Why is my flash not the same as daylight
john
taken by the flash do not look exactly the like as same colours taken
in bright daylight.
That's strange because years ago I had learnt (perhaps simplisticlly)
that the two light sources were the almost exactly same.
I'm now questioning if the colour settings on a digital camera for
flash and bright daylight are actually the same thing.
Any info on this please.
J
Michael
My point and shoot digital Nikon, when setting the white balance
manually, distinguishes between flash and sunlight and cloudy etc. So
no, they are not exactly the same thing. Close, yes.
--
Michael
Pat
Even if they were the same, on a small flash the ambient light could
create a color cast.
Don Klipstein
Even if overall color of the flash matches some form or another of
"daylight", with smaller capacitor value in a (usually) lower energy
"photoflash" the color rendering properties can still change.
With greater capacitance of the "energy storage capacitor", xenon
flashlamps tend to have spectrum closer to that of blackbody at color
temp. usually 5500 to 6000 K or so. When capacitor value gets smaller,
the spectrum of a xenon flashlamp can significantly change to one that
has altered color rendering properties - notably red objects being
rendered darker, along with slide/movie film sensing color temperature to
be higher than that sensed by human eyes, and slide/movie film also
sensing the overall color not only bluish but also greenish.
Usual smaller lower-energy "cheap" xenon flashlamp in my experience is
"BGA0013" or something that approximates that one.
Correlated color temp. of that one appears to me to be close to 6000 K
rather than the widely-spouted 5500 at usual voltages of/around 275-330
volts. Energy storage capacitors with value below or even around 47
microfarads can result in the spectrum changing to one disfavoring
daylight-like color rendition.
http://members.misty.com/don/xea.html#1
I suspect that dimming of camera flashes achieved by curtailing flash
duration can also change the spectrum if flash duration is shortened to
around or under that achieving majority of discharge of the minimum
capacitor value necessary to achieve "daylight-like" spectrum.
- Don Klipstein (d...@misty.com)
Pete D
"john" <no...@nowhere.com> wrote in message
news:Xns9BC22606...@194.177.98.144...
The sun is yellow, your flash is brilliant white.
Ray Fischer
>My compact digital camera has a tiny flash. I notice that colours
>taken by the flash do not look exactly the like as same colours taken
>in bright daylight.
>
>That's strange because years ago I had learnt (perhaps simplisticlly)
>that the two light sources were the almost exactly same.
Hmmm. The problem is that there really is no one "daylight" color.
The sun is yellow, the sky is blue, and the atmosphere filters out
blues. Depending on the time of day, the clarity of the air, the
humidity, and reflection from nearby objects, "daylight" color can
vary by quite a bit.
--
Ray Fischer
rfis...@sonic.net
Ioannis
Small Wattages or relatively low levels of illumination with various sources, in
my opinion tend to introduce color distortion.
To achieve a truer color rendering close to the one specified by the source
used, fairly high illumination levels are necessary.
A small (say 10 - 20W) "daylight" fluorescent, for example, will usually distort
colors and will not produce a true "daylight" illumination, unless some fairly
high illumination levels (at least 1,000 - 1,500 lux) are used.
My Nikon 2200 Coolpix has the same problem, too. I usually have to adjust the
contrast and color levels in Photoshop after I take a picture, otherwise the
picture is too gloomy :-)
> J
--
Ioannis --- "There's _always_ a mistake, somewhere".
Don Stauffer
The idea of a color temperature is primarily aimed at the light from a
thermal (broadband) source, that is, blackbody continuous emission
across the spectrum, rather than one or more individual spectral lines.
While the use of several judiciously picked spectral lines can
approximate a black body, there are not an infinite number of gases and
lines, so picking the right gases for a flash tube is a great art.
Flash tubes TRY to duplicate sunlight, but it is VERY, VERY hard to do
with line emission.
Sam
My understanding of daylight is that it's colour varies according to
environmental conditions. So it's impossible to have a flash calibrated to
match it, as it changes too much.
Of course, it's entirely possible that my understanding is flawed.
The Royal Spam
I was told by a lab technician that daylight balanced film was balanced for
June 21st in Washington D.C. at noon.
Sam
"The Royal Spam" <spam...@dontspamme.com> wrote in message
news:C5D1C2FA.15374%spam...@dontspamme.com...
I actually sprayed my coffee when I read that.
Celcius
"The Royal Spam" <spam...@dontspamme.com> wrote in message
news:C5D1C2FA.15374%spam...@dontspamme.com...
>
>
>>
My God! Is it actually always sunny at that time in Washington DC?
What if you live in London, Madrid or Havana... well maybe there's an
embargo in that last city ;-)))))
Marcel
Peter Irwin
>
>
> I was told by a lab technician that daylight balanced film was balanced for
> June 21st in Washington D.C. at noon.
He was confused. Mean Noon Sunlight was based on the average of
readings taken at noon on 21 June and 21 December 1926 at the National
Bureau of Standards in Washington. It was of the sun only and
not daylight (sun + sky). It is nearly equivalent to a colour
temperature of 5400K.
Photographic Daylight is 5500K light - not Mean Noon Sunlight.
Daylight at noon (sun plus sky) is usually closer to 6500K,
but has a lower colour temperature at other hours of the day.
5500 K is probably a pretty good average for daylight, but it
isn't directly connected to your story which is a confused remembering
of the definition of Mean Noon Sunlight.
Peter.
--
pir...@ktb.net
Kennedy McEwen
<no...@nowhere.com> writes
>My compact digital camera has a tiny flash. I notice that colours
>taken by the flash do not look exactly the like as same colours taken
>in bright daylight.
>
Flash, primarily individual emission lines of noble gases, can never
produce the same colours as daylight, a mixture of various filtrations
of the sun's blackbody emission.
