Nikon to announce new products on Sep. 1

Archived from groups: rec.photo.digital.slr-systems (More info?)

Just saw an invitation to a press conference on Sep. 1 where
Nikon is going to announce several new products. No details
yet - just the assurance that this will be interesting.

Hence, the field is open for speculation, which is probably what
Nikon wants. Just a new flash, lens or yet another P&S after
Canon's 5D announcement would be lame - rather than interesting.
The D100 is overdue for a replacement, and the D2x is not, so my
money is on that the main event will be the D200.

The $64000 question (although I hope it retail for less than that)
is whether it will be FF or not.
--
- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ]
------------------------------------------------------------------------
Kodak DCS460, Canon Powershot G5, Olympus 2020Z
------------------------------------------------------------------------

Archived from groups: rec.photo.digital.slr-systems (More info?)

johnf@panix.com (John Francis) writes:
> Gisle Hannemyr <gisle+news@ifi.uio.no> wrote:

>> The $64000 question (although I hope it retail for less than that)
>> is whether it will be FF or not.

> Not a snowball's chance in hell.

You're probably right.

> The D200 will be positioned against the EOS 20D, at a price point
> I'd guess somewhere around $1200 - certainly no more than $1500.
> You can't sell a 'FF' sensor at that kind of price.

According to a thread in the dpreview forum. Nikon "accidently" posted
the specs. on the web here:
http://nikonimaging.com/global/pro [...] 200_6p.pdf
It's gone now, but these people claim to have them:
http://www.infodigitalcamera.com/b [...] ication-2/

Summary:
12.4 Mpx DX format (3 fps) and 6.4 Mpx 2.0x crop (5 fps),
ISO 100-800 (???), CF Card and 2.5 in monitor,
3D Matrix, Center weighted and Spot metering,
9 AF sensors, instant on, 52 ms shutter lag.

If these are the specs, it sure looks as if it is positioned to
compete with the 20D. The only weird bit is 800 as the top ISO
sensitivity.

> Well, not for long, anyway. I suppose it's just conceivable that
> we'd see an APS-H (1.3x crop factor), but even that is unlikely;
> having a smaller sensor on your top-end model than on a midrange
> body seems a rather unlikely strategy.

I think Nikon's committment to DX-lenses rules out a 1.3x crop.
--
- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ]
------------------------------------------------------------------------
Kodak DCS460, Canon Powershot G5, Olympus 2020Z
------------------------------------------------------------------------

Archived from groups: rec.photo.digital.slr-systems (More info?)

In article <SlzPe.96466$G8.44450@text.news.blueyonder.co.uk>,
David J Taylor <david-taylor@blueyonder.co.not-this-bit.nor-this-part.uk.invalid> wrote:
>Gisle Hannemyr wrote:
>> The $64000 question (although I hope it retail for less than that)
>> is whether it will be FF or not.
>
>I sincerely hope they do provide a full-frame camera at some point in the
>near future - all this using large and bulky 35mm bodies and mounting toy
>sensors in them which limit wide-angle capability has gone on for too
>long! <G> (I don't think they will do that on September 1st, though).

I wonder how far the resolution of 1.5 crop sensors can be pushed without
seriously compromising dynamic range.

Suppose that Nikon creates a camera with interchangeable sensors: one
24 Mpixpel sensor that only works at ISO 200, and a 6 Mpixel sensor that
goes up to ISO 1600 or higher. (and a B/W sensor, and an IR sensor, etc.)

That may be cost effective as long as full frame sensors remain at least
twice as expensive as APS-C sensors.

It would make for an interesting shoot out if Nikon announces a 12 Mpixel
D200 for less than $2000.


--
That was it. Done. The faulty Monk was turned out into the desert where it
could believe what it liked, including the idea that it had been hard done
by. It was allowed to keep its horse, since horses were so cheap to make.
-- Douglas Adams in Dirk Gently's Holistic Detective Agency

Archived from groups: rec.photo.digital.slr-systems (More info?)

