Hi Dave!
I must say it seems hard to use! Is there general use documentation anywhere?
This is normal. No, we don't have a complete documentation for the core application yet; we'll be writing it during this month, just before the first commercial release. But once you get a bit accustomed, you'll see that the new interface follows the same paradigm and general rules as LE.
For instance, the processing console (whatever that is!) keeps popping up and covering up my image -- and in violation of usual Windows style you can't drag it around by its title bar.
Well, indeed PixInsight is not a Windows application at all :-)
Jokes apart, yes, the Processing Console window can be floated and moved. Double-click on its title bar (but not precisely over the title text), or click the down arrow icon at the top left corner and select "Floating". Then you can click on the title bar and move it around. The same is true for the rest of explorer windows: Process Explorer, Format Explorer, History Explorer, and Script Editor.
To hide the Processing Console when it is docked on the workspace (what we call
auto-hide state) you can hit the Escape key or click anywhere outside the console. Once you get accustomed, hitting Esc is a cheap price for the rich functionality that the console offers.
The processing console plays an essential role in the PixInsight interface. On one hand, the console is the "face" of the entire platform, where all processes communicate important information to the user, as progress information and numerical data that are often necessary to understand and control how processes run and what they do. It also allows you to stop/abort ongoing processes by clicking the Pause/Abort button.
The console also includes a command line where you can enter commands. This is an extremely powerful and flexible feature in PixInsight. For example, if you want to close all open images, just enter:
close *close is an
iinternal command; this means that its functionality is provided by the core application. The same is true of open, obviously:
open c:\MyImages\M31\core-tests*.fitwould open all FITS files that match the specified pattern.
All installed processes have also command-line functionality (some of them more than others, though). For example, to rotate 90 degrees counter-clockwise all images whose identifiers start with "M45":
FastRotation -r90 M45*or, to unsharp mask the active view with StdDev=2.5 and Amount=0.8:
UnsharpMask -s=2.5 -a=0.8If you get tired of typing "UnsharpMask", you can create an alias:
alias usm UnsharpMaskand now you can enter:
usm -s=2.5 -a=0.8with the same results as above. Of course, you can enter more complex things like:
usm -s=2.5 -a=0.8 Image01->*to apply the process to all previews defined in the Image01 image, for example. There are *lots* of things that can be made much easier and faster with the command line; for example, it can execute script files that are basically sequences of commands. This has been just a small demo to give you an idea of its power.
Microsoft said a few years ago that command-line interfaces are dead. Hahahaha!
Then I have some images whose levels need rescaling (because of the FIT format saved from Iris)
Use the Rescale process. You have it on the Process Explorer window, TransferCurves category. Or select the image and enter "Rescale" in the command-line ;-)
so the first thing I do is bring up an empty PixelMath window in PixInsight LE, and apply it to the image, and the rescale just works. It doesn't do anything to my image in Pixinsight standard.
PixelMath has been completely rewritten from scratch. It is a completely new process that accepts expressions in algebraic notation. With the exception of JavaScript support, PixelMath is perhaps the most powerful and versatile tool in PixInsight now. To do what you want (along with using the Rescale process, which is the obvious option), you can type the following expression in the RGB/K line of PixelMath:
$targetand ensure that the "Rescale result" option is checked. This will work as you expect. $target means "the current pixel of the target image".
A few more PixelMath examples. To add two images, Image01 and Image01:
Image01 + Image02Easy, right? If you want to make the process a bit more general, you can use:
$target + Image02which would add Image02 to any image where you apply the Pixelmath instance.
How about an hyperbolic arcsin stretch (which has so many adepts these days; I frankly don't know why)?
ArcSinh( $target )does the job. Let's put a bit more complex examples. Suppose that you want to apply a midtones adjustment of 0.35 (just as in the HistogramTransform tool) only to those pixels that have values greater than 0.12:
iif( $target > 0.12, MTF( 0.35, $target ), $target )The iif function works as follows:
iif( condition, do_this_if_condition_is_true, do_this_if_condition_is_false )Do you want to see a more complex thing with PixelMath? How about a transformation to polar coordinates?
r = Sqrt( Width()/2*Width()/2 + Height()/2*Height()/2 )*X();
theta = 2*Pi()*Y();
x = Round( Width()/2 + r*Cos( theta ) );
y = Round( Height()/2 - r*Sin( theta ) );
Pixel( $target, x, y )To try out the above expression, type "r, theta, x, y" in the Variables field, and copy the expression in the Expression Editor (click Edit button). Then apply the process to any image
With the internet connection required, there is limited time for me to learn how to use this (because I can't use it on the move), a manual is really needed.
I understand that this is an annoyance. But please take it positively: you can work with a fully functional application that doesn't impose you limits at all (file sizes, watermarks, disabled functions, etc).
As always, thank you for your feedback and interest.