Its that same effect that makes your wife/girlfriend want to return
those shoes/dress/coat/sweater because they don't match the rest of the
outfit when they get home, even though they were perfect in the store.
--
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's pissed.
Python Philosophers (replace 'nospam' with 'kennedym' when replying)
TKM
"Peter Irwin" <pir...@ktb.net> wrote in message
news:goh8jm$96n$1...@dns.ktb.net...
It is more scientific that what has been said. Daylight has been defined by
the International Commission on Illumination (CIE) and there's a good
article about that at: http://en.wikipedia.org/wiki/D65 Lighting folks,
however, specify two dimensions of light with respect to color:
chromaticity and color rendering. Color rendering takes into account (at
least somewhat) the width of the light source spectrum when measured as
output vs wavelength. Chromaticity (expressed as CCT) doesn't do that.
TKM
Alan Browne
The colour temperature of flash is very close to midday sun. But
sunlight photography is always affected by:
-angle of arrival. Midday is not the same as early morning, morning,
afternoon, late day. Your flash fires from near the lens - sunlight
arrives from different angles relative to the subject.
-haze, smog, dust ...
-clouds (from thin to thick)
-local objects 'colorizing' the light by reflection (trees, walls,
ground, ...)
--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.
-- usenet posts from gmail.com and googlemail.com are filtered out.
Alan Browne
The sun isn't yellow. Or everything would look yellow.
It has a yellow look to it (contrasted to the blue sky, in any case),
but it is quite white.
Alan Browne
> I was told by a lab technician that daylight balanced film was balanced for
> June 21st in Washington D.C. at noon.
Except for old French films which were daylight balanced in Paris.
British films were daylight balanced at Greenwich which is why, on
average, they tend to result in very punchy/contrasty films. (Greenwich
Mean Balanced Daylight Filn). It's now called Universal Film Daylight
Balanced Coordinated, a compromise backronym resulting from French and
British ISO delegates quarreling.
John McWilliams
It isn't that far fetched. T.o.D. makes a huge difference; seasons makes
a difference; and location can be important.
Weird thing is D.C. not 'zackly a sunny spot.
--
John McWilliams
ASAAR
>> I was told by a lab technician that daylight balanced film was balanced
>> for June 21st in Washington D.C. at noon.
>
> My God! Is it actually always sunny at that time in Washington DC?
> What if you live in London, Madrid or Havana... well maybe there's an
> embargo in that last city ;-)))))
Londoners are most likely to know, and there are probably always
some of them in Washington DC, Madrid and even a few in Havana,
because Mad dogs and Englishmen go out in the midday sun.
Twibil
>
> Londoners are most likely to know, and there are probably always
> some of them in Washington DC, Madrid and even a few in Havana,
> because Mad dogs and Englishmen go out in the midday sun.
So if you want to know about a strobe's color balance versus daylight
you should go ask a rabid Rottweiler?
Hmmm...
ASAAR
>> Londoners are most likely to know, and there are probably always
>> some of them in Washington DC, Madrid and even a few in Havana,
>> because Mad dogs and Englishmen go out in the midday sun.
>
> So if you want to know about a strobe's color balance versus daylight
> you should go ask a rabid Rottweiler?
>
> Hmmm...
Absolutely not! Don't you know the Rottweiler's motto?
"We own the night. Rrrrrr . . ."
You'd be better off asking a Chihuahua, or perhaps Dolce, Maria
Sharapova's sweet little pooch. That's a Pomeranian, by the way.
Kennedy McEwen
<alan....@Freelunchvideotron.ca> writes
>Pete D wrote:
>> "john" <no...@nowhere.com> wrote in message
>>news:Xns9BC22606...@194.177.98.144...
>>> My compact digital camera has a tiny flash. I notice that colours
>>> taken by the flash do not look exactly the like as same colours taken
>>> in bright daylight.
>>>
>>> That's strange because years ago I had learnt (perhaps simplisticlly)
>>> that the two light sources were the almost exactly same.
>>>
>>> I'm now questioning if the colour settings on a digital camera for
>>> flash and bright daylight are actually the same thing.
>>>
>>> Any info on this please.
>>>
>>>
>>> J
>> The sun is yellow, your flash is brilliant white.
>
>The sun isn't yellow. Or everything would look yellow.
>
;-)
Even under red, black and white darkroom light, everything looked black
and white. Even more so under completely monochrome (apart from
trivial, in this context, magnetic splitting) yellow sodium 'd' lines of
colour darkroom light, there was just white, black and shades of grey
between.
T'is a r'markable instrument, the human eye!
And there lies the problem.
Twibil
> On Mon, 2 Mar 2009 16:58:30 -0800 (PST), Twibil wrote:
> >> Londoners are most likely to know, and there are probably always
> >> some of them in Washington DC, Madrid and even a few in Havana,
> >> because Mad dogs and Englishmen go out in the midday sun.
>
> > So if you want to know about a strobe's color balance versus daylight
> > you should go ask a rabid Rottweiler?
>
> > Hmmm...
>
> Absolutely not! Don't you know the Rottweiler's motto?
>
> "We own the night. Rrrrrr . . ."
Thus my 6' fence and my two Rotties. (Why, my assault rifle and my
infantry training are practically useless these days!)
~Pete
Chris Malcolm
> Pete D wrote:
>> "john" <no...@nowhere.com> wrote in message
>> news:Xns9BC22606...@194.177.98.144...
>>> My compact digital camera has a tiny flash. I notice that colours
>>> taken by the flash do not look exactly the like as same colours taken
>>> in bright daylight.
>>>
>>> That's strange because years ago I had learnt (perhaps simplisticlly)
>>> that the two light sources were the almost exactly same.
>>>
>>> I'm now questioning if the colour settings on a digital camera for
>>> flash and bright daylight are actually the same thing.