In article <U6BPe.10666$p%3.41270@typhoon.sonic.net>,
scharf.steven@geemail.com says...

> Not weird at all, if it really is 1.5x and 12.4 megapixel. Nikon's taken
> a beating on the noise issues with the D2x at the higher ISOs--better to
> just leave it out completely than to have to deal with comparisons with
> the 20D.

What noise issues???

Oh, it's just you, Scharf. Moving on...

--
Save Photography | Shoot some film today!
Email: drop rods and insert surfaces

Archived from groups: rec.photo.digital.slr-systems (More info?)

burnsdavidj@yahoo.com wrote:

> I agree that the cap @ 800 ISO is a bit odd, the rest of the specs
> looks decent. A boost in mpxls over the 20D will give it some market
> legs, at least in the short term until Canon comes out with a 20D
> replacement (and thus the leapfroging game continues).

It's a pretty good marketing move, to emphasize your positives, while
preventing comparisons against the negatives. Most people will be
content with 800 ISO, so the reviews won't ding the omission of higher
ISO speeds nearly as much as they would ding noise at the higher ISO
speeds. Nikon needs to avoid comparisons of noise, against their Canon
competition, as much as possible. We all saw what happened with the D2x
high ISO noise in the reviews, and Nikon is determined to avoid a repeat
of this.

Archived from groups: rec.photo.digital.slr-systems (More info?)

SMS wrote:
> burnsdavidj@yahoo.com wrote:
>
>> I agree that the cap @ 800 ISO is a bit odd, the rest of the specs
>> looks decent. A boost in mpxls over the 20D will give it some market
>> legs, at least in the short term until Canon comes out with a 20D
>> replacement (and thus the leapfroging game continues).
>
>
> It's a pretty good marketing move, to emphasize your positives, while
> preventing comparisons against the negatives. Most people will be
> content with 800 ISO, so the reviews won't ding the omission of higher
> ISO speeds nearly as much as they would ding noise at the higher ISO
> speeds. Nikon needs to avoid comparisons of noise, against their Canon
> competition, as much as possible. We all saw what happened with the D2x
> high ISO noise in the reviews, and Nikon is determined to avoid a repeat
> of this.

Obviously you don't have a D2X or you'd know the noise "problem" is not
a problem at all. Canon use in-camera noise reduction which, when you
later try to enlarge the image makes it extremely hard to produce detail.

The Nikon approach is much preferred by experienced professionals. Use
of software like "Neat Image" or "Noise Ninja" during editing in
Photoshop produces far more detailed enlargements than from a
plasticized Canon hi-ISO image.

--
Douglas,
You never really make it on the 'net
until you get your own personal Troll.
Mine's called Chrlz. Don't feed him, he bites!

Archived from groups: rec.photo.digital.slr-systems (More info?)

pixby <pixby_douglas@hotmail.com> wrote:

> Obviously you don't have a D2X or you'd know the noise "problem" is not
> a problem at all.

Steven has been relentlessly pounding away with his lies about the D2x
"noise problem" since before it was even released, despite people who
(unlike him) have actually used it telling him that no such problem
exists. He is unwilling to let facts get in the way of his mindless
Canon cheerleading.

--
Jeremy | jeremy@exit109.com

Archived from groups: rec.photo.digital.slr-systems (More info?)

In message <430f6dfd$1@dnews.tpgi.com.au>,
pixby <pixby_douglas@hotmail.com> wrote:

>
>Obviously you don't have a D2X or you'd know the noise "problem" is not
>a problem at all. Canon use in-camera noise reduction which, when you
>later try to enlarge the image makes it extremely hard to produce detail.
>
>The Nikon approach is much preferred by experienced professionals. Use
>of software like "Neat Image" or "Noise Ninja" during editing in
>Photoshop produces far more detailed enlargements than from a
>plasticized Canon hi-ISO image.

I'm still waiting for some kind of proof of this. The noise
characteristic of the 20D looks full-spectrum to me; just weak. If what
you were saying were true, a histogram equalization of a RAW blackframe
would be dominated by low-frequency-noise, but it isn't.

