One upmanship and Canon's claim

On Sun, 8 Jul 2007 07:44:22 -0400, Rita Ä Berkowitz ritaberk2O04 @aol.com
wrote:
: RichA wrote:
:
: Somehow I suspect Mark is over dramatizing a bit. If Mark carries
: himself properly he will have no problems from the local "thugs"
: and "gangsters" that fill the wild streets. I know Mark will have
: a very safe and enjoyable time.
:
: I know it's hard to understand if you live in a nice, upscale
: community filled with "progressive liberals" who love all mankind,
: but at least TRY to stay current.
:
: http://allafrica.com/stories/200707040216.html
:
: When you consider incidents on a population/incident ratio the crime
: level is much lower than it is in New York, New Orleans, or DC. Mark's
: a big boy and he'll be fine.

Have you checked the figures lately? I'm not sure New York belongs on that
list anymore.

Bob

Roger N. Clark (change username to rnclark) wrote:
John McWilliams wrote:
Rebecca Ore wrote:
In article ,
"G.T." wrote:

This is why religiously following "18-month" rule eliminates
technical and
financial hardships. It's totally and utterly foolish to keep any
dSLR body
past its useful life expectancy of 18-months.

Very rich, very very rich in more ways than one.

It may be that beyond 18 months, the older camera body has almost no
resale value. I was re-reading a photography magazine from 1999 when
2.1 MP was considered state of the art and the DSLR was a Nikon
F5/Kodak hybrid.

That was true, but as the slope of the curve has changed, a top camera
bought today, or even two years ago, will hold significant value.

I agree. I got my 1D Mark II in summer 2004, sometime after
it was first introduced. It still works great, and at the moment
better then 1D3's, and still has over half its original value,
over 3 years later. The market has matured.

Roger

This is a good news bad news kind of thing. We bought a 20D 2 1/2 years
ago and it is still a very capable camera and holds up very well to the
30D. So it is nice that the camera is not totally obsolete. The down
side is that I can’t buy a FF digital for $1500. In the early years the
resolution of the cameras I bought went like this, 0.3 MP 1.2 MP 3.2MP
and 8MP. Each time going up by a factor of well over 2, that trend
ended in 2004, if it had not I would be shooting with a 24MP full frame
camera right now and I would have paid about $1000 for it.

At least flash memory has continued to fall at a good rate, and even
more important hard disk space has gotten way cheaper, otherwise I would
be spending more for my disk space then on my photo gear.

Scott

On Fri, 6 Jul 2007 23:16:46 -0700, "Mark²" mjmorgan(lowest even number
wrote:
: Peter A. Stavrakoglou wrote:
: It's not advisable to use canned air to clear dust out of a DSLR,
: especially on the sensor.
:
: The only problem with canned air is when you tip the can ...

And so Mark passes up another opportunity to tell us how he made out with his
1D3 at Canon's Irvine shop last week. I'm encouraged by that. Mark wasn't at
all reticent to say what he thought of the 1D3's autofocus problem and Canon's
initially passive reaction to it. His silence now suggests to me that the
folks at Irvine found a way to ensure that the AF problem won't spoil his
Africa trip. They would almost certainly have asked him not to advertise it
widely, lest all 1D3 owners start beating down their door.

Bob

Scott W wrote:
Roger N. Clark (change username to rnclark) wrote:
I agree. I got my 1D Mark II in summer 2004, sometime after
it was first introduced. It still works great, and at the moment
better then 1D3's, and still has over half its original value,
over 3 years later. The market has matured.

Roger

This is a good news bad news kind of thing. We bought a 20D 2 1/2 years
ago and it is still a very capable camera and holds up very well to the
30D. So it is nice that the camera is not totally obsolete. The down
side is that I can’t buy a FF digital for $1500. In the early years the
resolution of the cameras I bought went like this, 0.3 MP 1.2 MP 3.2MP
and 8MP. Each time going up by a factor of well over 2, that trend
ended in 2004, if it had not I would be shooting with a 24MP full frame
camera right now and I would have paid about $1000 for it.

