CMOS software binning

@umasscrew39 Hi Bruce. In tab 6 Integration all the way at the bottom you can set the scale factor to 0.5. Note that when using bin 1 lights, you need to use bin 1 calibration data as well.

Excellent................thank you!!

@wvreeven

Could you explain a little more the recommendation to bin Calibration (and where to do that) to correspond to a binning scale in Integration?

Is it recommended to change the Registration scale to match any change in the Integration scale?

Thanks

@jhart What I meant is that software binning by APP is applied very late in the integration process so you need to make sure that the lights and calibration input data all have the same binning. So if you have 2x2 binned lights then your calibration data MUST be 2x2 binned as well. If, for instance, you have 1x1 binned Lights and 2x2 binned calibration data then APP will not be able to perform the integration.

@wvreeven

Thanks.

Sorry to add comment to this quite old post, but I don’t want to do another thread on a similar question 😉

Still getting a little confused with the binning. On the technical side, my pictures are a little oversampled. I use a 585MC Pro on a 200/1000 Newton. I noticed that my latest pictures look a bit out of focus though I’m sure I checked that twice. So I blame it on the oversampling. Some say I should take images with 2x2, some say that binning in post would do the job. 
Any recommendations on that? I’m shooting mostly under a Bortle 5 sky, and the seeing is probably never “excellent”

CS, Sven

Hi Sven, @easylink

We will implement software 2x2 binning soon for users to directly software bin the data. But from my perspective, the best way is to simply bin while creating the integration, by setting the scale factor in 6) integrate to 0,5 and set the pixel interpolation filter to Catmull Rom-Spline instead of Lanczos-3 😉 Or if you want to Bayer Drizzle the data, simply leave the scale at 1.0 and set the drizzle droplets to 2.0. You can experiment with the scaling factor as well, depending on your data, maybe scale somewhere between 0,5-0,75 is already fine ? 

Some technical explanation, if you choose to software bin to start with your data, you are throwing away valuable data/information directly, the star centroids and thus alignment precision is directly affected and worse because of this, I see no reason to do so to be honest with any data. The only downside of not software binning is harddisk space I suppose. Adding software binning 2x will become an option soon in the Raw/Fits menu at least.

Mabula

Thanks again Mabula, for your help!

So I’d rather enable 2x2 in the camera? I did some darks and bias in 2x2 already to be prepared. But since now I haven’t tried to shoot with binning enabled. 
I regret a little bit that I sold the 6”. It would fit much better to the pixel size of my camera, but that’s also part of the learning progress I guess 😉

CS, Sven

Hi Sven, you can try to shoot with hardware bin 2x2 yes. But my point is that it is not really needed unless your stars are bigger than 10 pixels I would say 😉 If they are still reported as smaller than 10 pixel than shoot with bin 1x1. The star FWHM column in the frame list indicates the size of the average star in your images. So when shooting bin 1x1, simply reduce the final result with the scale factor 😉

Ok, maybe I’m looking for a solution in the wrong place. Maybe I was a bit out of focus though I focused before. If you don’t mind, I upload the processed mosaic. I can’t get rid off the gradient. Maybe the moon affected the shooting though I was using Duo-Narroband. 
The highest FWHM is 12, the lowest around 6. Most of them are below 10. 

This is the processed version, integrated at scale 0.5. To me it seems that the guiding was not perfect. Please be gentle, it is my first mosaic and APP project 😉

Hi Sven @easylink,

🙂 Did you try to remove the gradient with APP's Remove Light Pollution Tool? You should be able to get almost completely rid of it with the tool. I see most is gone already.

Yes, the moon... even with narrowband, there will always be some gradients actually. I think with those FWHM values, you really do not want to use bin 2x2. That is not needed. I see data with bin 1x1 with stars of 20 pixels, then you would say maybe use bigger pixels or start binning 😉

I will be very gentle, for a first mosaic and first APP project, that looks really awesome actually 🙂 !

Great work,

Mabula

Thanks Mabula! 😉 Yes, I tried the Light Pollution Tool. For now, I can’t get it any better. But I consider it as a test. I hope to get CS the next days when the moon disappears. Also, someone told me to think about getting an OAG. My Guidescope maybe is too small for the focal length of the telescope. Funny thing: first time ever I tried APP on the mosaic the 3 panels were aligned perfectly without any visible stitch. Now I always see them 😂