Interesting machinations at Kodak

Gordon Moat wrote ...

Thanks. I see you're certainly done some homework. I just wonder if your
sources are open-minded enough. As an example, I remember well the
distinguished researchers that just ten years ago claimed that cellular
phone penetration in western Europe would never exceed 25% due to usage cost
reasons. Now it is 65% in that specific country and over 80% in some.

I've seen some quite horrible market research studies recently, particularly
one on wireless messaging that covered 400 pages without even mentioning
competing technologies outside of the two over-hyped main ones that probably
were selected only to make the study sell well. Quite shocking if investors
let themselves be guided by such narrow-minded crap. (well in this case it
was hopefully obvious enough).

Some quick uninformed musings to get back on topic:
A digital 1,3 MP camera with USB today costs less than 50$. Last year I
guess you could only get a VGA (0,3 MP) for that price. Remarkable
progress. Still , film disposables seem to cost only 5$ or 10$ with flash.
So there's still a challenge for digital - I agree on that. But what is the
future price trend?

On the other hand, digital miniprinters cost only 150-300$ (no PC needed)
which I'd guess even a small village shop can bear. And they can provide
instant prints. Whereas the film disposables will always have to be sent
further away for processing. I wonder if there are studies about the price
premium that people are willing to pay for instant gratification?

Ten years is a very long time in the digital universe. So I'll bet on
digital. But as that Polaroid pack probably has an astronomical antique
value by then... shall we say a 25-pack of CD-ROMs instead?

See you in 2013 ;-)
/N

 

Is it a hardware limitation or merely a faulty processing algorithm
(firmware)? Both Canon and Kodak use 24mm x 36mm CMOS sensors...are there
two currently on the market?

 

A study on the silver market this week claimed that camera phones have
already this year passed the total number of digital cameras sold. If that's
true consumer digital cameras are dead now. The same study claimed film
sales in Asia have risen 10% so far this year.

Nick

 

Hardware is the root cause. One can determine this by
examining the sorts of artifacts that commonly appear
in 14n images, especially in the earlier samples. Blotchy
hair detail, for example. They're the sorts of artifacts
one can see when noise-reduction software has too much
shmootz to overcome. For examples, over-apply NeatImage
to a picture. As with unsharp masking, there are limits to the amount
of defects that noise-reduction software can fix; once you go
over those limits, the artifacts start appearing.

Yep. Canon's is home-grown, Kodak's is from a 3rd party. Somewhere
in a lab, Nikon's working furiously on one, and I imagine Sony and
Foveon are either in the race or thinking seriously about it.

stan

 

i've yet to see a camera phone that can beat a £25 webcam image let alone a
proper digital camera. Though from the adverts you'd think you had the
digital equivalent of a hassalblad.

 

Read somewhere that CMOS sensors were markedly inferior to CCD sensors. Is
there any truth in that? If so, why are the full frame sensors available
only CMOS?

Read somewhere that the predominant sensor array for the top flight medium
format digital backs was Kodak Blue Label (??). If so, what in the world
is Kodak doing with an inferior array for their own product.

Inquiring minds, an all that....

Bill Tallman

 

From: "William D. Tallman"

Several years ago this was true. Today all you have to do is compare the Canon
dSLR CMOS cameras' output to the competition to see it no longer holds up.

Here's Kodak's take on CMOS vs CCD sensors (they make both):

http://www.kodak.com/US/en/corp/researchDevelopment/technologyFeatures/cmo
s.shtml

Bill

 

There is a factor you guys may be missing. Computer technology price/
performance is improving so rapidly that developing countries may see
a jump over film right into digital. Right now the cost of a reasonable
computer is not much more than a basic SLR film camera. Even cell
phones have cameras these days, webcams can be had for <$20 US,
and the cost for digital photos blows film away, unless you print lots of
them. Why print them when computers are (will be) ubiquitous? Burn a
CD and give copies to your friends and family. The very same one hour
photo places can do digital, and there are more compelling reasons why
people will want computers anyway, such as education. What will it
look like in five to seven years? Only God knows for sure, but I will be
very, very surprised (maybe 'astonished' would be more accurate) if the
global film market is larger in 10 years.

 

Current technology brings these two much closer. There is still an edge with
CCD sensors, but at a cost.

Lower cost. Actually, with medium format digital backs, there are 36 mm square
sensors available as CCD, though they are very expensive. Some newer full
frame (almost) 645 sized sensors are coming out as CMOS, though at least one
CCD should be available soon. Much easier to change backs as technology
changes, unlike all in one digital SLRs.

With the all in one direct digital SLRs, like from Canon, the cost of
producing a CMOS sensor is lower. There is a microlens array in front of the
chip, to properly target the CMOS sensor sells, and this is not present on CCD
systems, but that extra bit still adds less cost than the overall cost of CCD
sensors.

Manufacturing yields will reflect costs. If manufacturing improves for either
type of chip, expect to see more of them, and a corresponding price reduction.
Probably sooner than later.

Kodak Professional digital gear often targets the photojournalist market.
Considering that newsprint has high dot gain, low total ink limits, and low
saturation limits, none of the Kodak gear is lacking. While that may be
humorous, be aware that Kodak has been very good at supporting these products,
and several updates have already been made available for their high end
digital gear.

The Kodak medium format digital backs are actually quite good. The costs are
still high, but they are quite competitive in that segment of the market.
Despite the high costs, medium format digital backs are about the only future
proof digital choices. Consider that many of these are available on lease and
upgrade plans, and the cost may be more reasonable. I doubt many amateurs
really get these. Renting is another option, and I have tried a few of these.
The image quality is noticeably better than all in one 35 mm size direct
digital SLRs.

