Thomas Ruckstuhl Art: February 2014

Thursday, February 27, 2014

Wednesday, February 26, 2014

Tuesday, February 25, 2014

Random Choice

60 min time-limited study. I went for a rather complex subject to make sure not be busy with details and work intuitively. Yes! A study were i see some improvement, finally.

Monday, February 24, 2014

Saturday, February 22, 2014

Landscape Study near Otterstadt

Friday, February 21, 2014

Self 1

The model would profit from exposure to sunlight and a shave.

Thursday, February 20, 2014

Wednesday, February 19, 2014

Monday, February 17, 2014

Sunday, February 16, 2014

Cordless drill driver without drill

Friday, February 14, 2014

Thursday, February 13, 2014

106 brushstrokes. I count and try to keep the number somewhat low. I do not really try to minimize it. The counting stops me from messing around with my brush and forces me to give more thought to each stroke in terms of color accuracy and the amount of paint needed.

CMPI Refinement

I did a some more color matching work and encountered a problem at evaluating the match of the grays. At low saturations the error in hue can become very large (up to 180) and can pull down the CMPI substantially. This can occur even if the matched color is not far away from target color in terms of overall distance in color space. To avoid this one could calculate this distance but i decided to keep it dead simple and come up with a quick and dirty remedy. Where hue error is beyond 50 i set it to the value of 50. To keep the tests comparable always 2 of the 10 squares have a (randomly generated) saturation below 10 and higher saturations for the other 8 squares.

The CMPI of my second test was 61.3. I will do dozens of these trainings in coming weeks and see whether the scatter of these tests results shows a tendency upward.

Wednesday, February 12, 2014

Rose Study

Tuesday, February 11, 2014

Color Matching Performance Index CMPI

The precise mixing of colors is the fundament of good painting. Alla prima paintings are done in one go and the colors mixed on the palette are transfered directly onto the canvas. Consequently the quality of color, or lack of it, is displayed mercilessly. Often, my colors are far off and i have to do corrections by either applying brushstrokes with better color on top or even use the palette knife to remove the paint. This is annoying as it leads to dirty color, slows me down and does no good to the end result. Therefore i have started to work on a simple training method to tackle the issue specifically. I can also use these trainings to measure my performance in color matching and whether i improve over time. It is quite plain....

I created 10 squares of randomized colour on the computer. I set the following limits for the colors in HSL color space:

Lightness : between 20 and 90 (to avoid colors too close to black (L=0) and to white (L=100)

Saturation: between 0 and 60 (to ensure that all colors are matchable with a standard palette of paints, such as the warm cool primary palette. Also, precise matching of highly saturated colors is somewhat less important than matching the grays)

Hue: 0 to 360, full spectrum.

Then I printed these 10 color squares and 10 blanks to be filled with paint.

For the painting part i set up simple rules.

- The 10 colors have to be matched within 10 mins. If not it's a fail... restart.

- I match the color i see on the paper on my palette and then apply it, without the possibility for any correction. I am also not allowed to hold the brush with the paint against the color square to compare it. I am free, however, to work sequentially. So i mix one color, apply it to the blank, mix the next one and so on. This way, i can to a certain degree learn from the mismatch of the prior squares.

My first results:

I knew it would not be easy but it was harder than i thought and the deviations felt shocking here and there. Striking is that all my mixes are too dark and the error there is quite substantial and the L-value is most crucial for painting. My saturation on the other hand seems quite okay ;-) Thanks to technology it is easy to quantify things a bit better. So I used the photo to determine the deviations of hue, lightness and saturation for each spot and calculated my average deviation from target color to be 13 in hue, 12.6 in lightness and 4.1 in saturation.

Finally, i define a color matching performance index (CMPI):

CMPI = 100 - <dH>/2 - <dS> - 2 * <dL>

(In words: hundred minus half of the average hue deviation minus average deviation in saturation minus two times the average deviation in lightness).

I derived this simple formular empirically after testing for a while. If you are an artist in doubt... hey trust me i am a physicist phd. If you are a scientist in doubt, i refer to my freedom as an artist. If you are both... great let's discuss!

The CMPI from my first test is 64.2, with plenty of space towards the ideal of 100.