>>>
>>> Any info on this please.
>>>
>>>
>>> J
>>
>> The sun is yellow, your flash is brilliant white.
> The sun isn't yellow. Or everything would look yellow.
> It has a yellow look to it (contrasted to the blue sky, in any case),
> but it is quite white.
What we consider to be white is the integrated illumination from both
the sun and the blue sky. Astronauts see the sun (with suitable sun
glasses :-) as white in a black sky. Down here at the bottom of the
atmospheric ocean the air takes a lot of the blue from the sun's
radiation and spreads it over the sky. That's why (as the French
impressionist painters famously "discovered") direct sunlight has a
golden tinge, and the shadows of direct sunlight a blue tinge. And
that is why artists involved in subtly colour work like what is called
white north skylight (in the Northern hemisphere). That's light which
has been diffued trough clouds so that the blue sky and golden sun are
reintegrated into white.
The reason these colour casts are not obvious to us, and why light
always looks a lot whiter than it really is, is that our eye and brain
operates a very effective auto white balancing, which in turn is the
reason why we need white balancing in our cameras, in order to make
things look like the colours we see, instead of the colours they
actually are in terms of spectral power.
--
Chris Malcolm
Don Klipstein
I have been in darkrooms before, and I also have been in areas
illuminated by LPS lamps before. And sometimes for hours rather than
minutes.
My experience is that if the light is red, I see everything im
black-and-red. And under LPS, my experience is seeing everything as a
ghastly black-and-yellow, with exception for a few reddish fluorescent
labels that manage to fluoresce from the 589.0/589.6 nm "D" lines.
Meanwhile, the color of sunlight varies. At zenith with nice clean
clear air, it's 5300-5400 Kelvin or so. During or near midday on clear
days in Philadelphia almost any time of the year, or even in midday on
moderately hazy days in late spring and early summer, I have found
sunlight to be slightly less yellowish (slightly more bluish) than
nominally 4100 K fluorescent lamps. At "high noon" on an especially clear
June or late May or early July day, I have seen Philadelphia sunlight
match the color of (or get very slightly more bluish than) several
nominally 5000 K CFLs.
Xenon photoflash also varies slightly in color temperature, even when
capacitance is sufficient to get visible spectral content mainly in a
continuum (which is usually true). 5500 K is a widely-mentioned number,
but I have seen some more like 6000 K. Some units have yellowish or
yellowish-orangish filtering "lenses", which I suspect changes the color
to something that appears white to 5500 K color slide film.
- Don Klipstein (d...@misty.com)
Don Klipstein
>john wrote:
>> My compact digital camera has a tiny flash. I notice that colours
>> taken by the flash do not look exactly the like as same colours taken
>> in bright daylight.
>>
>> That's strange because years ago I had learnt (perhaps simplisticlly)
>> that the two light sources were the almost exactly same.
>>
>> I'm now questioning if the colour settings on a digital camera for
>> flash and bright daylight are actually the same thing.
>
>thermal (broadband) source, that is, blackbody continuous emission
>across the spectrum, rather than one or more individual spectral lines.
> While the use of several judiciously picked spectral lines can
>approximate a black body, there are not an infinite number of gases and
>lines, so picking the right gases for a flash tube is a great art.
>Flash tubes TRY to duplicate sunlight, but it is VERY, VERY hard to do
>with line emission.
Xenon flashlamps generally produce nearly all of their visible light in
a continuous spectrum if the energy storage capacitor has sufficient
capacitance.
I give approximate minimum capacitor values for achieving such a
spectrum with several xenon flashlamps in:
http://members.misty.com/don/xea.html
(Missing figures on at least 2 flashtubes will be in tonight.)
Most photoflash units have sufficient capacitance.
- Don Klipstein (d...@misty.com)
Don Klipstein
Daylight balanced color slide film is supposed to be balanced for
"average"/"usual" combinations of direct sunlight and whatever-is-usual
fraction of the light from clear blue sky (not all of it).
Sunlight-plus-skylight should usually be fairly close in color
temperature to that of sunlight above the atmosphere - upper 5700's K.
I have seen some cite for color temp. of direct sunlight in Washington
DC, being 5200 K. I suspect that is at high noon on/near June 21st if the
air is especially clear.
- Don Klipstein (d...@misty.com)
Don Klipstein
>In rec.photo.digital The Royal Spam <spam...@dontspamme.com> wrote:
>>
>> I was told by a lab technician that daylight balanced film was balanced for
>> June 21st in Washington D.C. at noon.
>
>He was confused. Mean Noon Sunlight was based on the average of
>readings taken at noon on 21 June and 21 December 1926 at the National
>Bureau of Standards in Washington. It was of the sun only and
>not daylight (sun + sky). It is nearly equivalent to a colour
>temperature of 5400K.
Can you provide a cite supporting all of these details?
>Photographic Daylight is 5500K light - not Mean Noon Sunlight.
>Daylight at noon (sun plus sky) is usually closer to 6500K,
Maybe to color slide film if all possible UV is included and the film
wants some of it removed. I am aware of the numerous web pages citing
Kodak for some awfully high figures for color temperature of sunlight and
daylight.
I live a bit west of Philadelphia, northeast of Washington DC and
roughly or a bit under 1.5 degrees farther north in latitude than DC is,
and no way am I going to believe that a usual noontime combo of sunlight
and skylight in this geographic area is usually or even often more bluish
than 6000 K. I suspect someone got the story incorrectly somehow and
after that point it managed to spread.
Since sunlight above Earth's atmosphere has color temp. around 5800 or
in the upper 5700's K, I have a hard time believing that a combo of
midday direct sunlight and light from the sky is usually more bluish than
that, unless the sun is low enough to be noticeably dimmer than usual.
