›

PI vs. PS?

John Batts
in PixInsight
Hi Adam,

In response to your request for comments, there may not be another soul on the planet as well versed as you in both PixInsight and Photoshop. $64 question: When you process a new image for publication, do you now use PixInsight, Photoshop or a combination? Are there some types of images, say faint nebulae, where you prefer PI? It's an important question because the PixInsight learning curve is steep, but if Adam Block now uses it primarily, that's hugely attractive.

So, a suggestion for a new tutorial (it might be first on your list, and maybe even a freebie to attract new subscribers) might be a comparison between a PI- and PS-processed image. Perhaps use PI to calibrate the data, process the same calibrated data with both PI and PS, and compare the results. It would require a lot of effort, but if you (hopefully) find the PI image is superior, that would be a great reason to pursue PI.

As always, I find your tutorials straightforward and clear. Keep up the good work.
- John

P.S. I second the request for a PixInsight Horizons tutorial on narrowband, especially one incorporating H-a data.

Comments

  • BlockHeadBlockHead
    Hi John,

    For all images I have produced in the past 2? (maybe 1+ years), they have been done to the 97% level in Pixinsight. There are a few things that are special applications I have done outside of PI- but the answer to your question is yes I am primarily producing images in PI.

    A couple of things to note. First, my "thing" has been to move with the times and work with popular programs. I started with a precursor to Maxim DL  more than 20 years ago and have progressed through additional software such as Mira (anyone an remember this?? :) ), CCDStack, Photoshop, GIMP, Registax, Registar, (even a little IRAF)...etc etc. I am allowed the luxury to say my move to PixInsight is my modus operandi as a user that wants to understand all of the available tools as well as being a business decision. So it is not a comparative statement and I am not claiming anything is best. This is a trap I will not fall into! :)

    That being said- PixInsight is the only program I have used that in a single "Universe" of tools and techniques permits the rendering of an image from beginning to end. CCDStack, for example, wasn't designed to do this and so needs a complimentary program to make a "pretty" picture. And you cannot use Photoshop to properly calibrate astronomical images. So comparing PI to "Photoshop" doesn't make sense...and comparing to the combination of two programs... still keeps things apples and oranges.

    There is something else that might make my experience in PI go more smoothly with respect to Photoshop. In Photoshop I really did not do any terribly sophisticated and dynamic layers (like adjustment layers). So in PI I accomplish the same goals with a combination of masks and blends (all PS blending is fine to do in PI).

    But on the personal side the following tools are just so incredibly good I do not think I need to use Photoshop to do the vast majority of my work:  DBE, HDRMT, MLT/MMT, PCC and Pixel Math. That last one has opened up entirely new ways for me to interact with the data- and it has been extremely powerful. A second tier of tools include TGVDenoise, ArcSinhStretch, Mask making methods, and even MorphologicalTransformation. 

    HDRMT has probably influenced my style the most... I have always been a good user of DDP- and HDRMT is exactly the thing I always wanted. 

    Finally... another note- Unlike other programs I have directly created innovations in PI. I have created useful Pixel Math expressions...and new techniques (that to my knowledge are not "known."). Long ago I suggested the minimum filter in Photoshop should be be a circular kernel.... and they did take my suggestion! But in Pixinsight I feel I have made real concrete contributions that I hope will be used by others. Of course this is for you to decide after reviewing all of my material. :)

    Anyway, a verbose answer... But one last thing- you mention the learning curve is steep for PI. I think that is true when there does not exist enough explanatory content on how things work. This is precisely the thing I am trying to change- and this more than anything else I hope persuades you to get into PI.

    -the Blockhead

  • John Batts
    Thanks for your response, Adam, and keep up the good work!
Sign In or Register to comment.