I agree. I think there are two things limiting full frame
prices: 1)no competition (Canon is it), and 2) cost of a wafer
run is still very high. I remember seeing a figure that a
12-inch wafer run costs about $10,000 (if someone knows a better
number, please let me know). So how many full-frame sensors fit
on a 12-diameter disk (not many). Thus cost per sensor remains high.
From one news report I saw, Canon refined the process to get
far fewer defects per wafer, thus increasing yield. That allowed
the 5D price to be so low.

At least flash memory has continued to fall at a good rate, and even
more important hard disk space has gotten way cheaper, otherwise I would
be spending more for my disk space then on my photo gear.

Technology marches fast until a physical limit is reached.
For a while we were stuck max 500 GByte drives. That was when
the magnetic domain was spread on the surface of a disk
like a pancake. Then a couple of years ago, the write
vertically through the disk was refined (a process I read
about 10 years ago), so now disk drives can go up in size
about a factor of 10 or more. I just bought a 1 terabyte
USB drive that contains a single 3.5-inch form-factor drive
($440 at Best Buy), a Hitachi (formerly IBM). They will keep
dribbling out larger and larger drives, and we should see
5 terabyte drives in a few years.

But once a physical limit is reached, there is little room for
improvement. With electronic sensors, they are so good,
with such high quantum efficiency, there is little room for
improvement. We need a new technology, e.g. one that converts
each photon to a known energy so we know its precise wavelength.
Then we can do away with Bayer filters (or Foveon absorbers)
and improve sensitivity 3 to 5x (and our cameras will be
imaging spectrometers too). Oh, and make read and thermal
noise zero.

Roger

On Sun, 08 Jul 2007 08:17:03 -0600, "Roger N. Clark (change username to
rnclark)" wrote:
: But once a physical limit is reached, there is little room for
: improvement. With electronic sensors, they are so good,
: with such high quantum efficiency, there is little room for
: improvement. We need a new technology, e.g. one that converts
: each photon to a known energy so we know its precise wavelength.
: Then we can do away with Bayer filters (or Foveon absorbers)
: and improve sensitivity 3 to 5x (and our cameras will be
: imaging spectrometers too). Oh, and make read and thermal
: noise zero.

IOW, you're hoping for someone to prove that Werner Heisenberg was a crackpot?
;^)

Bob

Roger N. Clark (change username to rnclark) wrote:
Scott W wrote:

Roger N. Clark (change username to rnclark) wrote:

I agree. I got my 1D Mark II in summer 2004, sometime after it
was first introduced. It still works great, and at the moment
better then 1D3's, and still has over half its original value,
over 3 years later. The market has matured.

Roger


This is a good news bad news kind of thing. We bought a 20D 2 1/2
years ago and it is still a very capable camera and holds up very
well to the 30D. So it is nice that the camera is not totally
obsolete. The down side is that I can’t buy a FF digital for $1500.
In the early years the resolution of the cameras I bought went
like this, 0.3 MP 1.2 MP 3.2MP and 8MP. Each time going up by a
factor of well over 2, that trend ended in 2004, if it had not I
would be shooting with a 24MP full frame camera right now and I
would have paid about $1000 for it.


I agree. I think there are two things limiting full frame prices:
1)no competition (Canon is it), and 2) cost of a wafer run is still
very high. I remember seeing a figure that a 12-inch wafer run costs
about $10,000 (if someone knows a better number, please let me
know). So how many full-frame sensors fit
on a 12-diameter disk (not many). Thus cost per sensor remains high.
From one news report I saw, Canon refined the process to get far
fewer defects per wafer, thus increasing yield. That allowed the 5D
price to be so low.

That $10,000 per is about right (but can be as low as a few $000 or 10X
higher depending on the process and desired result. But a wafer is just
a step ... next is cutting and packaging (albeit lower risks).

