Difference image in postage stamps shows sources that have disappeared the same as new souces

[JohnAstronomer]

The difference images show sources that no longer exist the same as new sources. An example is https://lasair.roe.ac.uk/object/ZTF20acpdrkt/ this is a HPM star and as can be seen from the reference image the original source no longer exists - in fact it has moved to the position where it triggered the alert. However the difference image shows both objects which is not correct. The source that no longer exists should should be shown differently - I would suggest as black as the difference images in the ALeRCE explorer. The link below should slow the images for this transient from ALeRCE which is how I believe Lasair should show this.

It looks as though the maths in the image subtraction may be wrong as a missing object should be negative though of course that cannot be displayed.

[RoyWilliams]

Since the summer, we have been building a new Lasair, now in beta. Your source is there at
https://lasair-iris.roe.ac.uk/object/ZTF20acpdrkt/
We have a better FITS viewer. Click on the links where it says "This image is the most recent. Click on these links to inspect the science, ref, and diff images in detail."
Specifically https://lasair-iris.roe.ac.uk/fitsview/ZTF20acpdrkt_cutoutDifference/

[JohnAstronomer]

Thanks the images look good to me. I think I preferred the old page layout the light curve area is too short for some of the complicated curves such as CVs The ref, science & difference images really need to be shown at the same time - either on the front page as they were before or in a link. looking at the three is the best way to work out what is going on or objects such as this HPM star. Incidentally one of the other classifiers has it as an asteroid ! It appears to have no idea of it's absolute magnitude even though there is a good parallax in Gaia DR3 If I zoom into a part of the light curve with my mouse I don't seem to be able to zoom out again without reloading the page. I notice the comments have disappeared - is this deliberate ? They were a good way of passing on information about mis-classifications such as this case. John Murrell FRAS

…

On Thu, 2021-01-14 at 01:10 -0800, Roy Williams wrote: Since the summer, we have been building a new Lasair, now in beta. Your source is there at https://lasair-iris.roe.ac.uk/object/ZTF20acpdrkt/ We have a better FITS viewer. Click on the links where it says "This image is the most recent. Click on these links to inspect the science, ref, and diff images in detail." Specifically https://lasair-iris.roe.ac.uk/fitsview/ZTF20acpdrkt_cutoutDifference/ — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. [ { ***@***.***": "http://schema.org", ***@***.***": "EmailMessage", "potentialAction": { ***@***.***": "ViewAction", "target": " #204 (comment)", "url": " #204 (comment)", "name": "View Issue" }, "description": "View this Issue on GitHub", "publisher": { ***@***.***": "Organization", "name": "GitHub", "url": "https://github.com" } } ]

[RoyWilliams]

The ref, science & difference images really need to be shown at the same time
-- Not so easy, I struggled with the javascript to make that happen but failed

If I zoom into a part of the light curve with my mouse I don't seem to be able to zoom out again without reloading the page.
-- double click on it

I notice the comments have disappeared - is this deliberate ?
-- the old Lasair has been running 2.5 years. There have been 8 comments from you, 3 comments from Ken Smith, and 6 people who have made just one comment. Not very popular! We are thinking about how to improve the functionality.