New DX Sensor

[Jay Turberville]

The astonishing sharpness of Foveon sensor images is another
possibility. I hope that technology does not stall.

I would suspect a major challenge is getting dyes with the right
absorption spectra to adequately sense R, G or B, but not absorb
outside of those color ranges. Of course, the absorption ranges of
molecular dyes are not perfectly sharp, vary with temperature, and
can be affected by two photon processes, etc.

The Foveon sensor does not use dyes to distinguish colors in its R,G and B layers. They use the sensing depth of silicon. Different light frequencies tend to penentrate to different depths. Dyes might be used, however, with a prefilter that could be used to help fine-tune the response curves.

--
Jay Turberville
http://www.jayandwanda.com

 

[Julia Borg]

depends how you understand similarities. For example, trick a raw
converter into beleiving 5D file is 20D, and compare results with
processing same file as 5D file.

So you're saying that if two camera makers use identical sensors, but
converts to "raw data" in a different way than the other, then the two
sensor technologies are different?

no, I'm saying coffee and tea are both drinks, water is major component in both.

FYI, camera makers have little to no influence on many third-party converters. But when you start tuning and optimizing converter to process raw files from a given camera to the best you can - oh, you start to see those differencies...

--
Julia

 

[Julia Borg]

give me some numbers to support your suggestion of technology
similarities.
Namely, ratios of following parameters: pixel size, well capacity,
electronic gain, readout noise, thermal signal.

Scaling the same technology doesn't mean linear scaling of their
properties

oh yes. but the scaling should follow a pattern. say if pixel size is increased well capacity should be proportional to the increase of surface - for same technology?

--
Julia

 

[Miguel Furlock]

Hi Alan,

yes. I'm to sensing that there is a stubborn struggling for squared tools, something you and me and others will have observed in many other contemporary situations and matters. Architecture for example or monitors. That comes mostly from education towards the boxed and specialized knowledge and the need of humanity to get things together. It's a question of Lego, not of pixels.

And as our houses and lives are totally angled, the final product must have 90 degrees angles, wether it ends up against a wall, hanging or just looking from some harddisk to the screen.

Nethertheless, there is a point. With a digital back and 38 Megapixels there are no big heat problems and it is supposed that the upper frontier of these chips is at 120 Megapixels. These backs are not big, it's easy to see them evolving towards being attached to SLR's.

In fact the idea of having the chip inside a chasis is coming from the old idea of having a film in a dark room. This is typical in technological evolution. People get better instruments, but they take them back to the cove. They don't take things out of the cove and start to build up something totally new with the evolution gathered. Normally they don´t.

Have a look: All what we are now using comes directly from the past with only minor changes. Film has been substituted by sensor technology. What film roll or cassette filled, now is used for electronics.

But today you will not know where is the real difference. Machines still look nearly exactly the same as 20 years ago (a little bit rounder...). Have a look at the Alpha, with the Digiback on. Seeing this I believe that a very good camera in the next future will work with sensor and processing away from the machine, detachable. And then I hope that we can forget about the need to use two hands and a tiny grip when we use our tools.

There are so many other, a hundred times more evolved ways of construction a camera now that we got rid of the darkroom and chemicals and ... there is a real chance that someone brings up a totally new concept.

I would love to construct something like that. Far away from squares and much more ergonomic to people than the bodies we use actually.

Much closer to light and much more comfortable in shadows.

Who knows. Maybe some day I'll do? Must be funny to visit dpreview then and see how Phil would test it.

See you Alan. And congrats to your expo, I liked very much the framing you have done to the nature series. Very elegant.

Best wishes,

Miguel Furlock
http://furlock.com

Moving as smooth as possible on a thin skin of ice protecting me from myself

 

[Miguel Furlock]

--
Moving as smooth as possible on a thin skin of ice protecting me from myself

 

[Lee Jay]

give me some numbers to support your suggestion of technology
similarities.
Namely, ratios of following parameters: pixel size, well capacity,
electronic gain, readout noise, thermal signal.

Scaling the same technology doesn't mean linear scaling of their
properties

oh yes. but the scaling should follow a pattern. say if pixel size
is increased well capacity should be proportional to the increase
of surface - for same technology?

But not necessarily pixel spacing. They might be built on the same process technology which means some parts of the sensor do not scale with pixel area. The amps, for example, might be a smaller fraction of the pixel area allowing more room for the well.

--
Lee Jay
(see profile for equipment)

 

[Kevin Morris]

Gotta wade in here, sorry. I guess the part I don't understand is
when Nikon says "You have 2 choices: buy DX or pi$$ off," and you
think of that as a choice. Maybe it is a choice according to some
obscure definition in a dictionary, but it's not REALLY a choice,
is it. Certainly not in the sense that I understand the word.

What other sense of the word is there?

