night.

Thanks.

Link to comment
Share on other sites

20 minutes ago, Demonperformer said:

But it didn't ... twice!

At a guess, that's probably because the installer didn't have net access - it needs to be able to see online repositories to grab the relevant installation files to setup the Cygwin instance. The setup program for cygwin is really a package downloader and installer rather than a standalone install package (each Cygwin install may be different wrt to the packages it installs...). If you run a cygwin install on its own, you get the following options - to take the third option, you need to know exactly which packages you need to install to avoid any missing dependencies:

image.png.b05b39cd9e3d4da557b7272822d43a47.png

Link to comment
Share on other sites

6 minutes ago, coatesg said:

At a guess, that's probably because the installer didn't have net access

That would be my guess too.

I can't get my head around that mentality. This is a product whose sole purpose is to do something that could be done on the internet without needing to actually go there. Therefore, it is reasonable to assume that there are situations where you are installing it on machines that do not connect to the internet (ideal 'needed' ground for this software, I would have thought). So why don't they give you either the full package that you need with the installer - or at least tell you what else you will need, where to find it and where to put it [like with the index files]?

With the installer, if you've done option 2 onto the desktop's external usb drive, then doesn't running it again (from said external usb drive) on the laptop with option 3 cover it? [That's how I interpretted the link in my previous post].

Link to comment
Share on other sites

10 minutes ago, Demonperformer said:

I can't get my head around that mentality. This is a product whose sole purpose is to do something that could be done on the internet without needing to actually go there. Therefore, it is reasonable to assume that there are situations where you are installing it on machines that do not connect to the internet (ideal 'needed' ground for this software, I would have thought). 

With the installer, if you've done option 2 onto the desktop's external usb drive, then doesn't running it again (from said external usb drive) on the laptop with option 3 cover it? [That's how I interpretted the link in my previous post].

It makes a lot of sense from a linux point of view (remember Cygwin is used by AT, but is not maintained by the AT developers). Packages have inter-dependencies due to references against required libraries and the way that packages are built requiring certain versions - many of these libraries change on a regular basis due to bug/security fixes, or feature updates. It avoids the need for the AT guys to keep (essentially) their own copy of the cygwin distribution and to have to repackage it with their own software on a regular basis when there are security releases. There are also many instances where cygwin is already present on a client PC - you wouldn't want to force conflicts with any pre-existing Cygwin install, but rather let the relevant cygwin installer sort out itself what is needed.

As regards to the "option 3" method, yes, copying over the complete local package repository from a PC where it is installed, to one where it isn't and then running the cygwin installer against that local copy should work. It is a faff...

Link to comment
Share on other sites

OK, first test complete. Solving the M42 pic that SC has displayed. Log attached, but the upshot is it got it right (although when I first checked the coordinates I thought it had got it wrong because I was expecting -5:18:32 rather than 5:18:32S and so missed the 'S'). It did throw up a warning: Couldn't compute FAST_CWD pointer - not sure if that is (i) of any importance, (ii) a result of just copying cygwin files to the laptop - I'm guessing the anwser to (i) is 'not really' as it gave me the right answer.

Next step, re-read @stash_old's posts above to work out how to give SC an image of my choosing to solve ...

sc log1.txt

Link to comment
Share on other sites

Hi, Concerning the error message:

https://github.com/Alexpux/MSYS2-packages/issues/787

You can change the image in Sharpcap by going to Camera controls on the right hand side of the screen and then selecting the testing controls, and then selecting the image in 'Image'.

I think you're probably better off uninstalling what you've got and installing an up to date version of Astrotortilla, ASPS, or AnSvr. As mentioned before just install the indices you need for your focal length to save space.

 

Link to comment
Share on other sites

1 hour ago, cuivenion said:

I think you're probably better off uninstalling what you've got and installing an up to date version of Astrotortilla, ASPS, or AnSvr. As mentioned before just install the indices you need for your focal length to save space.

Dont think I would bother - Most haven't changed in yrs (last Astrotortilla update - a pre release - was in 2016). Ansvr still talsk about Win XP ! , All Sky Plate Solver I think again 2016. So chances are he is up todate :-)

His error is created by Cygwin  so ,IMHO,just update Cygwin (64bit version) and see how it goes. You change too many items and you risk introducing other problems - unless the package specifically says it need x.

Be careful as Robin says the logs says Astrotortilla but could be using Ansv or ASPS if installed - a bit confusing and something I hope Robin corrects over time. But as he supports the software and would know what to look for in the log it doesn't matter. This shouldn't matter if you have only installed one of the 3 programs.

