Colour worms v3

[mikemeister_admin]

Sorry Mike, but I do not seem to be able to turn off the worms showing on the curves.
Is it possible?
Chris

[derekfountain]

Sorry Mike, but I do not seem to be able to turn off the worms showing on the curves.
Is it possible?
Chris

I don't think it's possible and once the usefulness of the worms drops into place you won't want to. I used the beta for several weeks before it suddenly dawned on me one evening just how useful they are.

Tip: go to Lab mode, then put the mouse pointer on an area of the image that needs a general contrast boost. See the worm on the L curve showing you the area of the curve that needs steepening? OK, now right click and select Contrast Pin. You get two pins on the L curve, one at each end of the worm. Now put the mouse pointer in the lower margin of the L curve (i.e. just below the curve area itself) and click-drag right to rotate the curve around those two pins and increase the contrast in the area they represent.

It's a hugely useful technique, both for contrast and saturation adjustments.

[-default]

The only way to eliminate the worms would be to go to a point sample size, but that of course has other drawbacks. 

I agree with Derek that the worms are an improvement over the sample display, but I would consider  making it a user selectable option, if it's important to you.

[mikemeister_admin]

Thanks Derek, I'll have a play and see - I understand the contrast pin, but do not see what it has to do with worms. 

The worms do not seem to be dynamic -  the worms seem to show the value of other channels using the original image, not when I've made an adjustment.  So if I completely remove a channel (say green) it still shows as the same size worm as when I started.  I have yet to deduce what the length of the worms signifies and why (in RGB space) we need worms in the individual RGB channels, when it is showing in the Master one.  I obviously need to play and then understand.

Thank you for your reply - I'll go and digest the manual rather than be a monkey seeing what happens when I push buttons!
Chris

[derekfountain]

Thanks Derek, I'll have a play and see - I understand the contrast pin, but do not see what it has to do with worms. 

The worms do not seem to be dynamic -  the worms seem to show the value of other channels using the original image, not when I've made an adjustment.  So if I completely remove a channel (say green) it still shows as the same size worm as when I started.  I have yet to deduce what the length of the worms signifies and why (in RGB space) we need worms in the individual RGB channels, when it is showing in the Master one.  I obviously need to play and then understand.

Thank you for your reply - I'll go and digest the manual rather than be a monkey seeing what happens when I push buttons!
Chris

The contrast pinning puts an curve point at each end of the worm. I've just been tinkering with an image taken from a river bank which contains a lot of rippled water. I'd like to improve the contrast in that water surface. Moving the mouse pointer over the water, I watch the worm on the L curve and find a sweet spot where I have a long worm. That's the point to contrast pin because the worm shows me that I'll be steepening (and hence improving contrast in) the part of the curve that represents the water. Without the worm the only other way to do it would be the move the mouse over the water and try to remember the two extreme points where the curve's "bouncing ball" travels to.

I'm vague on the technicalities of how the worm is created. It's the sort of thing that you don't really need to understand to make use of. I do remember a long discussion in the test phase that discussed the merit of having worm(s) on the RGB master curve. I switch off the master curves (another CM3 feature!) so I don't see them, but maybe that's not the place for a worm-newbie to start...

[-default]

Derek is spot on with his description of how to use the worm - in particular I like the idea of a "sweet spot" where the worm gets longer because your sample contains a wider range of values.  I would add that the 11x11 sample size works well for this (this is set separately for the "roving" hue clock in the curves window than for the other samples).

The worms on the RGB curve are visually interesting because of the dynamics of seeing the various colors move around, and how they relate to one another.  It should also be also a cautionary note on the evils of using the master curve as anything other than a visual aid.  With worms, it is now visually obvious that changing one area of the RGB curve will usually bend each of the three worms differently, causing color shifts.  This is the motivation, discussed often in Dan Margulis's books, for adjusting the channel curves separately to add contrast without shifting the hue of the object in question.

Using a contrast pin allows you to do exactly that.  go to each of the three curves individually and adjust the contrast with a rotate move.

Another thing that is interesting about the worms on the composite curve is that they are shorter in wgCMYK than in RGB.  Being shorter, the worms for different objects in the image have less chance of overlap.  Less overlap means more  opportunity for selective adjustment, bigger curve moves, and finer control of individual colors than in RGB.

The worms in Lab are interesting as well, but that's probably enough for now - LOL.

BTW - Chris is right that the worms are based only on input values.  They represent the min and max channel values that the sample is picking up.  Their shape changes as you move the curve around, but not their horizontal range.