Ciao!

Gordon Moat
Alliance Graphique Studio
<http://www.allgstudio.com>

 

Consider that electrical power distribution is still not wide spread enough to
service devices needing recharging or plug-in. A one time use camera, and that once
a month trip to the city, could capture some moments.

Industry computer statistics (VARBusiness, and CRN) indicate that the average home
computer purchase, software, hardware, peripherals, etc. is about $1800. While
within reach of many of us, this is still not low cost. Even lowering the costs for
developing countries still makes computers a luxury/business/government item.

The same was said of magazines when the internet took off in 1995. You probably
remember the "paperless office" as well. Currently, there are more magazines in
publication, and paper usage indicates that even more printed matter is being
generated. However, your statement does indicate the dominant usage of digital
camera buyers, in that few print anything (11% indicated by PMAI).

Definitely a well thought out reply. You are welcome to join the bet, and e-mail me
in ten years. I still get the feeling I will be getting lots of Polaroid 665 P/N in
ten years.

I am sure the computer industry would like emerging markets to boom in computer
sales. Unfortunately, some analysts, and futurists, see less of a need for
computers, especially in the home. There is an expected growth in small portable
devices that do more, especially smarter phone technology, or handhelds that work
better with phones. Development of better battery technology limits the release of
some devices, but we should see big changes in ten years.

Ciao!

Gordon Moat
Alliance Graphique Studio
<http://www.allgstudio.com>

 

I remember some of those projections as well. I usually have a minimum of three
sources prior to forming a pattern of projection. With some things, I have many
more times sources than that. the only way to lessen bias, is to increase the
sampling.

Investors largely look at quarterly performance, and often ignore longer term
research. Too much of "buy on the rumour, sell on the news" drive the stock
market. There is more money in potential, than in tangible items. Anyway, it is
somewhat high risk for a company to pursue directions that will show profits
many years down the road. They do that, but largely keep that information to
limited distribution.

Don't forget battery life, or the ability to recharge a device. Weatherproofing
is a side issue, but this is getting better.

How fast are they? Can they generate profits fast enough? Consider that some
mini lab gear might be available at low, or no cost, just to begin exploiting a
market. The company providing that gear makes money on supplies, like paper,
chemicals, inks, etc. Xerox already offers a business model like this for their
high end printers, so the cost to the business is only monthly expenses on
supplies. Xerox makes out on this, since the contract requires supplies be
purchased from them. AGFA, Kodak, and Fuji could easily do this, if they are
not doing it already.

Largely a Western phenomenon. This is more cultural, and to properly do
business in non Western markets, it is very important to consider the cultures.
Treat them like Westerners, and failure may be more likely.

Who the hell is going to be using CDs in 10 years? Even DVDs may be obsolete
then. Have you heard of fluorescent disk technology, or solid state cards?
Fewer moving parts for future technology, and a greater difficulty in copying.
CDs will only be useful for Xmas ornaments, or drink coasters, like the current
AOL CDs.

Anyway, I will stick to the Polaroid 665 P/N. Likely that Fuji will have
acquired Polaroid by then, so I will accept a similar product from Fuji.

See you then.

Ciao!

Gordon Moat
Alliance Graphique Studio
<http://www.allgstudio.com>

 

Interesting. I did not know they were selling that fast. I may be off on predicting
those devices to take over in two to three years. Perhaps within a year, small
digital P&S sales might be a niche market.

I think film sales increases are more likely a result of increased travel. In the
last couple years, drops, or stability of film sales has been more tied to travel
and holiday usage. A good example of this is high film sales in New York in
September these last two years.

Of course, the greatest film sales is often one time use cameras, so it hardly
reflects usage of anyone on this news group. I cannot remember ever buying one of
those.

Ciao!

Gordon Moat
Alliance Graphique Studio
<http://www.allgstudio.com>

 

One thing for sure about ten years from now:

George Preddy will still be making Foveon posts.

 

The link is broken, but your point is made. Thanks!

Bill Tallman

 

Here's Kodak's take on CMOS vs CCD sensors (they make both):

Probably just need to cut-and-paste since it overflowed to the next line ...
try this one, I think it's short enough to not wrap to the next line ...

www.kodak.com/US/en/corp/researchDevelopment/technologyFeatures/cmos.shtml

 

The CMOS units have variation in gain, offset, and noise for each pixel.
You can still make CCDs even worse though.

Kodak had a 24x36 CCD in their KAF series I think, in the mid 90s or
so. It may have not been available with anti-blooming though.

--
Paul Repacholi 1 Crescent Rd.,
+61 (08) 9257-1001 Kalamunda.
West Australia 6076
comp.os.vms,- The Older, Grumpier Slashdot
Raw, Cooked or Well-done, it's all half baked.
EPIC, The Architecture of the future, always has been, always will be.

 

Heh, yep I was blind... This works. Kodak suggests (in 2001) that CMOS has
a ways to go. But now the 1Ds and the Kodak counterpart are running CMOS!?
It has got to have come a long way in a very short time, I think.

Thanks for the link!

Bill Tallman

 

Gordon Moat wrote:

<snip>

Hey, thanks, Gordon!! That was as lucid an explanation as I've run across
recently.

So, as the full frame DSLRs are CMOS and that technology is based on a more
robust manufacturing tech, it does seem reasonable that as that latter tech
improves and yields increase, we may see a 1Ds replacement at a lower
price, presumably with even better specs. This is very good news for us
guys that bought into EOS for that reason.

Good job!

Bill Tallman

 

is the cmos sensor what is being using in the newer flatbed scanners
like the Epson 1640? --Loren