Siril 'Tilt' command

[Zummerzet_Leveller]

Wondering if anyone has used the 'Tilt' command in Siril yet and what results you may have had?

Just tried it out on an old image I know was a fair bit off and it seems to work well.

Here's a more recent image where I checked things over a little better.

[ONIKKINEN]

I use Siril all the time and didn't know about this, is this a recent addition in 1.0? Looks like it works pretty well on my shots too.

[Zummerzet_Leveller]

9 hours ago, ONIKKINEN said:

I use Siril all the time and didn't know about this, is this a recent addition in 1.0? Looks like it works pretty well on my shots too.

Yes it's new functionality.  They've also brought in Pixel Math and more Photometry features.

https://siril.org/download/2021-11-20-siril-1.0.0-rc1/

[vlaiv]

I think that you need nice uniform star field for that to work properly.

[Zummerzet_Leveller]

15 minutes ago, vlaiv said:

nice uniform star field

Here's some results from individual subs on the Iris Nebula (an ok starfield I think) through the Redcat with the 2600MC.  Quite a bit of variability in these, so either my image train had movement or the tilt calculation is a bit flakey.

[vlaiv]

3 minutes ago, Zummerzet_Leveller said:

Quite a bit of variability in these, so either my image train had movement or the tilt calculation is a bit flakey.

It can be down to SNR and FWHM calculations.

I believe that it calculates tilt by simply looking at FWHM of stars (this is why it needs uniform star field so that FWHMs are on average equal) and then finding average in each "corner" (or rather zone of sub).

Many faint stars introduce quite a lot of noise in FWHM measurement.

Do you have any set of subs where target is not in direction of Milky way? Fewer stars and clean background could help

[Zummerzet_Leveller]

6 minutes ago, vlaiv said:

Do you have any set of subs where target is not in direction of Milky way? Fewer stars and clean background could help

Not with the 2600 and Redcat.  Here's one with SW150P and Canon 100D from earlier in the year.  Interesting that only one corner is off when compared to the other three.  I had the standard focuser without a 2" ring clamp at the time.

If we ever get some clear skies again I'll try the 2600 rig out to the north.  Seems like it could be a useful tool to get indications of tilt, not sure it could be used in a live situation to resolve the tilt though.

[alacant]

Hi

Here we are with the rc3 from git. Quicker and more convenient than astap. Any part of the sky you like. Works best on linear. 

Here's an eos700d frame.

Cheers and HTH

 

Edited December 10, 2021 by alacant

[Zummerzet_Leveller]

Useful to know @alacant. I shall continue to play.

[lock042]

That should always be applied on linear image.

[ONIKKINEN]

I'll add my tests to here. They show what i was already expecting just based on looking at the images, but would have appreciated a tool like this when testing in the field. First 3 pictures are with the stock VX8 focuser, the last is with a Baader diamond steeltrack fitted and shows significantly less tilt. Now that i know the mechanical issues are sorted i can start dialing in tilt with a tilter plate. But also, the last sub is of poor quality so it might not make a good test subject so this can change. But the tool works well from what i can tell.