First Attempt with Pixinsight

[MChandaue]

Hi this is my first attempt at processing with pixinsight. Really poor data only about 30 mins of lights, same for darks, no flats or bias. Taken in heavily light polluted skies. I couldn’t do anything with this in photoshop. This is a very basic workflow, dynamic background extractor to remove most of the light pollution, colour calibration, denoise and then stretch. Finished off the levels in lightroom as I still haven’t got used to the histogram and curve tools in PI. 

It’s kind of good news/bad news. Well pleased with what I got from data I’d given up on. The bad news is I’m definitely going to have to buy PI once the trial runs out because it’s a joy to use despite the learning curve. Once I know what I’m doing with it and have clear dark skies to get some good data the sky is the limit. 

[Marky1973]

Nice image! Have you got "Inside Pixinsight"? Very good book on PI if not!

[pc387]

That looks really nice (certainly not poor as an end/interim result), I'm sure there's some more colour that could be teased out?

I was in the same boat as yourself, and there are some excellent tutorials by Harry Page, Light Vortex, You Tube etc that got be 10% proficient in a very powerful piece of software; but it covers 90% of my needs and worth every penny.

 

pc387

[MChandaue]

Not got any books yet as this is my first day with the trial so just watched some tutorials on youtube but will definitely invest in the book.

[MChandaue]

It’s not cheap but it is definitely worth the money. With very little effort you can really get amazing results. I found it far easier to use than photoshop and I can only imagine what’s possible once I learn how to use more of the tools. Given the time I think I could do a lot with this data. I will probably reprocess this many times as I learn what I’m doing. Can’t wait to get some clear skies so that I can get some good data with lights, flats, darks and bias. 

[Stargazer33]

Nice first try. Some lovely dust lanes in those spiral arms.

I think you've clipped the blacks though. Your background is too black and the galaxy extends way out past where yours stops. If you look at your histogram you will probably see the black end (left hand side) doesn't come all the way down to the base line, but goes off the graph some way up the left hand edge. This means you've clipped your blacks and lost some of the finer detail (hard won data).

HTH

[ollypenrice]

The Pixinsight/Photoshop debate is an interesting one. Honestly I think it comes down to the way a particular imager's mind works. For me my mind works like Photoshop and I find that intuitive. Choose the one in which you feel comfortable. Being promiscuous, in imaging at least, I use both.

To play the devil just a tiny bit (can't resist!) PI is quietly introducing Ps functions previously deemed to be against its rather fundamentalist principles. No no, I didn't say that!

Your image is going in the right direction but it is massively black clipped, something on which both PI and Ps would certainly agree.

Olly

[saac]

 

Hi MCHandauM, I bit the bullet and purchased a licence for PI last month and have not regretted it. It is such a well put together piece of software and is very well supported with tutorials (videos books etc). It is expensive but in the long run I think it makes sense. I'm slowly getting up to speed , trying to process a capture of M13 at the moment, so lots to learn. It's good to have company on the learning curve.  Good image you have there and good luck with PI.  

Jim

[MChandaue]

Yes this is black clipped that was in Literoom. If I spent a bit more time with DBE I could have avoided that. My first attempt I spent a lot of time in DBE and it really nailed the light pollution but a bit further on I clicked the wrong thing and completely lost the image so had to start from scratch.

This time I only spent a few mins in DBE and as a result it did a decent job but still kept the sky a bit too bright. I bought down the blacks in lightroom to compensate but overcooked it a bit but on the whole was reasonably happy with the result. This was about 20 mins from opening the stacked image to the saved file here.

When I have time and a better idea of what I’m doing I will reprocess this image to see my progress. 

As far as the PS vs PI debate I think if you already know PS then it’s going to be the better tool. If you know neither PI will get you a decent image quicker than PS because of things like the auto stretch and the very good gradient removal and noise reduction tools. 

Mark

[Lead_weight]

You might try Astro Pixel Processor. It's also got a free trial, and is a lot more straight forward for processing. The steps to process are linear, order in the interface 1-9, and you just follow the steps to process your images. I've been finding I can do all my calibration and integration way easier in this program, and the results are really good. I also find it easier for removing light pollution and calibrating the color for the image. But I still do a few steps in PixInsight occasionally as it's got a more robust and varied tool set. But also, APP is a fraction of the cost.

[MChandaue]

I may have a look at APP although I found PI very simple to use although that’s thanks to the very good tutorials. Of course to get the most out of it the learning curve is very steep. StarTooks also looked pretty good and again a fraction of the price of PI. 

[StuartJPP]

I don't want to get into a debate regarding PI and APP but I extended my APP trial, thanks to Mabula, to try it multiple times but I just couldn't get on with it. PI consistently produced a better stack for me using the same data set which to me is the important bit, it is the foundation to work on.

These were just my findings and it may well be that I am doing something wrong in APP but eventually the trial ran out and hence I gave up.

So I am still a PI and (old) PS man, keep at it and you will get there.

Using PI on its own is possible but quite painful if you are not using it very often or if you are not a math's professor (using PixelMath).

 

[wimvb]

That's a very nice image. Others already commented on the processing, but I'm more interested in the gear used. You've got star elongation in the corners. Is that a lens effect or a spacing issue?

Taking flats is important, especially if you have a short fl system in light pollution. Without them, you'l have multiple gradients, that will be difficult to remove even with DBE.

[wimvb]

On 13/09/2018 at 20:52, ollypenrice said:

To play the devil just a tiny bit (can't resist!) PI is quietly introducing Ps functions previously deemed to be against its rather fundamentalist principles. No no, I didn't say that!

When PI introduces brushes and erasers, we know who won.

Or when PS introduces à trous wavelets, for that matter.

(there, I said it)

[ollypenrice]

10 minutes ago, wimvb said:

When PI introduces brushes and erasers, we know who won.

 

Yes indeed, and the winner is... common sense!

This PI doctrine for only touching the data using tools derived from the data reminds me of those various sects of monk who weren't allowed to handle money (so they wore gloves) or weren't allowed to draw blood (so they went into battle with clubs.)

lly

[iwols]

just to add i cant get on with either ,but thats just thicko me ,find that when pi does a good job it does a great job,but app makes it a whole lot easier and more consistant,bought both licenses but stuggling with both really

[Marky1973]

3 minutes ago, iwols said:

just to add i cant get on with either ,but thats just thicko me ,find that when pi does a good job it does a great job,but app makes it a whole lot easier and more consistant,bought both licenses but stuggling with both really

Yep, that's the problem - one size just doesn't fit all - you really do have to experiment to find the best solution for you. I always struggled with more in-depth stuff on PS, but found some of the stuff I did know, helped tweak efforts in PI...although I am currently limited to GIMP until  I decide whether or not to buy PI outright!

[MChandaue]

I think the star elongation in the corners is because I’m not using a field flattener. That’s my assumption but I’m still very new to this. I didn’t use flats for this and DBE did a great job on the gradient but there is a dust bunny that is hidden by the black clipping. I will give APP a try but my current plan is to take it so far in PI then finish off in PS. However that plan may change once I’ve tried the APP trial.