I also see so often how daylight compares to various articial light
sources of various color temperatures. For one thing, overcast sky
usually appears to me less bluish than every nominally 6500 K fluorescent
that I ever got to compare to overcast sky. Sunlight-plus-skylight
usually appears to me less bluish still.
Yes, I am aware of cited measurements of light from overcast sky over
snowy ground having color temp. in the mid 7,000's K - and I have a hard
time believing that this is a usual color temp. for daylight as opposed to
one on the bluish side, short of bluishness of clear sky (varies widely,
often as low as 7500 K, often as high as 20,000 K, and chromaticity is
often anywhere from impressively close to that of blackbody to .01-plus
"distance unit" on 1931 CIE diagram displaced from "blackbody locus"
towards green. To a lesser extent, clear sky has greater extremes to more
blue than infinite color temperature, purpler than blackbody, and outright
pinkish. If urban air quality is poor, then in midday most of the sky can
be clear and have color temp. in the 6,000's.
>but has a lower colour temperature at other hours of the day.
>5500 K is probably a pretty good average for daylight, but it
>isn't directly connected to your story which is a confused remembering
>of the definition of Mean Noon Sunlight.
I have a very hard time believing yellowed only to 5400 K from 5800 or
upper 5700's above the atmosphere. Especially since my experience is that
only the least yellowish noon sunlight gets to or barely past 5000 K at
about 39.9 or 40 degrees north latitude in a western suburb of
Philadelphia. " 'High noon' sunlight here appears to me to range from
4400 to 5100 when not made lower by air pollution. On smoggier days even
in the city (in Philadelphia), sunlight at "high noon" usually appears to
me to range from 4200 to 4600 K. I do see smog being mainly a summertime
issue, and affecting color of sunlight more at time of day when the sun is
low or moderately low, at a time of year when the sun gets closer to the
zenith at "high noon" than it does on an "equinox day".
>Peter.
- Don Klipstein (d...@misty.com)
Don Klipstein
The above Wiki article on D65 appears to me generally good, with the
exception of:
"D65 corresponds roughly to a mid-day sun in Western Europe / Northern
Europe"
I have yet to see midday direct sunlight that bluish anywhere in my life,
and I have been above about 70% of the atmosphere in early summer at high
noon at 40-41 degrees north latitude, and I have been above 74% of the
atmosphere at high noon at about 37-38 degrees north latitude in late
October. I have also experienced high noon sunlight near Orlando in early
July with especially clear atmosphere (and other conditions common there),
around 28-28.5 degrees N latitude.
And my several July experiences in Toronto and some northern suburbs
thereof (at least 44 degrees north latitude) suggest to me that direct
sunlight does not get more bluish farther north from where I live. My
Florida experience suggests that direct sunlight gets slightly more bluish
when the sun gets closer to zenith.
I look at things like this a lot because I am a lighting-and-color nut,
and noontime direct sunlight (with exclusion of light from the sky)
appears to me to have a high rate of being in the range of 4200 to 5100 K.
When sun is closer to zenith than is possible at 40 degrees latitude or
when altitude is far above sea level, I see more bluish but short of the
upper 5700s or 5800-or-so of direct sunlight above the atmosphere.
Combo of sunlight and skylight, with sun higher and brighter, appears
to me to do well at approximating 5500 K or so. With sun within 30-40
degrees of zenith and effort taken to include rather than exclude light
from the sky, or on horizontal plane with sun at zenith, I see 5700 or
5750 or so (over 5500, under 6000, generally/mostly not significantly
exceeding 5800) being reasonable.
Horizontal plane with lower sun I so see fair chance to get past 6000
or maybe around or over 6500. (Heck, at 1 minute after sunrise or 1
minute before sunset with especially clear air, it appears to me that a
horizontal plane illuminated by unobstructed clear sky can easily
experience overall illumination color temp. exceeding 8000 K.) However,
horizontal plane with low sun (sun much less than 30 degrees above
horizon) is something I would dispute being a good representation of
"direct sunlight" or "direct sunlight plus sky light".
- Don Klipstein (d...@misty.com)
Peter Irwin
> In article <goh8jm$96n$1...@dns.ktb.net>, Peter Irwin wrote:
>>In rec.photo.digital The Royal Spam <spam...@dontspamme.com> wrote:
>>>
>>> I was told by a lab technician that daylight balanced film was balanced for
>>> June 21st in Washington D.C. at noon.
>>
>>He was confused. Mean Noon Sunlight was based on the average of
>>readings taken at noon on 21 June and 21 December 1926 at the National
>>Bureau of Standards in Washington. It was of the sun only and
>>not daylight (sun + sky). It is nearly equivalent to a colour
>>temperature of 5400K.
>
> Can you provide a cite supporting all of these details?
>
Manual of photography. Focal Press 9th edition page 25.
I own a hard copy, but google books is convenient online.
It is a good solid book, successor to the Ilford Manual.
Peter.
--
pir...@ktb.net
extremewanderer
>
> - Show quoted text -
colours are dependant on the light.
There is something called colour rendition index (CRI)
http://en.wikipedia.org/wiki/Color_rendering_index
some flourescent tubelights and cfls actually heighten blues.
Kennedy McEwen
<d...@manx.misty.com> writes
red lit darkroom and a few years later a similar amount of time in a
gold filtered sodium lit darkroom. I can assure you, since you cannot
discern red/yellow from white under such light, after a few minutes
everything just appears white and shades of grey. When you watch an
image appear in the developer dish, it is a black and white image, not
black and red. Similarly, monitoring colour prints develop they also
appeared black and white, not black and yellow, or any other colour.
>>T'is a r'markable instrument, the human eye!