What the recent Canons are known to do is measure and cancel some of its
own known error; never heard of anything about trying to filter random
noise of the sensor.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

Archived from groups: rec.photo.digital.slr-systems (More info?)

<JPS@no.komm> wrote:

> Words like "problem" are loaded. I prefer more objective terminology.
> I suspect that the intensity of noise in the RAW data is probably a
> little bit greater with the D2X, making it a little noisier than
> something like the 10D in low, but well-balanced light,

Measurements have the D2x about equal to the 1Ds2 in noise, up to ISO 800.
But the point is that, at low ISO, noise just isn't a factor in DSLRs any
more.

> but that the D2X might be competitive in greatly unbalanced light sources,
> as it digitizes each color channel separately, so the weakest link in the
> chain of noise is eliminated (assuming, of course, that your WB setting
> on the D2X is close to what you're shooting under).

On the D70, you could sometimes see the blue channel start to fall apart
if you had to boost it quite a lot (like setting white balance to 2000K,
or otherwise) but I have yet to see that happen on the D2x. You get to
general shadow noise before you get to that point, and the shadow noise
only starts to happen about where you'd expect it to. I can't compare
it to anything from Canon from experience, but compared to anything I've
used, the D2x is very good in wacky light.

--
Jeremy | jeremy@exit109.com

Archived from groups: rec.photo.digital.slr-systems (More info?)

In message <11gv4v53eonn2d@corp.supernews.com>,
Jeremy Nixon <jeremy@exit109.com> wrote:

> <JPS@no.komm> wrote:

>> Words like "problem" are loaded. I prefer more objective terminology.
>> I suspect that the intensity of noise in the RAW data is probably a
>> little bit greater with the D2X, making it a little noisier than
>> something like the 10D in low, but well-balanced light,

>Measurements have the D2x about equal to the 1Ds2 in noise, up to ISO 800.
>But the point is that, at low ISO, noise just isn't a factor in DSLRs any
>more.

It's not perticiularly visible in the lower ISOs, unless you boost the
shadows. At high ISOs, though, all the DSLRs are still lacking, to
various degrees. People who only shoot landscapes or architecture on a
tripod at low ISO, or with studio lighting have no idea how dark the
conditions are for people shooting wildlife under a canopy of foliage,
or at dusk or dawn. Animals generally don't spend a lot of time in
bright sunlight, and when they do, they are usually very spooky. You
can generally get much closer to animals after the sun has set than
mid-day. I arrived at a local park the other day about 2.5 hours before
sunset, and saw a foe and fawn feeding in a field, 150 feet from me.
They immediately took to the woods, as I got within 100 feet of them. I
went to the flower garden, to shoot hummingbirds, and upon my return
after dark, the foe and fawn allowed me within 10 feet of them without
any sign of distress; of course, ISO 1600 shots at the slowest
hand-holdable shutter speed and the lens wide-open, even at 10mm, were
grossly underexposed. Only the flash shots I took were well-saturated,
but they looked like garbage.

>> but that the D2X might be competitive in greatly unbalanced light sources,
>> as it digitizes each color channel separately, so the weakest link in the
>> chain of noise is eliminated (assuming, of course, that your WB setting
>> on the D2X is close to what you're shooting under).
>
>On the D70, you could sometimes see the blue channel start to fall apart
>if you had to boost it quite a lot (like setting white balance to 2000K,
>or otherwise) but I have yet to see that happen on the D2x. You get to
>general shadow noise before you get to that point, and the shadow noise
>only starts to happen about where you'd expect it to. I can't compare
>it to anything from Canon from experience, but compared to anything I've
>used, the D2x is very good in wacky light.

We probably won't see this feature from Canon for a while, as none of
the reviewers are raking them over the coals for the poor incandescent
and deep-blue-sky-shade performance (poor compared to what it could be;
not absolutely poor compared to other cameras, per se).

--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

Archived from groups: rec.photo.digital.slr-systems,rec.photo.digital.zlr (More info?)