The obvious one. "I have product A and product B, which do you prefer?" That's a choice. As opposed to "Buy this product or I'll shoot you." Is not really a choice. Nikon's current offerings amount to "Buy product A or pi$$ off!" Again, not really a choice.

Did you not have a choice when you bought nikon?

No! I was looking to a major system upgrade and fell in love with the F4. I had no choice but to buy it.

Have you no choice in your next purchase?

A difficult question. Since investing in the Nikon system I have made a significant investment in lenses and accessories. Nevertheless, I was seriously considering the 5D when this new hint of a Nikon FF started to appear. I can afford to wait another year or so, so maybe we'll know more by then.

You fanboy, zealot types are really something else.

This comment is completely uncalled for.

That's your opinion. I disagree. What is uncalled for is the
canon fanboys that won't quit posting that drivel here. The
constant canon superiority for this or that, fanboy evangelism, is
what is uncalled for.

I agree with your sentiment, especially when people say things like "Just wait 'til Nikon delivers FF, then we'll gloat." Those statements are just childish. But I have no problem with Canon users joining these debates. I find I learn the most from the people I disagree with.

You seem to be taking some
people's desire for FF to be a personal affront. Those of us that
want FF don't expect it to be a Nikon exclusive.

That's ad hominem nonsense.

You've stumped me here. This seems to be a unique interpretation of "ad hominem".

I couldn't care less, if nikon makes a FF or not and whether you
want one or not.

Then why do you join in in all the FF threads and often belittle the people who do want FF.

On a personal note: I get a kick out your posts Kerry. Although we argue a lot, I can't help but think that if we were neighbours we would be good friends.

--
Tuktu Sijuktei
'Please tell me if the lens cap is on.'

 

[Julia Borg]

The amps, for example, might be a smaller
fraction of the pixel area allowing more room for the well.

in light of the above, please explain your statements "5D's sensor is 256% the size of the one in the 20D. This means at the same f-stop, it captures 2.56x as much light" and "Well capacity is around 50000e for 20D and around 85000 for 5D"

--
Julia

 

[Lee Jay]

That's what this is about? The difference between 82000 and "about 85000"? You should try entering the real world once in a while.
--
Lee Jay
(see profile for equipment)

 

[Hamming]

FWIW, let me add that the column amplifier was shared between two pixels on the 20D. (and my personal read/analysis of 20D and 5D indicate that the 5D is a 20D with a larger sensor and a poorer AA filter normalized to Nyquist frequency).

Full-frame contains APS-C and while I have no such immediate aspirations, Julia Borg will be the proud owner of a Nikon FF by 2007, there's my prediction for what's next

Cheers!

The amps, for example, might be a smaller
fraction of the pixel area allowing more room for the well.

in light of the above, please explain your statements "5D's sensor
is 256% the size of the one in the 20D. This means at the same
f-stop, it captures 2.56x as much light" and "Well capacity is
around 50000e for 20D and around 85000 for 5D"

--
Julia

 

[Julia Borg]

Julia Borg will be the proud owner of a Nikon FF by
2007, there's my prediction for what's next

This one seems easier, and does not require any experiments or calculations, right?

But you are wrong here - no way.

If you do not see different color, different character of noise, very different pattern at high ISO you still have to explain why the same technology takes a year to hit the market.

--
Julia

 

[Hamming]

If you do not see different color, different character of noise,
very different pattern at high ISO you still have to explain why
the same technology takes a year to hit the market.

My guess is that they got the yield to a point where they were comfortable with the pricing (after all, nobody expected a consumer FF at this price so early) -- it has always been the case with CMOS processes that yields in the same lithographic process get better with time.

 

[Raul]

Especially if/when
the "holy cow" of raw will be understood obsolete

Will you please explain what you mean by the above?

With very kind regards

Raul

 

[Julia Borg]

Especially if/when
the "holy cow" of raw will be understood obsolete

Will you please explain what you mean by the above?

Hi Raul

some things can be done better in analog, like white balance and gamma correction. many noise reduction schemes are applicable in analog easier. resulting digital file will not be "raw" as we understand it now.

another option is to use sensors with logarithmic response instead of linear. if you want linear data from them (say, to apply white balance the way it is done now) you need to concert data to linear.

I think "raw" is a pretty poorly defined term. for example, when Nikon applies "pre-conditioning" to sensor data - is it still raw or not?

--
Julia

 

[Charles Musgrave]

Scaling the same technology doesn't mean linear scaling of their
properties

oh yes. but the scaling should follow a pattern. say if pixel size

It follows a pattern, but isn't necessarily linear.

is increased well capacity should be proportional to the increase
of surface - for same technology?

I don't know how linear the capacitances of these photodiodes are
with their surface areas. With very small wells the capacitance will
definitely be non-linear.