Doesn't sound like your's are as remarkable as most folks. ;-)
Which may be advantageous in your line of work
Alan Browne
> In article <ULOdnVDlbN6JyjHU...@giganews.com>, Alan Browne
> <alan....@Freelunchvideotron.ca> writes
>> Pete D wrote:
>>> "john" <no...@nowhere.com> wrote in message
>>> news:Xns9BC22606...@194.177.98.144...
>>>> My compact digital camera has a tiny flash. I notice that colours
>>>> taken by the flash do not look exactly the like as same colours taken
>>>> in bright daylight.
>>>>
>>>> That's strange because years ago I had learnt (perhaps simplisticlly)
>>>> that the two light sources were the almost exactly same.
>>>>
>>>> I'm now questioning if the colour settings on a digital camera for
>>>> flash and bright daylight are actually the same thing.
>>>>
>>>> Any info on this please.
>>>>
>>>>
>>>> J
>>> The sun is yellow, your flash is brilliant white.
>>
>> The sun isn't yellow. Or everything would look yellow.
>>
> Actually, if the sun was yellow everything in daylight would look white.
> ;-)
>
> Even under red, black and white darkroom light, everything looked black
> and white.
Hmm. To me a red flooded DR looked black and red.
> Even more so under completely monochrome (apart from
> trivial, in this context, magnetic splitting) yellow sodium 'd' lines of
> colour darkroom light, there was just white, black and shades of grey
> between.
Perception aside, the sun is much more white than yellow. That was the
point. eg: in white light, objects colours are revealed. In monochrome
light, everything is from black to shades of that colour.
Alan Browne
> In rec.photo.digital Alan Browne <alan....@freelunchvideotron.ca> wrote:
>> Pete D wrote:
>>> "john" <no...@nowhere.com> wrote in message
>>> news:Xns9BC22606...@194.177.98.144...
>>>> My compact digital camera has a tiny flash. I notice that colours
>>>> taken by the flash do not look exactly the like as same colours taken
>>>> in bright daylight.
>>>>
>>>> That's strange because years ago I had learnt (perhaps simplisticlly)
>>>> that the two light sources were the almost exactly same.
>>>>
>>>> I'm now questioning if the colour settings on a digital camera for
>>>> flash and bright daylight are actually the same thing.
>>>>
>>>> Any info on this please.
>>>>
>>>>
>>>> J
>>> The sun is yellow, your flash is brilliant white.
>
>> The sun isn't yellow. Or everything would look yellow.
>
>> It has a yellow look to it (contrasted to the blue sky, in any case),
>> but it is quite white.
>
> What we consider to be white is the integrated illumination from both
> the sun and the blue sky. Astronauts see the sun (with suitable sun
Blue sky only shows in the shade (aka: "open shade"). The contrast
ratio of sunlight to blue sky fill is about 4 stops (16:1) so its
contribution to an exposure is pretty much nil in the sunlit portion of
the image which is what this question is about: flash v. sunlight (not
blue-sky fill light).
> glasses :-) as white in a black sky. Down here at the bottom of the
> atmospheric ocean the air takes a lot of the blue from the sun's
> radiation and spreads it over the sky. That's why (as the French
> impressionist painters famously "discovered") direct sunlight has a
> golden tinge, and the shadows of direct sunlight a blue tinge.
Only in golden hour light (or at least late afternoon to golden hour) is
anything yellow tinged.
The blue is simply open shade lighting under blue skies. Here, with
lots of snow on the ground open shade blue is anything but subtle.
Alan Browne
> Daylight balanced color slide film is supposed to be balanced for
> "average"/"usual" combinations of direct sunlight and whatever-is-usual
> fraction of the light from clear blue sky (not all of it).
The ratio of blue fill to sunlight is about 1:16. Virtually nothing for
a sunlight exposure.
frank tarkinsen
This proves your perception wrong. The sun is neither white nor yellow. You
have lived under a bright green star all your life so you perceive green
shifted light as white, proving the other person's point.
Chris Malcolm
>> Daylight balanced color slide film is supposed to be balanced for
>> "average"/"usual" combinations of direct sunlight and whatever-is-usual
>> fraction of the light from clear blue sky (not all of it).
> The ratio of blue fill to sunlight is about 1:16. Virtually nothing for
> a sunlight exposure.
Very little effect on exposure. Quite a significant effect on white
balance.
--
Chris Malcolm
Don Klipstein
>Don Klipstein wrote:
>
>> Daylight balanced color slide film is supposed to be balanced for
>> "average"/"usual" combinations of direct sunlight and whatever-is-usual
>> fraction of the light from clear blue sky (not all of it).
>
>The ratio of blue fill to sunlight is about 1:16. Virtually nothing for
>a sunlight exposure.
Just to toss in some numbers, I will figure out what one gets with 1
part of 12,000 K (my estimate of "average blue sky") plus 16 parts of 4600
K (my estimate of "average midday sunlight at 39.9-40 degrees latitude").
Result of my calculations here is 4825 K - 225 K more blue than direct
sunlight.
I try again for 1 part 15,000 K plus 16 parts 4800 K. I figure there
about 5055-5060 K, 255-260 K more blue than "sunlight alone".
I try once more for one part 2,000,000 K plus 16 parts 5200 K. All
16-to-1 ratios I am here considering to be photometric ratios as opposed
to radiometric.
The extreme of 1 part 2 million K to 16 parts 5200 K with 1:16 ratio
being photometric, I calculate to correlated color temp. around 5575 K.
I may decide to not argue against as high as 5600. The light from the
blue sky in this extreme case boosts color temp. of direct sunlight by
375-400 K.