On Fri, 26 Aug 2005 07:17:06 GMT, "David J Taylor"
<david-taylor@blueyonder.co.not-this-bit.nor-this-part.uk.invalid>
wrote:

>Gisle Hannemyr wrote:
>> Just saw an invitation to a press conference on Sep. 1 where
>> Nikon is going to announce several new products. No details
>> yet - just the assurance that this will be interesting.
>>
>> Hence, the field is open for speculation, which is probably what
>> Nikon wants. Just a new flash, lens or yet another P&S after
>> Canon's 5D announcement would be lame - rather than interesting.
>> The D100 is overdue for a replacement, and the D2x is not, so my
>> money is on that the main event will be the D200.
>>
>> The $64000 question (although I hope it retail for less than that)
>> is whether it will be FF or not.
>
>I sincerely hope they do provide a full-frame camera at some point in the
>near future - all this using large and bulky 35mm bodies and mounting toy
>sensors in them which limit wide-angle capability has gone on for too
>long! <G> (I don't think they will do that on September 1st, though).

What some people need to do is try not to act like old fogies who for
whatever irrational reason can't seem to let go of the "35mm" film
size. It has NOTHING to do with digital sensors and legacy lenses do
not do justice to a full frame digital sensor anyway.
-Rich

>
>On the ZLR front, I would like to see a move from the 8MP sensor down to a
>less-noisy 7MP sensor or something which gains low-light capability, with
>an improvement to the telephoto aperture on both the Coolpix 8400 and
>8800, and the addition of image stabilisation to the 8400.
>
>David
>


"Bittorrents are REFUNDS for all the BAD movie products Hollywood
never gave us refunds for in the past"

Archived from groups: rec.photo.digital.slr-systems (More info?)

On Fri, 26 Aug 2005 23:25:36 +0100, Tony Polson <tp@nospam.co.uk>
wrote:

>Gisle Hannemyr <gisle+news@ifi.uio.no> wrote:
>
>>Just saw an invitation to a press conference on Sep. 1 where
>>Nikon is going to announce several new products. No details
>>yet - just the assurance that this will be interesting.
>>
>>Hence, the field is open for speculation, which is probably what
>>Nikon wants. Just a new flash, lens or yet another P&S after
>>Canon's 5D announcement would be lame - rather than interesting.
>>The D100 is overdue for a replacement, and the D2x is not, so my
>>money is on that the main event will be the D200.
>>
>>The $64000 question (although I hope it retail for less than that)
>>is whether it will be FF or not.
>
>
>No, it will not. Nikon has a full frame DSLR under development but it
>will be in the pro series (D3?) and won't be announced until PMA 2006
>at the earliest. Shipping won't be before Q3/2006.

Tony,

How do you know this? Do you go to same shabu-shabu restaurant as
some of the senior Nikon execs? Been there when they have all
been "singing" after drinking just a wee bit too much?

Pricing?

Archived from groups: rec.photo.digital.slr-systems (More info?)

SMS wrote:

> Gisle Hannemyr wrote:
>
>> If these are the specs, it sure looks as if it is positioned to
>> compete with the 20D. The only weird bit is 800 as the top ISO
>> sensitivity.
>
> Not weird at all, if it really is 1.5x and 12.4 megapixel. Nikon's taken
> a beating on the noise issues with the D2x at the higher ISOs

And Nikon has chosen not to accept the poor WA performance Canon FF cameras
have. I know you are obsessed with high ISO noise performance but people
that actually USE a camera rather than calculating the specs are more
impressed with good optical performace edge to edge.
--

Stacey

Archived from groups: rec.photo.digital.slr-systems (More info?)

Father Kodak wrote:

> Which older Nikon lenses wouldn't do justice to a
> full-frame digital sensor?

You'll have to wait for a full frame camera to find out. Just look at the
examples from canon's wide zooms to see how poorly a digital FF sensor can
react to use with a "legacy" lens. At least the newer Nikon zooms are being
designed with a longer exit pupil distance so they at least have a good
chance of being FF friendly.

