Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - mcgillca

Pages: [1] 2
Hi - I've been wondering about buying a Mac mini 2018 for Pixinsight use. However, looking at the benchmarks, I'm a bit surprised by the swap speed.

Users have reported Blackmagic reports disk read and write speeds for the 1TB disk at around 2.5GB/s (its a bit slower for write speeds for smaller disks), but the PI benchmarking data reports only 764 MB/s.

I was a little surprised since my wife's iMac with a 51GB SSD PI benchmark reports speeds of around 2.4GB/s, which is in line with the Blackmagic speed test.

I then experimented with a RAM disk on my wife's computer - Blackmagic reported about 5GB/s, whilst the PI benchmarking reported only about 1.5GB/s.

There is clearly something more complicated going on here since I would have expected better results with the RAM disk.

In each case, I used 4 temp file instances, so this shouldn't have changed.

Any thoughts?


General / Unmaximising the real time preview window
« on: 2018 November 15 02:25:45 »
Several times recently, I've found I've unintentionally maximised the real time preview window.

When this happens, i haven't been able to find a way to either close the window or resize it.

The best I've been able to do is to move the red line (dividing image area from icons) to the right to get more room for images behind.

Anyone know how to either close or resize the real time preview window when this happens?

Thank you,


Bug Reports / Flat correction in One Shot colour work flow?
« on: 2017 December 27 05:01:27 »
Hi - I recently got a one shot colour camera to use with a mobile scope (I normally use a mono camera and filter wheel).

While I've been getting images that are not bad for a suburban location outside of London, I've been struggling with the colour balance - my images are looking too red.

I've spent the morning trying to work out why - turns out that for whatever reason, the red pixels in my flats have about half the exposure of the green and blue pixels. As a result, when the flat is applied, the red pixels are scaled to about twice the value of the blue and green pixels - this seems to persist through colour calibration.

I enclose two images of M31 - the first using the standard flats and about 5 hours of data - as you can see, the image is pretty good, but there is a red/brown tinge to the galaxy, hardly any blue at all.

For the second, I used Python to multiply red pixels by a constant amount in the flat so that the median value was about the same for red, green and blue pixels, then reprocessed one night's subs.
The processing is not as good (it was just done quickly), but as you can see, the colour balance is much better.

This made me wonder if the process for correcting flats for OSC is correct - should the various colours be scaled independently to avoid this effect?


Hi - I've been experimenting with the new photometric colour calibration tool.

I have an image of M31 which is very red - see

Is there a way to choose a different white reference point to get a better colour balance? For example, if I used the Elliptical Galaxy as the white reference point, I would expect this to have an inherently redder colour than the spiral galaxy reference point, so that if this was selected, I would get a bluer image since the colour is measured wrt the new reference "white point".

When I did this, I got a very similar image, and I very similar again when I used the standard colour calibration tool.


Hi - now that we have the excellent PCC tool, would it be possible to propagate the required information on image RA/DEC/Pixel Size and focal length from the subs to the integrated image?

That would make it easier to use the tool and remove sources of error.


Hi - I'm trying to use the Aperture Photometry Script to calibrate my system.

Firstly, congratulations - this is an impressive script!

I took images of a blue/red pair over several hours to look at second order extinction coefficients. Since the stars are quite bright, I'm limited to about 20s exposure time to keep these in a linear range. Unfortunately, this is in the part of the sky with very few stars and the script is unable to plate solve.

Is there any way to get the script to carry out the photometry anyway? I have several hundred images and can't do this by hand .. :).

Thank you,


Hi - I am trying to use the AperturePhotometry Script to calibrate my system using Landolt star fields.

I have found that in a couple of fields, the script things the stars are in slightly the wrong places, and hence not obtaining the correct flux.

I've included an example below.

The original fits file is at the location:

Any ideas?

Thank you,


Image Processing Challenges / Galaxy Trails in NGC 5907
« on: 2016 July 29 08:17:53 »
Hi - I've been trying to use Rogelio's techniques in "Lessons from the Masters" to try and bring out the dwarf galaxy trails near NGC 5907 - see APOD

This is the best I've been able to do so far - there are hints of the trails in all the right places, but that's about it! The data is 35 hours of Luminance from an RC8 shot in absolutely dark sky (22 mag/arcsec^2), so its unlikely I'll be able to improve on the data much. It may be that these are simply too dim to image well with amateur equipment  :-[.

Any thoughts to improve?

If anyone would like to try, the raw data is in my drop box location:


Bug Reports / FluxEstimator Bug
« on: 2016 April 03 11:05:05 »
Not sure if this is just me - I'm trying to use the FluxEstimator tool.