On a given wafer there will be randomly distributed defects. As the
sensors get larger the probability of a defect hit gets larger as well.
As sensors get smaller, more sensors "miss" defects. This is why the
P&S cameras are so damned cheap ... you get well over 1000 sensors per
wafer (I would guess yield at somewhere 'tween 1500 and 2000).

For a FF you would be lucky to get 50 ...before defects and packaging
steps (which include the RGB mask and possibly microlenses...)

(and our cameras will be imaging spectrometers too). Oh, and make
read and thermal noise zero.

Boltzman says no-can-do. You could always add heat removal but it's
unlikely that you can cool an entire portable camera sensor to milli-K
levels.

Cheers,
Alan

--
-- 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.

Robert Coe wrote:
On Fri, 6 Jul 2007 23:16:46 -0700, "Mark²" mjmorgan(lowest even
number wrote:
Peter A. Stavrakoglou wrote:
It's not advisable to use canned air to clear dust out of a DSLR,
especially on the sensor.

The only problem with canned air is when you tip the can ...

And so Mark passes up another opportunity to tell us how he made out
with his 1D3 at Canon's Irvine shop last week. I'm encouraged by
that. Mark wasn't at all reticent to say what he thought of the 1D3's
autofocus problem and Canon's initially passive reaction to it. His
silence now suggests to me that the folks at Irvine found a way to
ensure that the AF problem won't spoil his Africa trip. They would
almost certainly have asked him not to advertise it widely, lest all
1D3 owners start beating down their door.

Bob

You'e good. -Except that there are still problems. I've wanted to type it
all out, but it will take some doing... Long story. Very interesting,
though. I'll try to do it today...though I'm REALLY struggling to be ready
in time for my flght tomorrow...

--
Images (Plus Snaps & Grabs) by Mark² at:
www.pbase.com/markuson

Mark² mjmorgan@cox wrote:

Carrying a handgun would drop me right into an ex-soviet prison...

Be secure. Beware of other drivers. Invest into passive security.
Never have to worry about bad streets or mud holes or finding
no parking space. Never worry about the occasional firefight
in the neighbourhood again. Make sure noone will steal your
vehicle and escape unnoticed. Sleep asured, knowing that only
very professional thieves can enter your vehicle --- and that
the police will notice in time. Crush your competition completely.
Drive a tank.
APDS-ammo optional.

-Wolfgang

Robert Coe wrote:
On Sun, 08 Jul 2007 08:17:03 -0600, "Roger N. Clark (change username to
rnclark)" wrote:
: But once a physical limit is reached, there is little room for
: improvement. With electronic sensors, they are so good,
: with such high quantum efficiency, there is little room for
: improvement. We need a new technology, e.g. one that converts
: each photon to a known energy so we know its precise wavelength.
: Then we can do away with Bayer filters (or Foveon absorbers)
: and improve sensitivity 3 to 5x (and our cameras will be
: imaging spectrometers too). Oh, and make read and thermal
: noise zero.

IOW, you're hoping for someone to prove that Werner Heisenberg was a crackpot?
;^)

Bob
Not at all. ;-)

Alan Browne wrote:

(and our cameras will be imaging spectrometers too). Oh, and make
read and thermal noise zero.

Boltzman says no-can-do. You could always add heat removal but it's
unlikely that you can cool an entire portable camera sensor to milli-K
levels.

It need not violate any physics, and I should have said
effectively zero, not absolutely. Read noise in the
best consumer cameras is already under 4 electrons
at room temperature and reports are reportedly even lower
in the 1D Mark III. With more accurate 18-bit converters
and on-pixel electronics, 1 electron read noise is possible.
Thermal noise keeps getting better with each generation
sensor. Cooled CCDs reached the level of one electron/hour thermal
dark current a long time ago, and DSLR CMOS dark currents are
less than a thousand electrons/hour already at ambient temperatures
(and noise is square root dark current). These are great values
already, even without any improvements!
Line and pixel uniformity is improving too with each generation,
but is already outstanding. Most digital camera imaging
has noise limited by photon noise, not other sources, so even
with improvements, most images will see little difference.

Roger