--

Stacey

Archived from groups: rec.photo.digital.slr-systems (More info?)

Stacey <fotocord@yahoo.com> wrote:

> You'll have to wait for a full frame camera to find out. Just look at the
> examples from canon's wide zooms to see how poorly a digital FF sensor can
> react to use with a "legacy" lens. At least the newer Nikon zooms are being
> designed with a longer exit pupil distance so they at least have a good
> chance of being FF friendly.

It would be interesting, once the 5D is here, given the poor example Canon
posted, a test with that Canon zoom compared to the Nikon 17-35 mounted on
the 5D via an adapter. I for one would love to see how much of the problem
can be solved by the lens design. It would make it that much easier to see
how viable full-frame really is at the present time.

--
Jeremy | jeremy@exit109.com

Archived from groups: rec.photo.digital.slr-systems (More info?)

In article <430f73cb$0$11076$e4fe514c@news.xs4all.nl>,
Bart van der Wolf <bvdwolf@no.spam> wrote:
>
>"Philip Homburg" <philip@pch.home.cs.vu.nl> wrote in message
>news:8bbporlunfv5gj9i2hcc33b8p0@inews_id.stereo.hq.phicoh.net...
>SNIP
>> I wonder how far the resolution of 1.5 crop sensors can be pushed
>> without seriously compromising dynamic range.
>
>Since Dynamic range is determined by potential well depth and noise,
>the well depth limitation of/by small sensor elements seems to be the
>issue to me (noise reduction can only be improved to a point). And
>well depth is limited by design to roughly scale with sensel
>size/area.

I know how it scales, but are current designs at the limit?

Another thing (I have no idea if that would work), as far as I understand,
you get photon shot noise when photons are turned into electrons. What
happens if you use the CMOS circuitry to store just 1 electron for
every 10 'photon-related' electrons?


--
That was it. Done. The faulty Monk was turned out into the desert where it
could believe what it liked, including the idea that it had been hard done
by. It was allowed to keep its horse, since horses were so cheap to make.
-- Douglas Adams in Dirk Gently's Holistic Detective Agency

Archived from groups: rec.photo.digital.slr-systems (More info?)

"Stacey" <fotocord@yahoo.com> wrote in message
news:3nahppFkpvrU3@individual.net...
> Father Kodak wrote:
>
>> Which older Nikon lenses wouldn't do justice to a
>> full-frame digital sensor?
>
> You'll have to wait for a full frame camera to find out. Just look at the
> examples from canon's wide zooms to see how poorly a digital FF sensor can
> react to use with a "legacy" lens. At least the newer Nikon zooms are
> being
> designed with a longer exit pupil distance so they at least have a good
> chance of being FF friendly.
>
> --
>
> Stacey

Again, Stacey, the 17-40 isn't a "legacy" lens, it postdates the 1D
somewhat. Which doesn't explain the WA shot on Canon's website, I'll agree.
But I still want to know why, if digital res is just equal to film, the
performance of the WA lenses significantly worse on full frame digital than
on film. After all, the sensor size is similar.

--
Skip Middleton
http://www.shadowcatcherimagery.com

Archived from groups: rec.photo.digital.slr-systems (More info?)

Stacey wrote:

> Father Kodak wrote:
>
>
>>Which older Nikon lenses wouldn't do justice to a
>>full-frame digital sensor?
>
>
> You'll have to wait for a full frame camera to find out. Just look at the
> examples from canon's wide zooms to see how poorly a digital FF sensor can
> react to use with a "legacy" lens. At least the newer Nikon zooms are being
> designed with a longer exit pupil distance so they at least have a good
> chance of being FF friendly.
>

OTOH, that's as much a function of sensor design (or more properly the
algorithm for recording sensor data) as any "problem" with the legacy lens.

Canon could make their full frame sensor more compatable with their
existing wide angle lenses, but why would they, since it would make it
harder to motivate you to buy new wide angle lenses for your new FF
digital slr.