--
Julia

--
Chasm

 

[Julia Borg]

Scaling the same technology doesn't mean linear scaling of their
properties

oh yes. but the scaling should follow a pattern. say if pixel size

It follows a pattern, but isn't necessarily linear.

and who says it is?

I don't know how linear the capacitances of these photodiodes are
with their surface areas.

maybe you can look it up, within same technology?

--
Julia

 

[Charles Musgrave]

I would suspect a major challenge is getting dyes with the right
absorption spectra to adequately sense R, G or B, but not absorb
outside of those color ranges. Of course, the absorption ranges of
molecular dyes are not perfectly sharp, vary with temperature, and
can be affected by two photon processes, etc.

The Foveon sensor does not use dyes to distinguish colors in its
R,G and B layers. They use the sensing depth of silicon. Different
light frequencies tend to penentrate to different depths. Dyes
might be used, however, with a prefilter that could be used to help
fine-tune the response curves.

Interesting. As you can tell I haven't looked at the Foveon sensor very
closely! I wonder what the variability of penetration depth is for a
particular wavelength. It would be interesting if someone used
quantum confinement using quantum dots as dyes...

--
Jay Turberville
http://www.jayandwanda.com

--
Chasm

 

[Julia Borg]

That's what this is about? The difference between 82000 and "about
85000"? You should try entering the real world once in a while.

would be nice to know how you came up with this figure

--
Julia

 

[Radoslav]

So, when Nikon does come out with a full-frame camera, the
17-35/2.8 will be like an 11-24/2 would be on a DX camera (again
ignoring the 1/6 stop). You can go through this exercise with all
lenses.

exactly. see my thread

http://forums.dpreview.com/forums/readflat.asp?forum=1021&thread=16155253

With the DOF issue it mainly is the lack of a narrower DOF you get on a DX system.

So in the end Nikon has two options. Going FF, or making some pro f2,0 zooms and 1,4 primes for the DX sensor. I am betting on the last option, but even going to FF would mean that in the long end they'll have to come up with new f2,8 zooms with better resolving power (more lp/mm).

Gr.
Radoslav

 

[Kerry Pierce]

What other sense of the word is there?

The obvious one. "I have product A and product B, which do you
prefer?" That's a choice. As opposed to "Buy this product or I'll
shoot you." Is not really a choice. Nikon's current offerings
amount to "Buy product A or pi$$ off!" Again, not really a choice.

Well, you have choices, but the choices you have are not what you want. That's not the same as not having a choice.

Did you not have a choice when you bought nikon?

No! I was looking to a major system upgrade and fell in love with
the F4. I had no choice but to buy it.

hmmm, well, that's interesting.

Have you no choice in your next purchase?

A difficult question. Since investing in the Nikon system I have
made a significant investment in lenses and accessories.

I don't consider purchases of any of this stuff to be investments. Of course, I understand that the monetary loss of changing systems would be unwelcome, but it's better than being unhappy, IMO.

Nevertheless, I was seriously considering the 5D when this new hint
of a Nikon FF started to appear. I can afford to wait another year
or so, so maybe we'll know more by then.

I sincerely hope that this "hint" isn't just a cruel joke. It's one of the reasons that I don't think these FF discussions are useful here. Nobody outside of nikon knows what's going on with it. Giving false hope is not a kind or constructive thing to do, IMO.

I agree with your sentiment, especially when people say things like
"Just wait 'til Nikon delivers FF, then we'll gloat." Those
statements are just childish.

Agreed.

But I have no problem with Canon
users joining these debates. I find I learn the most from the
people I disagree with.

Lots of canon users are pretty cool people. Unfortunately, they aren't the ones you usually see posting here.

You seem to be taking some
people's desire for FF to be a personal affront. Those of us that
want FF don't expect it to be a Nikon exclusive.

That's ad hominem nonsense.

You've stumped me here. This seems to be a unique interpretation of
"ad hominem".

heh. It's ad hominem because the statement alluded to my personal feelings about people who want FF, as being the basis for my response to the fanboy. Not true, not pertinent, and diversionary, thus is within the definition of ad hominem, as I understand it.

I couldn't care less, if nikon makes a FF or not and whether you
want one or not.

Then why do you join in in all the FF threads and often belittle
the people who do want FF.

I don't join all the FF threads and have never been against FF. I read what strikes my fancy. I only make a negative reply when I see something misleading or in error.

Generally, the only people that I have unkind words for, are the instigators, flamebaiters. I don't make a habit of belittling people, without provocation.

On a personal note: I get a kick out your posts Kerry. Although we
argue a lot, I can't help but think that if we were neighbours we
would be good friends.

I think that is probably true. Adults can disagree without adversely affecting their relationship. Well presented arguments are often a learning experience, which I do enjoy.

--
my gallery of so-so photos
http://www.pbase.com/kerrypierce/root