Re: Digital vs 35mm camera optics
From: Jamie Carter (jacarter3_at_onebox.com)
Date: 08/18/04
- Next message: Ralph Sansbury: "Re: P10 Acceleration: Light Speed Doesn't Extrapolate"
- Previous message: David Littlewood: "Re: Digital vs 35mm camera optics"
- In reply to: Bob Monaghan: "Re: Digital vs 35mm camera optics"
- Next in thread: AES/newspost: "Re: Digital vs 35mm camera optics"
- Reply: AES/newspost: "Re: Digital vs 35mm camera optics"
- Messages sorted by: [ date ] [ thread ]
Date: 18 Aug 2004 07:41:10 -0700
This discussion has moved from the optical scientist to the realm of
the consumer. I worked at Kodak in the (very) early 80s when the Kodak
"disc" camera made its debut. Back then, Kodak still did a great deal
of film process and print work in high volume plants in house (not the
local supermarket process center business plan). They actually
analysed the content of thousands of home photos for foreground versus
background content, features sizes (spatial content), lighting and a
variety of other factors to weight the design of the aspheric
objective. The lens MTF characteristics were designed to enhance the
subjective enjoyment and perceived quality of the "typical" consumer
while making the optics as cheap as possible (Kodak never made money
on cameras; they were only for burning Kodak film, processing
chemicals and photographic print papers). Since the objective had such
a short depth of focus, the disc format was used to keep the emulsion
in the focal plane. The moral to this story is that the consumer
camera is designed for consumer use with the least cost.
I have never had any doubt that a 2.5 megapixel camera or higher was
grossly oversampling the resolution transfered by the digital camera
lens to the sensor. While the "number of pixels" and "bigger means
better" are things any consumer will gladly believe, the quality of
the image is very difficult for them to objectively assess or even
understand. If you look at anyone's digital photos, you will note that
they are dominated by many effects that degrade resolution including,
but not limited to, camera motion, depth of field, lighting, limited
dynamic range, and noise. To the consumer, if it looks good on the
computer monitor (which usually requires resampling to the limited
screen resolution), it's a great photo.
In other words, all those people buying 3, 4, and 5 megapixel cameras
with little, auto-focus lenses are fooling themselves. The
manufacturer knows these will be used for snap-shots and the quality
of the lens does not need to match the sampling resolution (which is
also affected by Bayer color filter arrays and anti-aliasing filters)
and it makes very little business sense to give up profit for superior
optics. If you don't believe me, check out the cameras that
professional photographers use. These aren't the consumer cameras at
Best Buy or Circuit City. They have interchangeable lenses and start
at prices well over $1000.
Personally, my "high end" digital camera has only 2.1 megapixels. I
chose it for the Pentax lens and small f/no. Subjectively, I can't
tell the difference between its pictures and those of my brother's 4
megapixel Cannon. One specification that is very difficult, or
impossible, to find on a consumer digital camera IS the lens
resolution (or any other spec. that goes beyond zoom ratio, focal
lengths, and aperture size). Many people believe the the lens
resolution must be equivalent to the sensor resolution and we
certainly don't want them to believe any differently.
Just my $0.02 worth.
James Carter
- Next message: Ralph Sansbury: "Re: P10 Acceleration: Light Speed Doesn't Extrapolate"
- Previous message: David Littlewood: "Re: Digital vs 35mm camera optics"
- In reply to: Bob Monaghan: "Re: Digital vs 35mm camera optics"
- Next in thread: AES/newspost: "Re: Digital vs 35mm camera optics"
- Reply: AES/newspost: "Re: Digital vs 35mm camera optics"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|