Link to comment
Share on other sites

2 hours ago, Demonperformer said:

OK, first test complete. Solving the M42 pic that SC has displayed. Log attached, but the upshot is it got it right (although when I first checked the coordinates I thought it had got it wrong because I was expecting -5:18:32 rather than 5:18:32S and so missed the 'S'). It did throw up a warning: Couldn't compute FAST_CWD pointer - not sure if that is (i) of any importance, (ii) a result of just copying cygwin files to the laptop - I'm guessing the anwser to (i) is 'not really' as it gave me the right answer.

Next step, re-read @stash_old's posts above to work out how to give SC an image of my choosing to solve ...

sc log1.txt

One thing I did notice from your log is that you have a wide fov - i.e. a lot of index's - Once you get it working you should look to reduce this if you can(depends on your kit of course) as this will speed things up. Plus maybe start looking into the memory options of Sharpcap which as you are running on a I5 processor with 4gb memory you maybe able to provide more for Sharpcap - maybe a future memory upgrade to compliment your SSD LOL !

It would be nice if the profiles for Sharpcap could allow for different index profiles so that they matched your kit profile at any one time - dont think it does that now. Thats if I understand how the Platesolve works in Cygwin.

Link to comment
Share on other sites

Yes, will definitely need to reduce the number of index files.

A 16'*11' image of NGC 188 (including a green box to confuse the issue) took 15 min to solve. A wide-angle shot of cygnus centred on NGC 7000 is still running after just over an hour (working on objects 41-50 at the moment).

Link to comment
Share on other sites

Calculated my requirements with the aid of CCDcalc & the lightvortex setup page. Not sure that the 20% of minimum isn't a little overkill, but I can always adjust the files when I have some live results. For my main setups the files will be:

Star71+1600: 175'*132' > 175'*26' = 4013>4007
Star71+224: 49'*36' > 49'*7.2' = 4009>4004
8SE+Teleconverter+224: 14'*10.5' > 14'*2.1' = 4005>4000
8SE+Teleconverter+1600: 51'*38' > 51'*7.6' = 4009>4003
Canon+50mm lens: >3 degrees so use 41xx files

This should seriously reduce the time needed to solve!

 

Link to comment
Share on other sites

Must admit I am thinking of moving towards a 2 stage Plate Solve using a 2nd outfit (guide scope maybe) to do the initial blind plate solve when using something my 9.25 inch SCT.  

Be interested on any timings you get when you try your "profiles" out for real!

You can ,of course, "hone" you lists by looking at the log for actual indexes used per profile - I believe this idea was in one of the tutorials I read when I first used Astrotortilla

Link to comment
Share on other sites

Interesting test using same image and blind solving using astrotortilla and ASPS - The latter is faster 22 secs to 32 secs by AstroT!

Both tests were on the same PC and with the same indexes. I even tried tweaking Astrotortilla "downscaling" from 0 to 4 (4 was found to be best ,0 gave 200secs) but ASPS still came out on top. As pictures are the proof see below. Image was from my DSLR Canon 100d and small Frac (627mm)

PC was old Visa Dual Core 2gb memory and SSD

 

testsolve.jpg

L_0100_ISO800_60s__14C.JPG

Link to comment
Share on other sites

13 hours ago, Demonperformer said:

A 16'*11' image of NGC 188 (including a green box to confuse the issue) took 15 min to solve.

 

12 hours ago, stash_old said:

Be interested on any timings you get when you try your "profiles" out for real!

First try caused major panic. I had obviously cropped this sub, because I put the scope/camera settings into CdC calc and determined I needed index files 4004>4010, but it was taking forever to solve. So I went back and checked the log of when I solved it before and it said it solved using index file 4002! Checked the 16'*11' settings - they require 4000>4006 - so restarted the attempt.

My intial thought is that the 'gain' in speed will be much less with a small image. The time taken on indeces 4000>4005 is much greater than the time spent on 4014>4019 and so the saving on losing the high number files will be that much less.

Link to comment
Share on other sites

AT has started not performing as required ... it has placed NAN in the middle of Gemini, with a field size about 1/8 of the actual image. So I am looking at installing either ASPS or ANSVR. Neither of these has run perfectly, but I am hopeful about ANSVR - the installer is about 10x bigger than that of ASPS (may have cygwin with it rather than having to download it separately), and the only (as far as I know) reason that it failed is because I prevented it from downloading all the index files again. This seems like a far more 'usual' problem for installation, so I have emailed the contact email address for advice on how to skip this step successfully.

Ever the optimist!

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.