Archived from groups: rec.photo.digital.slr-systems (More info?)

On Sat, 27 Aug 2005 10:46:40 -0400, no one
<noone@nowhere.nohow.invalid> wrote:

>Father Kodak wrote:
>
>
>> Wow. I still do film and I carry either two or three bodies, K64,
>> Ekta pushed to 3200 and a print film at 100 or 400. One of the
>> reasons for me to go digital is that I would need only one body. Now
>> I would need two!
>
>Doesn't matter if you're doing film or digital, two bodies is a good
>idea because Murphy was an optimist.

Film bodies make a great backup. Especially if you already own them
and you already have a scanner.

Archived from groups: rec.photo.digital.slr-systems (More info?)

"Dimitri Cohen" <leecher@leechers.org> wrote in message
news:UN2Qe.2288$Wd7.1253@newsread1.news.pas.earthlink.net...
>> Again, Stacey, the 17-40 isn't a "legacy" lens, it postdates the 1D
>> somewhat. Which doesn't explain the WA shot on Canon's website, I'll
>> agree.
>> But I still want to know why, if digital res is just equal to film, the
>> performance of the WA lenses significantly worse on full frame digital
>> than on film. After all, the sensor size is similar.
>
> The problem is with photo elements on digital sensors. They react to light
> differently when it doesn't fall at the 90 degree angle. As you get
> further away from the center of the projected by the lens image, the angle
> at which the rays hit the surface of the sensor deviates from the 90
> degree angle further and further. Canon somewhat have a solution to the
> problem by placing a microscopic lens in front of each light element right
> on the sensor, thus converting the light right before it hits the element.
> Others, (Olympus with the 4/3 format of theirs), approaches this problem
> by redesigning the lens to achieve parallelism of all light rays projected
> onto the sensor, which, I must add, is an extremely difficult process.
>
> Film, on the other hand, doesn't really care (or probably not as much) at
> which angle the light hits the reactive surface. Hence, the same lenses
> that performed quite well (quality of image wise) on film bodies will now
> perform somewhat worse on a digital body with the same size sensor.
>
> I understand why Nikon take their time to come up with an FF sensor body.
> Their goal is to retain the quality of the image while allowing their
> customers use the same glass arsenal that was used for the 35mm film
> bodies. I'm guessing they will have to come with a similar solution like
> Canon (or maybe redesign the sensor totally) since changing the lenses
> would, IMHO, be less desirable for the already existing crowd of Nikon
> users.
>
>
> Dimitri
>
>
> P.S. I'm not partial to anyone but I love my 20D
>
Thanks, I don't know how many times I've voiced that question, and you are
the first to provide a cogent answer. Now I understand it better, it leaves
the question of why Canon, when they designed the 16-35, 24-70 and 17-40, in
chronological order, didn't take that into consideration. These lenses were
released post digital, and in their blurbs, Canon even talks about them
being suitable for digital use...

--
Skip Middleton
http://www.shadowcatcherimagery.com

Archived from groups: rec.photo.digital.slr-systems (More info?)

Dimitri Cohen wrote:

>
> Film, on the other hand, doesn't really care (or probably not as much) at
> which angle the light hits the reactive surface. Hence, the same lenses
> that performed quite well (quality of image wise) on film bodies will now
> perform somewhat worse on a digital body with the same size sensor.


Yes THIN emulsion films don't care about this at all, the old thick emulsion
films did have an issue similar to what digital sensors have to deal with
and weren't nearly as sharp, especially with wide angle optics.


--

Stacey

Archived from groups: rec.photo.digital.slr-systems (More info?)

Jeremy Nixon wrote:

> Stacey <fotocord@yahoo.com> wrote:
>
>> You'll have to wait for a full frame camera to find out. Just look at the
>> examples from canon's wide zooms to see how poorly a digital FF sensor
>> can react to use with a "legacy" lens. At least the newer Nikon zooms are
>> being designed with a longer exit pupil distance so they at least have a
>> good chance of being FF friendly.
>
> It would be interesting, once the 5D is here, given the poor example Canon
> posted, a test with that Canon zoom compared to the Nikon 17-35 mounted on
> the 5D via an adapter. I for one would love to see how much of the
> problem
> can be solved by the lens design.

