This Tutorial is based on http://www.astrosurf.com/buil/iris/tutorial19/doc40_us.htm with a very welcome helping hand from Stephane Poirier

Important: switch off the automatic Wallpaper changer (should you have one installed).

I imaged Comet 17/P Holmes with K3CCDTools and have an AVI as input.
In K3CCDTools I found the time of capture of the first and last images of the AVI:

First image 22:11:55
Last image: 23:09:59
Elapsed time: 58 minutes and 4 seconds = 1.034482759 hour

In K3CCDTools I applied the Dark Frame, aligned on the stars, sorted the frames on their original sequence number and exported the Frame Collection as 179 BMP's: filename k, starting with 1, suffix 1 digit.
Note: D:\Temp-iris is my default IRIS work folder.

I loaded these BMP's into IRIS with "File|BMP conversion" with output kk and I selected b/w.

Now I have 179 fits: kk1.fit – kk179.fit

Now it is time to use IRIS commands via this button, which gives a popup window:

Type the following command:
> setspline 1 and hit the Enter key. Nothing visible happens, but that is OK.

Next draw a small rectangle around one the core of the comet and do:
> register kk kkk 179

Registration is done and now we want to find out the displacement of the comet.

Load the first (registered)image:
> load kkk1

Draw a small rectangle around the core of the comet and in the Output window you see coordinates popping up: we want the x=333 y=206 values.

Now load the last image:
> load kkk179

Again you draw a small rectangle around the core of the comet and in the Output window you see coordinates popping up: we want the x=341 y=189 values.

So the displacment in x is 8 pixels and in y it is +10 pixels.
This all during a period of (see above) 1.034482759 hour

Now we manually calculate the displacement in pixels per hour, both for the x and the y shift and we find dx and dy.
DX= 7.7
DY= 9.7

Now we execute the following command:
Syntax : [trans2 input output dx dy #images] :
>trans2 kk kkk 7.7 9.7 179

Now stack the registered set of images. The simplest method:
>ADD_NORM KKK 179

And I get this:

To be on the safe side I save this temporary result as t1
>save t1

My result was heavily overexposed, so I used the Threshold slider:

This looks better :o)

Saved as tt1
>save tt1

Draw a rectangle around the core and note the coordinates (similar to above)
I found: x=330 y=208

Now it is time for the Rotational Gradient:

Syntax: RGRADIENT [XC] [YC] [DR] [DALPHA]

Christian ‘IRIS’ Buil says:
"Typical values of [dalpha] go from 2° to 20°. For [dr], the values are from 0 to 2 (negative values are also significant). Finally, note that it is necessary to know the position of the center of the object with precision. It is good to experiment with several values of the parameters to have a precise idea of the structures that are revealed."

I used:

>rgradient 330 208 0 15

Result:

A very vague comet, so I played with the Threshold sliders:

I saved this as:
>savebmp final

Here is the final result, not very nice but the procedure works …