It does appear to me, based on literature that I have exposed myself to,
that sunlight alone achieves 5200 K for high extreme achievement at
Washington DC. The sun gets to 15.5-16 degrees from zenith there at "high
noon" in mid/late June. And I suspect that the light from blue sky on
such a clear day, overall from the sky, is more like 20,000 or 17,000 K
rather than pushing infinity. However, I find correlated color temp.
using this extreme combo (1 part 20,000 K plus 16 parts 5200 K with 1/16
ratio being photometric) to be around 5525 K or so. Light from blue sky
boosts correlated color temp. from that of sunlight by 325, maybe 330 K.
I suspect that color temp. boost from light from blue sky adding to
direct sunlight is maximized when the sun is low, dimmed, and having low
correlated color temperature below 2,000 K.
- Don Klipstein (d...@misty.com)
Don Klipstein
I appreciate the link/reference, thanks and whatever related "happy
stuff", but I still have difficulty believing that average "high noon"
ditrect sunlight with exclusion of light from sky/clouds at any low
elevation place on planet Earth at latitude 38.5 degrees exceeds 5000 K.
I suggest research into derivation of "Standard Illuminant B" -
supposedly noontime sunlight somewhere, with color temp. in the mid or
upper 4800's.
I do appreciate your cited source saying that 5500 K is "photographic
daylight".
I would question the statement a paragraph or two later that "near noon,
the combination of light from the sun, sky and clouds has a color
temperature in the region of 6000 to 6500 K."
Color temp. of sunlight above Earth's atmosphere is slightly less than
6000 K, and I have a hard time believing that most midday daylight on
sunnier days in sunnier areas of the world is more-blue than sunlight
above the atmosphere that scatters shorter wavelengths more than longer
wavelengths.
- Don Klipstein (d...@misty.com)
Don Klipstein
I appreciate this story!
If I had to work 8-plus hours per day 5 or more days per week with
illumination being bandwidth-restricted to monochromatic versions of
yellow to red, I might find "black-and-yellow" or "black-and-red"
"monochromatic vision" to be "effectively black-and-white". Though I
suspect that at Hour 9 I will still realize and know every minute that the
illuimination is not white but yellow/red/whatever.
>>>T'is a r'markable instrument, the human eye!
>Doesn't sound like your's are as remarkable as most folks. ;-)
>Which may be advantageous in your line of work
I suspect advantageous to me - though I suspect that I differ from
others in the area where the eyes connect to - the brain. I like to sense
when my eyeballs do "white balancing $#!+".
- Don Klipstein (d...@misty.com)
Don Klipstein
Sunlight above Earth's atmosphere is well enough known to be described
by color temperature in the upper 5700's to around 5800 K.
Direct sunlight at Earth surface appears to me to have correlated color
temperature lower, and to have chromaticity more green than "blackbody
locus" by merely a few thousandths of a "distance unit on 1931 CIE
chromatiocity diageram", to as greenish as around ".015 distance unit",
when the sun is merely several degrees above the horizon and has
correlateed color temp. around 3,000 or in the mid-upper 2,000's Kelvin.
- Don Klipstein (d...@misty.com)
Kennedy McEwen
<d...@manx.misty.com> writes
> If I had to work 8-plus hours per day 5 or more days per week with
>illumination being bandwidth-restricted to monochromatic versions of
>yellow to red, I might find "black-and-yellow" or "black-and-red"
>"monochromatic vision" to be "effectively black-and-white". Though I
>suspect that at Hour 9 I will still realize and know every minute that the
>illuimination is not white but yellow/red/whatever.
>
towards the illumination colour within seconds. That is why some
optical illusions work so well.
eg. the Inverted Flag illusion that you can find all over the net with
many different national flags,
the black and white Spanish Castle colour photo
http://www.johnsadowski.com/big_spanish_castle.php
or the Lilac Chaser illusion
http://en.wikipedia.org/wiki/Lilac_chaser
They work because the cones in your eye, which provide colour vision,
tire of the same constant stimulus and your view eventually nulls out to
grey. Remove the stimulus and you get the reverse colour image for a
while.
These illusions are just stimulating part of your eye but the effect is
just the same when all of your retina just gets one colour - you just
see a black and white image after a while, but get a flash of reverse
colour when the white light comes on.
>>>>T'is a r'markable instrument, the human eye!
>>Doesn't sound like your's are as remarkable as most folks. ;-)
>>Which may be advantageous in your line of work
>
> I suspect advantageous to me - though I suspect that I differ from
>others in the area where the eyes connect to - the brain. I like to sense
>when my eyeballs do "white balancing $#!+".
>
You can't tell when your eyes are doing the white balancing - it just
happens. I presume you can't experience any of these famous illusions.
Don Klipstein
I did look at the castle one - very impressive. And I am aware
of negative afterimages. However, it appears to me that eyes only
partially white-balance.
I have before spent hours at a time in a room illuminated by nothing but
low pressure sodium light, and it still looked to me yellow rather than
white. And when my eyes adapt to pure or nearly pure red light (free of
wavelengths shorter than reddish orange or orange), the shift I see is
changing to orange. Sometimes orangish yellow after getting exposed
enough to red that has some significant orange spectral content - but I
don't see white.
- Don Klipstein (d...@misty.com)
Kennedy McEwen
<d...@manx.misty.com> writes
>
> I did look at the castle one - very impressive. And I am aware
>of negative afterimages. However, it appears to me that eyes only
>partially white-balance.
>
image of the castle, complete with white clouds on a bright blue sky and
green grass, when the monochrome image pops up.
Don Klipstein
>In article <slrngr40c...@manx.misty.com>, Don Klipstein
><d...@manx.misty.com> writes
>>
>> I did look at the castle one - very impressive. And I am aware
>>of negative afterimages. However, it appears to me that eyes only
>>partially white-balance.
>>
>If it was only partial balancing then you wouldn't get a full colour
>image of the castle, complete with white clouds on a bright blue sky and
>green grass, when the monochrome image pops up.
