View Full Version : Having some issues with greyscale oil painting
andymania
May 12th, 2008, 11:42 AM
I am currently working on a black and white oil study on wood panel. Now I am using Van Dyke Brown (Semi-transparent) and Titanium white only. What I discovered is that values 50% to 0% are great and they have just enough warmth as opposed to using regular black and white. However values above 50% that have more brown and less white tend to be way too warm and brown and too transparent.
1. How can I avoid the increase in chroma when going into darker values?
2. Is there an alternate brown or other color I dont know about that is opaque and has the same consistency as the Titanium white? I know Mars black does but it is much too cool.
Any ideas greatly appreciated as I am currently stumped.
dcorc
May 12th, 2008, 12:34 PM
Mix your black and white to get a set of (blu-ish) value-steps. Similarly, mix your brown and white to get the same (brown-ish) values. Then mix matching value steps together to get dead-neutral greys.
Dave
andymania
May 14th, 2008, 12:03 AM
Wouldn't it be easier to just mix the black and brown together (making that your new black) and just add white? That's what I am doing now and its seems to work fine.
dcorc
May 14th, 2008, 03:07 AM
Doing it the way I suggested gives more control, (and you'd also learn more from the process of doing it, too).
Puck
May 14th, 2008, 07:52 AM
I've used burnt umber and tit white - which when mixed loses alot of the warmth of the umber (though I've only done it with acrylics, not oils). But there's no way to avoid the increase in chroma as you're using less tit white and the warmth will come through, though I've always thought this was one of the nice things about using a brownish greyscale underpainting to layer glazes on (slightly warmer darks and cool lights).
Though like you said in your third post, mix a black that works for you and use that for a greyscale study (I'm liking burnt sienna and ultramarine blue at the moment - having fun playing with warm and cool transitions by tweaking the mixture).
Elwell
May 14th, 2008, 08:12 AM
Wouldn't it be easier to just mix the black and brown together (making that your new black) and just add white? That's what I am doing now and its seems to work fine.
That is fine, as long as you don't mind still having some color shift as you move up and down in value. If you want absolute neutrals all along the line, though, dcorc's way is the way to do it.
andymania
May 14th, 2008, 10:18 AM
Ok Dcorc,
I'll try it definitely. Im just trying to figure out how that method exactly works and how it eliminates chroma shifting as opposed to just mixing even amounts of brown and black in the first place. Why should mixing a brownish value 5 gray and bluish value 5 gray give you a completely neutral value 5 gray as opposed to just adding the right amout of white to a blue/brown mixture that is evenly mixed?
But I'll definitley give it a go!!
-Andy
Elwell
May 14th, 2008, 12:46 PM
Because, A: different pigments have different tinting strengths, so will effect a mixture disproportionally, and B: adding white, especially titanium white, shifts color towards purple-blue, so higher values may need more warming than lower ones.
vBulletin® v3.8.2, Copyright ©2000-2012, Jelsoft Enterprises Ltd.