I have been quiet about this latest thread about color
"given different light, amount of weathering, time of
day....". But the old myth of color having scale has
been repeated so often it has been taken by some as a
truth. Color does not have a scale. Colored light does
come in differing intensity, etc. etc., but it does
not scale. (Maybe at an electronic microscopic level!)
If you change the colors on differing scale models, at
most, only one will be correct. I leave with this
paradox; any panel of an O scale car is smaller than a
whole HO car. How can you reconcile this? If this
truism is correct, a 10 panel car would need to be
painted differently than a welded side car. A real
(12inch/foot scale)86' boxcar under this theory would
have to be painted darker than a 40' car.
-Skeptically yours,
-Andy


--- Charlie Vlk <cvlk@...> wrote:

Andy Carlson writes:


"Color does not have a scale. Colored light does
come in differing intensity, etc. etc., but it does
not scale."

I believe the point is, while color might not change, a human's perception of does...depending on the amount of other light being received at the same time. Hence, if one is so close to a painted object that light from other reflective surfaces does not appear in view, the color one see's will be more dominant than it will if viewed from further away IF, from that view, reflective light from other quite different objects is seen at the same time.

While I can certainly see the need and interest in striving to determine the color of an original car...particularly for newly painted or new condition cars...the many variables introduced to modify the perception and actual color of an in service car seem to make stringent adherence to the original color less important. IMO. IOW, when I view a photo of a train of PFE cars and note untold numbers of different shades of PFE orange or recall watching a very grey "appearing" ex-C&O 2-8-4 2716 [ even though it was painted with rather black paint ]in the afternoon sun, color accuracy seems not as necessary on items operating on a layout in which changing variables ARE present...i.e., weather, variable light, dirt etc.

Mike Brock

--- Mike Brock <brockm@...> wrote:

I believe the point is, while color might not
change, a human's perception of does...depending on
the amount of other light being received at the same
time.

I think the most fundamental difference in perceived
color is light intensity. An average train room is lit
about 7 or more F-stops less than the good ol
outdoors. In a marginally lit indoor train room,
lightening the color could be argued as being best FOR
THESE CONDITIONS. However, if one is to view the
exquisit models members of this list are creating
under real sunlight, I maintain the most accurate
color is the one the prototype used.

Another point; blending paints to achieve color
matches has an inherent problem- using multiple
pigments causes multiple color perceptions when viewed
under varying light sources. I learned this when first
painting with Accu-Paint color mixes suggested to me
as accurate for SP&S Locomotives. The green must have
looked real good to the suggester. I found that
viewing the finished loco in sunlight the green was
horrible. Under incandescent light it was close to
acceptable, and of course under florescent it was
worse still. Following a suggestion from Jack Parker,
who said the real railroads "pullman green" was a
blend of black and chromium yellow (both, incidently,
are mono-pigmented) I mixed my own 2 color PG from
Automotive lacquers, and now have a green which views
favorably under the 3 differing light sources. Less
pigments means less variability. If the green looks
too dark, increase the lighting. In the real world,
dark objects look dark in the waning light.
-Andy Carlson

Andy, I agree totally. As I illustrated with this pic a while ago.
Same car, same day, same roll of film -- change of time only.

http://groups.yahoo.com/group/STMFC/files/nyc_40ft_DD_box_Essex_1998.JPG

In the real world, dark objects look dark in the waning light.
-Andy Carlson

Tim O'Connor <timboconnor@...> -->> NOTE EMAIL CHANGE <<--
Sterling, Massachusetts

Andy Carlson notes:

"I think the most fundamental difference in perceived
color is light intensity. An average train room is lit
about 7 or more F-stops less than the good ol
outdoors. In a marginally lit indoor train room,
lightening the color could be argued as being best FOR
THESE CONDITIONS. However, if one is to view the
exquisit models members of this list are creating
under real sunlight, I maintain the most accurate
color is the one the prototype used."

Perhaps. A couple of things. First, perception indoors is not driven just by the amount of light, it is driven just as much by the type of light. I use florescent warm white to help produce the very reddish look of southeastern Wyoming. Oddly, I paint outside and I've certainly encountered the different appearance of painted objects under 5300 kelvin [ if I recall the temp of sunlight correctly ] and my indoor florescents. Next, as I've said before, the color is very dependent upon atmospheric conditions and sun angles. Having spent a great deal of time shooting photos of Daylight 4449 and "J" 611, I can tell you that the appearance does change. Again, at Toccoa, GA, at around 5PM I turned to watch 2-8-4 2716, painted black, move by at about 100 yds away and was amazed at how grey it APPEARED. Last...while there may be those that build and paint for viewing in direct sunlight, I don't know any.

Regarding Tim's photos and "waning" light, I guess I don't quite get the point. Of course in darkness dark subjects are going to appear dark. So are light colored objects. OTOH, sun angles produce very different results. At midday the side of a black engine will look damn dark. At 5 PM in May in Georgia the side of a black engine will look quite light colored...some shade of grey... if viwed with the sun to your back. The other side will look damn black. Same thing will happen to any other color.

Every 15 months we all seem to get on here with our color ideas. No problem...just a fact. Each modeler has the opportunity to go out in the real world with some painted object, turn it in different directions to the sun, cover it with dirt or dirty water and turn it again. Then they can do it at different times of the day. They can decide for themselves if they think the object's appearance changes with changes in sunlight, sun angle, dirt etc. They then have the difficult task of replicating the colors they think they see inside a train room lit up by all kinds of possible light sources. It's what we do...I guess.

Mike Brock

Andy-
The effect of the perception of color by the human eye varying by the size
of the object being observed is a function of the amout of light received by
the rods in our eyes....and has been the subject of scientific articles (the
one I recall was in Scientific American within the last five years.....
Yes, the panel on an O Scale car is smaller than a whole HO car. but the
entire O Scale car is going to deliver more light (percieved color) than the
HO car can deliver. You might be right about the hicube 86' car vs. the 40'
car, although at that scale the amout of light being dilivered is not as
great a difference as between a 1/1 car and a 1/87 car.
Others have also mentioned the effect of indoor lighting and different
temperature light sources.....
Proof of this is paint your PRR locomotives with REAL PRR paint.... you will
never see any of the green in such a small sample...yet the prototype, when
clean, has a definite greenish tint to it....
Sometimes you have to use artistic license to make the eye see what you want
it to see......that is why it is called modeling rather than prototype
reductioning!!!
Charlie Vlk