It appears to me that 1 of 2 mutually exclusive possibilities is true,
and I do not know which one applies here:
1. The pre-exposure has its hues having saturation greater than those in
the desired image, to achieve the overcorrection necessitated by white
balancing of human vision being only partial.
2. The negative afterimage in the eye resulting from the pre-exposure
causes color sensation less than full but sufficient to make a spectacular
impression despite hues in the pre-exposure having saturation only same
(in opposite hue direction) as that of the "desired colors".
Keep in mind that if I spend a few hours somewhere illuminated by
nothing but incandescent light of color temp. around 2800K, I do not find
that light white but somewhat yellowish. However, my vision does
white-balance in response to ~2800 K incandescent lighting or similar
color CFL lighting even for hours to an extent where ~3400K appears "pure
white".
- Don Klipstein (d...@misty.com)
Kennedy McEwen
>In article <6D+0vuAb...@kennedym.demon.co.uk>, Kennedy McEwen wrote:
>>In article <slrngr40c...@manx.misty.com>, Don Klipstein
>><d...@manx.misty.com> writes
>>>
>>> I did look at the castle one - very impressive. And I am aware
>>>of negative afterimages. However, it appears to me that eyes only
>>>partially white-balance.
>>>
>>If it was only partial balancing then you wouldn't get a full colour
>>image of the castle, complete with white clouds on a bright blue sky and
>>green grass, when the monochrome image pops up.
>
> It appears to me that 1 of 2 mutually exclusive possibilities is true,
>and I do not know which one applies here:
>
>1. The pre-exposure has its hues having saturation greater than those in
>the desired image, to achieve the overcorrection necessitated by white
>balancing of human vision being only partial.
>
>2. The negative afterimage in the eye resulting from the pre-exposure
>causes color sensation less than full but sufficient to make a spectacular
>impression despite hues in the pre-exposure having saturation only same
>(in opposite hue direction) as that of the "desired colors".
>
So your brain, having been swamped by stimulation of one sensor type to
the point of disregarding it, suddenly notices the change. Which is
exactly the common explanation of the effect and assumes that something
in the eye-brain combination tires of stimulus, making the ye see grey
even when it is red, or yellow, or green or blue or....
People hear in perfect pitch because of the way the ear works. The eye
doesn't work the same way, and people who claim to see in perfect pitch
are <polite on> kidding themselves <polite off>.
Alan Browne
> This proves your perception wrong. The sun is neither white nor yellow. You
> have lived under a bright green star all your life so you perceive green
> shifted light as white, proving the other person's point.
Hardly.
Alan Browne
No. Read carefully what I wrote above - it was not about amount of
exposure but about making an exposure in sunlight where the blue from
the sky would have no effect on the daylight lit portion of the image.
IOW: If you are exposing something in direct sunlight, the 'blue' fill
from the sky will only reveal itself in the shadows. You will not see
it in the portions of the image that are in direct sunlight.
This is precisely why for "white balance" reasons we _would_ add a
warming filter for portraits shot in open shade but not in open sunlight
for daylight balanced films.
With digital cameras I just set 5500K for daylight and 7500K for open shade.
Alan Browne
> In article <o_Odna3vF8fpuy3U...@giganews.com>, Alan Browne wrote:
>> Don Klipstein wrote:
>>
>>> Daylight balanced color slide film is supposed to be balanced for
>>> "average"/"usual" combinations of direct sunlight and whatever-is-usual
>>> fraction of the light from clear blue sky (not all of it).
>> The ratio of blue fill to sunlight is about 1:16. Virtually nothing for
>> a sunlight exposure.
>
> Just to toss in some numbers, I will figure out what one gets with 1
> part of 12,000 K (my estimate of "average blue sky") plus 16 parts of 4600
> K (my estimate of "average midday sunlight at 39.9-40 degrees latitude").
>
> Result of my calculations here is 4825 K - 225 K more blue than direct
> sunlight.
+/-225K around "sunlight" is pretty much meaningless in all but the most
color critical (eg: advertising of color specific merchandize; some
scientific work) photography.
> I try again for 1 part 15,000 K plus 16 parts 4800 K. I figure there
> about 5055-5060 K, 255-260 K more blue than "sunlight alone".
>
> I try once more for one part 2,000,000 K plus 16 parts 5200 K. All
> 16-to-1 ratios I am here considering to be photometric ratios as opposed
> to radiometric.
>
> The extreme of 1 part 2 million K to 16 parts 5200 K with 1:16 ratio
> being photometric, I calculate to correlated color temp. around 5575 K.
> I may decide to not argue against as high as 5600. The light from the
> blue sky in this extreme case boosts color temp. of direct sunlight by
> 375-400 K.
Still pretty meaningless in the sense of my para above. Shown an
individual photo there are few people that would notice it was "bluer"
than "nominal". (Which is not the same as a side-by-side comparison).
> It does appear to me, based on literature that I have exposed myself to,
> that sunlight alone achieves 5200 K for high extreme achievement at
> Washington DC. The sun gets to 15.5-16 degrees from zenith there at "high
> noon" in mid/late June. And I suspect that the light from blue sky on
> such a clear day, overall from the sky, is more like 20,000 or 17,000 K
> rather than pushing infinity. However, I find correlated color temp.
> using this extreme combo (1 part 20,000 K plus 16 parts 5200 K with 1/16
> ratio being photometric) to be around 5525 K or so. Light from blue sky
> boosts correlated color temp. from that of sunlight by 325, maybe 330 K.
At the certain risk of repeating myself...
> I suspect that color temp. boost from light from blue sky adding to
> direct sunlight is maximized when the sun is low, dimmed, and having low
> correlated color temperature below 2,000 K.
Color temps below 4500K or so begin to be "warm" (emotional warm as
opposed to K warm) looking in a photo. Incandescent lighting is about
3800L, candles are down around 2000K or so.
