HSB and no noise. At all.

The Curvemeister 3 beta test. Please post and discuss bugs, screen shots, suggestions, and any other information about your testing. This board is only visible to Curvemeister 3 beta testers.
mikemeister_admin
Posts: 4927
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Wed Oct 31, 2007 1:50 am

This is amazing. I just opened a very colorful image in CM and applied a fair amount of saturation to it. Then I took the image into PS, duplicated it, and turned off the CM layer. Then I applied saturation from the Hue/Sat slider to match the CM adjusted image. The Noise from Hue/Sat was tremendous. The noise from CM non-existent.
I went back into CM, and as I raise the Saturation, the noise disappears, as I lower it, it becomes more obvious.
What kind of magic is this?
...true, I kind of blasted out some detail, but that's beside the cool point.

Gloria

-default
Posts: 1916
Joined: Thu Mar 26, 2015 1:53 am

Postby -default » Wed Oct 31, 2007 4:37 am

I can't say what happened for sure without seeing the image, but that's never stopped me before.

It's not unusual that when you want to bump saturation, you want to do so in the colors that are less intense, and leave the ones that are already there alone.  In Curvemeister you can do this by moving the middle of the saturation curve upward, leaving the white end of same curve in place.  This protects the already intenst colors from becoming blowing out, and at the same time adds more oomph to the medium colors. 

OTOH, in Photoshop's Hue/Sat you have basically one slider to move, sort of like using contrast instead of curves or even levels.  Moving it will increase the color intensity (and the noise) of the more colorful areas.

Of course, if I saw the actual image, it could be something else entirely. :-)

On a related note.  Dan Margulis has said a couple of times that one of the great uses of masking is to allow saturation to be increased selectively in bright areas, leaving dark areas alone. 

To tell the truth, I haven't tried this yet.  Here's what Dan said, a few days ago discussing CM's new masking:

"It adds a lot of power to intensify color. I've realized in the last few years that we look for intense colors in the quartertone range, more intense than the camera ever provides. We can use LAB or similar to help get there, but the problem is that it can make the darker areas too saturated. We'll see how far they carry the ball." - DM

I would add that they is ... us! :-)



mikemeister_admin
Posts: 4927
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Wed Oct 31, 2007 10:19 am

I'll post the images, and if you can't see here, I'll email you the files. Kind of remarkable, at least it was last night :)

ggroess
Posts: 5342
Joined: Wed May 24, 2006 2:15 am
Contact:

Postby ggroess » Wed Oct 31, 2007 12:30 pm

Gloria,
This example might be a good Wiki article.  Describe how you did it and what the differences are...
If you don't want to write it on the wiki site...write it in word with lot's of examples and I'll see what I can do to get it posted.

Greg

mikemeister_admin
Posts: 4927
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Wed Oct 31, 2007 11:29 pm

I just got back to this, and haven't quite figured it out yet. I have to go find some good noisy images on another hard drive. But I am posting the two close up sections. Here is all I did. I moved the saturation curve in CM in the middle of the curve from 45 to 65. I could see the noise smooth out. It looks like I applied a filter of some kind.
I used a hue/sat layer in PS for the second example. I think it was about 20 points, but I just looked at the CM version and tried to match it. Huge noise increase; broke apart completely in some areas. The CM file (Mikedoesntbelieveit1) is with this post. The PS file (Mikedoesntbelieveit2) is posted separately.

As far as the Wiki article, Greg, if this turns out to be consistent and explainable, sure I'll describe and send you some examples. :) (Yep, don't like writing on the Wiki site.)

mikemeister_admin
Posts: 4927
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Wed Oct 31, 2007 11:31 pm

Hue/sat in PS

-default
Posts: 1916
Joined: Thu Mar 26, 2015 1:53 am

Postby -default » Thu Nov 01, 2007 6:00 am

Something interesting is definitely happening here.  I'll need a little time to think about this, but one thing I noticed, in Photoshop, is that the Brightness of the star shaped noise area changes significantly with the saturaiton adjustment.  It seems to me that an HSB or HSL saturation adjustment would change only the saturation.

mikemeister_admin
Posts: 4927
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Thu Nov 01, 2007 8:13 am

As a quick note, I am not getting the same result with other images I've tried, of high ISO NYC street night scenes...

mikemeister_admin
Posts: 4927
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Thu Nov 01, 2007 8:25 am

Gloria, there's a nagging thought in the back of my mind saying that somewhere I read that PS H/S adjustment layers increase noise more so than a&b channel adjustments. IIRC, it was something about H/S creating haloing.

It's not unusual that when you want to bump saturation, you want to do so in the colors that are less intense, and leave the ones that are already there alone.  In Curvemeister you can do this by moving the middle of the saturation curve upward, leaving the white end of same curve in place.  This protects the already intense colors from becoming blowing out, and at the same time adds more oomph to the medium colors.
This is really intriguing.

ggroess
Posts: 5342
Joined: Wed May 24, 2006 2:15 am
Contact:

Postby ggroess » Thu Nov 01, 2007 12:18 pm

First time you write for the Wiki it can be interesting...

I just copied the styles mike used and tried to learn a bit more as I went along...It was ugly at first but It gets better...

Greg


Return to “Curvemeister 3 Beta”

Who is online

Users browsing this forum: No registered users and 26 guests