Re: the freight car colour subject

Schuyler Larrabee

Mostly, I've decided not to participate in these
threads, since I >know< what's right, and the rest
of you can form your own erroneous opinions
without my help . . . 8^)))

But Dave in this message has made more sense to me
than the vast majority of previous posts. He's
analyzing the color WITHOUT regard to the light
impinging on the surface. Now, yeah, he's in
trouble already because he's analyzing the color
from a photo, and there's all the variabilities
introduced by THAT process, but at least he's
analyzing the color of an image he accepts as

It has always seemed to me that if you mix the
colors to be "correct" in daylight, and then take
the model into a fluorescently-lit room, the
colors will be perceived as "correct under this
lighting." After all, everything else there is
also being seen under the fluorescent lights, and
they, too, are driven off "correct" to "correct
under this lighting" just as much as the model's
color. Now, maybe my eyes aren't clicking on all
8 cylinders (or color perception mechanisms), but
to me, if it looks all right in daylight, my eyes
will accept the changed conditions under the

Anyway, proper models are weathered and that
changes the whole game a great deal.


-----Original Message-----
From: Dave Nelson [mailto:muskoka@...]
Sent: Monday, May 17, 2004 1:16 AM
To: STMFC@...
Subject: RE: [STMFC] the freight car colour

Rob, I think it's hopeless "To Know the Correct
Color (tm)"
of anything.
But that said, I personally find it advantageous
to have
scanned and color sampled various paints and
photos. For
instance, I learned that Modelflex
(MF) Light Tucsan -- a color I find acceptable
as a base for
creating PRR's Freight Car Red, is very low on
the cyan scale
and high in magenta.
Floquil's (FQ) ATSF Brown is basically even on
those two
pigments. And of course, darker. So if I
wanted to
approximate that FQ paint with MF I know it will
have to
include some very small quantity of blue (e.g.,
MF GN Big Sky
Blue) added to whatever I use as a tone.

I can also take a color photo of mine of an old
WP boxcar,
sample the browns (trying to pick one that
covers a large
area), and see the color is virtually identical
to Pantone
506CV. Now that can be communicated and
understood by all.
It doesn't say I can mix that paint, or that if
I do it will
look good on the car. But it does tell me
there's more
magenta in that color than any of my
out-of-the-bottle brown
paints... and that MF Maroon Tuscan might work
as a base into
which I add some Light Tuscan. Might work,
might not, but at
least I get pointed in a direction. And
somebody else can
understand what I'm trying to do and approximate
it with some
other mfgr's paint that they prefer.

What I never understand in this on-going debate
is why the
side that say's "You're a fool (tm)" to any
question/statement about getting more precise
than "it was
reddish brown" cannot grasp the logic needing to
go a few
steps farther along in our understanding of

What's so wrong with that?

Dave Nelson

-----Original Message-----
From: Rob Kirkham [mailto:rdkirkham@...]

Another way of puting the question is this: If I
accept that
the perception of colour is variable, that
colour is
fluctuating in light and with weather and
weathering, with
dust and grime and with the paint mix and
changing chemistry
over the decades, etc, etc., is it reasonable to
that the subject is a hopeless one, not worthy
of any study?

Yahoo! Groups Links


Join to automatically receive all group messages.