Re: [photo-3d] Re: Halogen vs Florescence

[Paul Talbot <ptww@xxxxxxxxx>]

Ray Moxom quoted and wrote:
> 
> > The light from a tungsten halogen bulbs looks more natural.  [drt]
> 
> That is because it is more natural.
> 
> Even when the colour temperature of a tungsten halogen bulb and a florescent
> lamp are the same, the tungsten halogen bulb has a continuous spectrum of
> light while the florescent lamp tends to have a number of distinct
> wavelengths of light. This means that the florescent lamp may miss some
> portions of the spectrum. The tungsten halogen bulb therefore has a better
> colour rendering index (CRI) than the florescent lamp.
> 
> CRI is a unit of measure that defines how well colours are rendered by
> different illumination conditions in comparison to a standard such as a
> thermal radiator or daylight. The tungsten halogen bulb is a thermal
> radiator

I believe it is inappropriate to discuss all fluorescent light
sources as though they are the same.  There is a very wide variety
of color temperatures and CRI values of commonly available fluorescent
tubes.  See http://retirees.uwaterloo.ca/~jerry/orchids/tubes.html for
a list of values for many tubes from major makers.

There is a new generation of fluorescent lamp known as "full spectrum"
(goes beyond "wide spectrum") that feature extremely high CRI values.
While a standard fluorescent may have a CRI of about 48 to 70, there
are a number of full spectrum tubes with CRI of 90 and color temperature
of 5000K.  See, for example, the GE "Sunshine" tubes.

Other manufacturers may be pushing the envelope even farther.  Here
is a site that is offering a 5000K fluorescent tube with claimed 98
CRI!!  See http://www.budgetlighting.com/fs/fs.html

I don't have the CRI values for the "CCFL" untra-thin light
tables that I mentioned, but they are also color-corrected
lights.  They should not be lumped together with non-corrected
fluorescent light sources when discussing color accuracy.

I think the term "natural" in this discussion is being used in
more than one way, which is serving to confuse matters further.
In fact, GE uses the term "natural" to describe their fluorescent
lamps that are in the 3000 K temperature range, *not* for their
"sunshine" 5000 K tubes.  But their color strip samples under the
"natural" light look distinctly red-shifted to me.  "Natural" is
not the same as "accurate compared to bright daytime sunlight."
"Natural" light can come from many different sources and have
vastly different characteristics; as a standard for accuracy
it is meaningless.  An object viewed in the light of a forest
fire started by lightning would be viewed by a "natural" light
source, but would look nothing like it would look under daytime
sunlight.

Certainly many people, including DrT, may find the warm balance
of a 3000 K light source more pleasing, but the thread was started
by someone seeking a light source for color accuracy.  If the
standard for accuracy is daytime sunlight (the most logical choice,
IMO), then of course daytime sunlight itself is far and away the
best light source to use.  My non-expert opinion is that when an
artificial light source must be substituted, color-corrected
fluorescent sources with high (90+) CRI values, such as lights
used for professional-grade light tables, the new CCFL light
panels, and the full-spectrum light tubes are closer to daytime
sunlight than are most incandescent light sources, which are
much more red/orange than daytime sunlight.

Paul Talbot