Thanks for the numbers above. Quite interesting.
Peter Irwin
>> sunlight.
>
> +/-225K around "sunlight" is pretty much meaningless in all but the most
> color critical (eg: advertising of color specific merchandize; some
> scientific work) photography.
>
to convert the Kelvin values to MIREDs. MIRED means "million reciprocal
degrees" and is obtained by dividing 1 000 000 by the Kelvin colour
temperature. The advantage of this system is that a difference of
a certain number of Mireds is subjectively comparable and requires
the same strength of colour conversion filter.
4600K is 217 mireds
4825K is 207 mireds - difference 10 mireds (very slight)
compare the difference between a 3200K lamp and 3400K photoflood:
3200K is 312.5 mireds
3400K is 294 mireds - difference 18 1/2 mireds (slight but enough
for a conversion filter to make a noticeable
difference on slide film.)
Peter.
--
pir...@ktb.net
Don Klipstein
>In article <slrngr8lq...@manx.misty.com>, Don Klipstein
><d...@manx.misty.com> writes
>>In article <6D+0vuAb...@kennedym.demon.co.uk>, Kennedy McEwen wrote:
>>>In article <slrngr40c...@manx.misty.com>, Don Klipstein
>>><d...@manx.misty.com> writes
>>>>
>>>> I did look at the castle one - very impressive. And I am aware
>>>>of negative afterimages. However, it appears to me that eyes only
>>>>partially white-balance.
>>>>
>>>If it was only partial balancing then you wouldn't get a full colour
>>>image of the castle, complete with white clouds on a bright blue sky and
>>>green grass, when the monochrome image pops up.
>>
>> It appears to me that 1 of 2 mutually exclusive possibilities is true,
>>and I do not know which one applies here:
>>
>>1. The pre-exposure has its hues having saturation greater than those in
>>the desired image, to achieve the overcorrection necessitated by white
>>balancing of human vision being only partial.
>>
>which would be impossible in a normal intensity image.
I dispute that in the case where desired saturation is short of 100%,
which I suspect to be true with this castle scene.
>>2. The negative afterimage in the eye resulting from the pre-exposure
>>causes color sensation less than full but sufficient to make a spectacular
>>impression despite hues in the pre-exposure having saturation only same
>>(in opposite hue direction) as that of the "desired colors".
>>
>So your brain, having been swamped by stimulation of one sensor type to
>the point of disregarding it, suddenly notices the change. Which is
>exactly the common explanation of the effect and assumes that something
>in the eye-brain combination tires of stimulus, making the ye see grey
>even when it is red, or yellow, or green or blue or....
I never had red or low pressure sodium yellow or blue-LED-blue appearing
to me white or gray, even after hours with one of these being sole source
of illumination. Even when cooped up in a house illuminated by nothing
but incandescents or incandescent-color CFLs for 6-plus hours after dark,
such warm color lighting does not appear to me to be white, but yellowish
white.
>People hear in perfect pitch because of the way the ear works. The eye
>doesn't work the same way, and people who claim to see in perfect pitch
>are <polite on> kidding themselves <polite off>.
When in a room illuminated by low pressure sodium light for a few hours,
I see the 589-589.6 nm very orangish yellow light as different from how it
appeared initially, but far short of white/gray. Its apearance to me
changes to a less orangish shade of yellow. With red light, the
appearance to me merely shifts from red to orange. Light from blue or
green LEDs merely shifts appearance to me to a lighter shade of blue if
that is the only light I see for hours. Yes, I have been there!
- Don Klipstein (d...@misty.com)
Alan Browne
> In rec.photo.digital Alan Browne <alan....@freelunchvideotron.ca> wrote:
>> Don Klipstein wrote:
>>> Result of my calculations here is 4825 K - 225 K more blue than direct
>>> sunlight.
>> +/-225K around "sunlight" is pretty much meaningless in all but the most
>> color critical (eg: advertising of color specific merchandize; some
>> scientific work) photography.
>>
> If you are going to compare colour temperatures, it really helps
> to convert the Kelvin values to MIREDs. MIRED means "million reciprocal
> degrees" and is obtained by dividing 1 000 000 by the Kelvin colour
> temperature. The advantage of this system is that a difference of
> a certain number of Mireds is subjectively comparable and requires
> the same strength of colour conversion filter.
I have nothing against Mireds but for this discussion staying in the
base unit seems more useful.
> 4600K is 217 mireds
> 4825K is 207 mireds - difference 10 mireds (very slight)
Exactly. Most people will not see the difference unless in a side by
side comparison - (which I noted in my prev. posting).
> compare the difference between a 3200K lamp and 3400K photoflood:
>
> 3200K is 312.5 mireds
> 3400K is 294 mireds - difference 18 1/2 mireds (slight but enough
> for a conversion filter to make a noticeable
> difference on slide film.)
Again, viewing an individual image, nobody would be able to say, "this
is cool/warm" v. some nominal expectation.
Side by side, one would notice.
Doug
big snip
>
> When in a room illuminated by low pressure sodium light for a few hours,
> I see the 589-589.6 nm very orangish yellow light as different from how it
> appeared initially, but far short of white/gray. Its appearance to me
> changes to a less orangish shade of yellow. With red light, the
> appearance to me merely shifts from red to orange. Light from blue or
> green LEDs merely shifts appearance to me to a lighter shade of blue if
> that is the only light I see for hours. Yes, I have been there!
>
> - Don Klipstein (d...@misty.com)
I'm a lurker on this list, but have learned a great deal here. I also have
an interest in photography and lighting in general. I'm wondering if the
your perception of the color of white light never rebalances because you
have taught yourself to automatically be aware of the color of the light
and not because there is something "different" about your vision.
-Doug