On start up, the values for the wavelength and width are set to zero - and when you try and enter a number, it just tells you that there is an error and the  current number is zero.

Using PI on Windows 10.

Any suggestions?

Thank you,


Bug Reports / Problems with DynamicPSF with synthetic star profile
« on: 2015 October 27 03:29:31 »
Hi - not sure if this is a bug, or something unusual I'm trying to do.

I've been trying to understand an unusual aspect ratio pattern in my telescope (an 8" RC). I think its because the primary is misaligned, causing coma across the field.

In an attempt to test this, I've created some synthetic star profiles using ray tracing. When I import these into PixInsight and try to use the DynamicPSF tool to analyse the aspect ratios, PixInsight stops responding.

I've included a small portion of a fits file with just 1 synthetic star, and this exhibits the problem.

Any idea what I'm doing wrong? I have the same problem in CCDInspector - it does not recognise the stars.

Thank you,


Hi - I've started messing with scripts (I thought I'd see if I could code up something to produce HR diagrams).

I've never coded in PJSR before, but the language looks relatively simple. Understanding the PJSR objects in PI looks a lot harder!

I managed to produce a small toy script with interface by using some simple scripts (AberattionInspector) as a basis, and I can see more or less how they've worked, but looking at other scripts its clear that there are a significant number of objects with lots of methods and data. Is there any way to find out what is there, and what they do, other than go through other's scripts?

I've been through this entire section, and I note that Juan first mentioned that PJSR documentation was on the to-do list in 2009, so I'm guessing it won't be available soon  :)

I've also seen that reference to a Parameters object - looks like a global object created by PI - is this right? Any documentation for it, and any other interesting global objects?


Tutorials and Processing Examples / Herzsprung Russell Diagrams
« on: 2015 January 06 07:40:57 »
Hi - not sure where to post this, but I had some fun this morning creating a HR diagram for M67.

Had some fun constructing the Herzsprung-Russell diagram. Also loaded the HR diagram from Johnson and Sandage (1955) - Ap.J. 121, pp 616-627, for comparison.

To do this, I used the FWHMEccentricity tool on each of the master R, G and B images and exported CSV files for each. These contain the parameters for a Moffat fit to each star: x and y location, X size, Y size, rotation angle, amplitude and background.

I found the definition of a Moffat function in the PixInsight documentation (the Dynamic PSF tool), then used Wolfram Alpha to integrate the expression over X and Y to get the total flux from the star For Beta = 4, the standard fit, turns out its A * (Sigma X * Sigma Y) * Pi/3.

I then created an index for each star based on its location: i+NX * j to allow me to match up stars from each filter.

Finally, I used ImageSolver to get the exact co-ordinates of the frames and AnnotateImage to label the Tycho stars in the image. I found one that wasn't saturated and looked up its magnitudes in Vizier. These were used to calibrate the magnitudes.

From there, I just used Excel to plot the data.

Not sure its science (the original paper is 60 years old), but interesting to see that we can now go 3 magnitudes deeper with an 8" RC and a CCD than Sandage could go with an 82" scope and a photo-electric detector.


General / Choice of Default Profiles for Colour Management
« on: 2014 December 08 10:26:21 »
Hi - I recently got a wide gamut monitor, and decided to try processing with the ProPhoto RGB profile, rather than the sRGB profile I'd been using before.

I have found that the ProPhoto images:
  • Require less saturation enhancement
  • Have a somewhat green blue colour, rather than straight blue

See for example, the two images of M15. The first has been processed using ProPhoto RGB as the profile, the second was assigned sRGB at the beginning. They've been processed almost identically - the colour saturation has been increased more for the sRGB (I applied a gentle saturation curve 5 times for sRGB, 3 times for the ProPhoto version).

Is this expected? I have found it gives quite interesting colours in galaxies, but less pleasing colours for clusters!


General / Best monitor for PixInsight processing?
« on: 2014 July 07 12:23:36 »
Hi - I am looking to upgrade my monitor. Any suggestions for a good monitor to use for astro photo processing?

I'm based in the UK and am desk space limits me to about a 24" screen.

Any thoughts?


Bug Reports / Problem with printing
« on: 2013 December 23 03:39:51 »
Hi - I've managed to print successfully to an A4 printer (a Canon MG18500) with a Win 7 64 bit, and the latest version of PI.

However, I'm having problems when printing out to a 4x6" image.

I select the size in the Canon printer preferences dialogue, this gets copied correctly to the PI print dialogue, but when I select the icon for "Expand Image Horizontally", PI hangs - it goes into a loop using about 13% of my CPU. The only way out is to kill the process and start again.

I've now realised that this only happens when I also select the landscape orientation in PI.

Not sure why it works with the A4, but not the 4x6.


Pages: [1] 2