Pixinsight - BPP vs Manual

[Aidi]

Knowing there is quite a good following in here with users using PI.. Whats the thoughts on using BPP for calibration, registration, stacking etc vs doing it manually?

Are the results from BPP quite a way off from doing it manually? Is one really better than the other, or just personal preference?

I must admit, i just shove everything into BPP after blinking the subs to get rid of any excessive trails etc. Generally i'm quite pleased with the results. I know the process for doing it manually and have tinkered with it, but not really tried doing a full set of subs from start to finish.

[XCalRocketMan]

I have used the BPP method on a number of occasions, but I prefer the manual route as I have more control over how things work. I have a list of process icons that I have previously set up and simply execute each in order following my generic process. I have a Workspace for Calibration and Integration, Postprocessing OSC, Post Processing LRGB/NB and a spare that I place testing processes. Most of these pre-filled out processes come from various internet sources, including Warren Keller's "Inside PixInsight" (highly recommended).

[KathyNS]

I try to keep my processing simple. For the vast majority of my images, I use BPP. I'll fiddle around with manual steps if there is a compelling reason to do so, but mostly I am happy with the results I get from BPP.

I'd rather spend my time working on my images than learning advanced PI processes. I doubt if I use even 10% of what PI can do. But it gives me nice results with just the basics.

[UlteriorModem]

I use BPP almost exclusively. With master calibration files. Saves a lot of time and effort. Once configured I think it can do just a good a job as I can manually. Once configured that is

One key thing is to cull your subframes either visually with Blink and or with the Subframe Selector Script. Garbage in garbage out

I just wish there was a way to save all the settings in the script. I tried making a process icon but all it did was open the script with the default settings.

[Aidi]

Glad it’s not just me then!

Not sure why saving the process is not working for you Tom, pretty sure it works on my pc... I’ll test and let you know

[jmfloater]

After checking all lights using blink I use BPP for calibration of light, dark and flat frames only . Then I run Cosmetic Correction, Alignment, Normalization and Integration manually. Doesn't really matter as long as you are happy with the results.

[Stuart]

I used to be manual for everything, but lately I've been doing more BPP up until calibration. I haven't seen much difference and it's way faster. I put everything in and then I get a sandwich.
However, I stop at calibration. After that I'll go "manual" to cosmetic correction, then Subframe Selector, then star alignment and then integration.

So I guess I do a hybrid approach.

[STEVE333]

I must admit, i just shove everything into BPP after blinking the subs to get rid of any excessive trails etc.

Just FYI - I never remove a frame of data just because of satellite/airplane trails. I live in the flight path for San Jose International Airport, so, almost every set of data has at least one or more trails. If you have a reasonable number of Light frames to stack (probably 15 or more) the Imageintegration step will completely remove the trails but keep the rest of the data. This will improve your final image.

That's been my experience for the last 3 1/2 years, and, I've never seen any trails in the stacked data.

Steve

[bobharmony]

My vast experience (having integrated two objects ) is that I am using the manual process to understand what is going on. My first process took a couple of days, my second one got done in about two hours. I will try out the BPP at some point, but for now will continue with the manual method. As Kathy said, I am scratching the surface of PI, and liking what I am seeing so far.

Bob

[Ben Cartwright SASS]

I downloaded the trial version of PI and am going to try it, the question is BPP

All the images I am getting from the Slooh scopes are already calibrated. What I seem to see is that if you try to use BPP it wants to force you to use master flats and darks etc, I don't have them since they are already done when I get the fits files.

[jmfloater]

Ben, if they are already calibrated then the next step I would do is cosmetic correction though this also may be unnecessary. Do a star alignment on the images and then integrate them. Unless I missed something you should be ready to combine them at that point.

[Stuart]

HI Ben, Agree with John. If your images are calibrated no need to use BPP. If you don't want to do Cosmetic Correction or SFS, then just go straight to star alignment and image integration

[ram]

I used to use BPP all the time to produce an integrated image from the raw captured frames but this year I've started doing things manually after running into some issues with Sharpcap's dark frame capturing (had to manually create a master dark but I was able to figure this out with help and by redoing each of the steps in BPP manually). These days I find it fun (?) to check the first frame and compare statistics to the master dark frame, and then make sure everything is happening correctly at every step of the way (i.e., examine the statistics, see below). It turns out there's nothing I'm doing manually that goes beyond what BPP does but I feel I messed up on the previous issue (trusting Sharpcap's master darks which for some reason aren't reliable I find - it's happened twice now) by just pushing buttons without thinking that I now see the value in doing the steps and checking the result of each step.

I also think if you are going to use SubframeSelector, then I don't see how BPP helps. SubframeSelector is done after calibration (and debayering if you are using an OSC) and if you're doing this step manually, you can do the remaining two steps manually (especially depends on how you use the results of SS, see below).

In the case of M31, I did use it to throw out some frames but for every other target I integrate them all anyway so all this does is give me an idea of what the data looks like. This is useful in tempering expectations - for instance, if I have really good statistics then I should be able to produce a certain kind of image compared to having bad statistics.

The way I use PI is very similar to the way I use many GUI tools, i.e., I use the Unix file system to organise things on the Mac at the prompt and use a lot of command line tools - so different directories/folders for different things and BPP sets up things its own way but by doing the steps manually I can organise things my way and especially with SS I can have directories containing the top half exposures (according to my criteria) or the top 75% or what have you.

--Ram

[Stuart]

Hi Ram,
You can use BPP to calibrate the images by clicking "calibrate only" and then go manually from there.