DSS vs PixInsight stacking

[ssavignac]

Hello,

I'm trying to use more and more functionality of PI.

Now I'm tring to stack with PI and With the same set of images I'm not able to get the same result as with DSS.

First PI get less then 10 min to process the whole thing and DSS took 1:30 hrs. I Like PI for that.

BUt when using Noise evaluation script in PI for both result.

I always get better result with DSS.

I don't know what to try to get better result.

Samy

[pfile]

what type of camera? if OSC, did you use superpixel debayering in DSS and some other form in PI?

what stacking algorithm are you using in DSS? in PI? what registration method in PI (auto?)

need more information to answer this kind of a question.

rob

[ssavignac]

I have a t3i filter, skywatcher 8 inch, barlow 2x
I took, 8 light 300 sec iso 800 , 8 dark, 100 bias

In DSS
 register setting 2%, reduce noise with median filter.
Stacking light : Median sigma kappa clipping Kappa =2, Number of iteration =5
Stacking Dark : hot pixel and dark optimisation
Stacking biais: median  Kappa =2, Number of iteration =5
 Result channel    0= 18   :   1=30  :  2=23

In PI 1.8 rc7 ,Bactch processing script 1.33
Debayer RGGB,VNG
Registering : bcubic spline: clamping 0.14, max star 500, noise reduction 3
Integration: average winzorised sigma clipping sigma low 4-sigma high 3
Result channel    0= 13   :   1=44  :  2=29

Samy

[pfile]

what debayer method in DSS? same?

most likely it is down to the interpolation method. depending on what DSS uses that can have a big effect on the noise in the final image. PI's interpolation algorithms are pretty good.

rob

[ssavignac]

HI,

I don't see this option in DSS.

What is the standard parameter you use in PI for a Canon DSLR Stacking.

Samy

[pfile]

in DSS the debayer settings are under Raw/FITS DDP Settings. in DSS i don't think you have a choice of interpolation algorithms, and i don't know offhand what interpolation algorithm DSS uses.

the debayer method you choose i guess depends on what you are trying to do. if you shot Ha with a DSLR, then probably SuperPixel makes sense. if your tracking/guiding was not good then also SuperPixel can make sense as the smaller final image hides the imperfections a little bit. i guess you need to experiment with the other algorithms.

if you are talking about what interpolation algorithm to use, well, again you have to experiment and see what works best. i'm guessing that DSS is using a less sophisticated interpolation method which has the effect of smoothing the subexposures, so you trade decreased spatial resolution for a decrease in noise. PI's default interpolation method of lanczos-3 preserves details really well, but on my images i find that despite lowering the clamping factor i still get ringing. so i usually use bicubic spline.

rob

[ssavignac]

Thanks Rob,
I'll try that

Samy