I for 2 would like to see this done.. I'm tired of having to deal with film,
shooting with my mamiya 645 and would LOVE to see a system that would
replace it. From what I've seen so far, the 5D with canon optics aren't
going to do it.
--

Stacey

Archived from groups: rec.photo.digital.slr-systems (More info?)

In article <mfs1h1tvvsosah42u88svp9s0butfir1gr@4ax.com>,
Father Kodak <dont_bother@IDontCare.COM> wrote:
>>
>>>It's a nice idea, though, as long as the rest of the electronics is
>>>upgrade able to accommodate larger files from future hi-res sensors.
>>
>>I guess that the limits of resolution will be reached soon (though there is
>>the 'Bayer' factor). If you compute DoF using the size of a single sensor
>>element in a 16 Mpixel camera as CoC, you get close to nothing.
>
>
>Are you saying that the circle-of-confusion size is much smaller than
>a single sensor element, or the other way around?

I'm saying that the traditional CoC size for 35mm is a lot bigger than a
single sensor element.


--
That was it. Done. The faulty Monk was turned out into the desert where it
could believe what it liked, including the idea that it had been hard done
by. It was allowed to keep its horse, since horses were so cheap to make.
-- Douglas Adams in Dirk Gently's Holistic Detective Agency

Archived from groups: rec.photo.digital.slr-systems (More info?)

"Stacey" <fotocord@yahoo.com> wrote in message
news:3ncvnoFsv1dU3@individual.net...
> Skip M wrote:
>
>> Canon even talks about
>> them being suitable for digital use...
>>
>
> Suitable isn't the same thing as "optimal". Remember, Canon is in the
> business of selling people things, They sell you a 17-40, then 4 years
> later they come out with a "digitally optimized" version and get to sell
> another batch of the same focal length lenses. Watch and see if this
> doesn't happen.
>
> --
>
> Stacey

Here's what the Canon Camera Museum blurb has to say about the 17-40 L we've
been talking about...
"Featuring a new optical system design, the new lens delivers an expanded
zoom range of 17-40mm, with the wide-angle-end 17mm focal length making it
ideally suited to digital photography." I wonder what their idea of "works
ok, but not great" is...
I know they're talking about focal length, but still...

--
Skip Middleton
http://www.shadowcatcherimagery.com

Archived from groups: rec.photo.digital.slr-systems (More info?)

In article <q5zmr2ntde.fsf@nelja.ifi.uio.no>,
Gisle Hannemyr <gisle+news@ifi.uio.no> wrote:
>philip@pch.home.cs.vu.nl (Philip Homburg) writes:
>> I'm saying that the traditional CoC size for 35mm is a lot bigger
>> than a single sensor element.
>
>But the sampling theorem tells you that the best resolution you can
>hope to resolve is /two times/ the sensor pitch. And in most modern
>cameras, there is an AA-filter in front of the sensels that filters
>out any signal with a frequency higher than the nyquist limit (= two
>times the sampling frequency).

I am going to ignore the AA filter. There is basically no camera with a MTF
of zero at nyquist. Check the resolution targets at dpreview.

At nyquist (actually, slightly below nyquist) you can resolve a complete
cycle: black and white. This means that the CoC should have a frequency that
is double the nyquist frequency. In other words, the CoC has to be smaller
than the sensor element size (however, two times the sensor element size is
good for Bayer pattern sensors).

>The CoC-value used for 135-format film was typically 25 µm. Because
>the crop factor, the corresponding CoC for a 1.6x crop digital sensor
>is 16 µm. As it happens, the sensor diameter in the 6 Mpx DSLRs are
>around 8 µm, tow times that is 16 µm, so 16 µm is also the best
>resolution this camera can have according to the sampling theorem.

For a full-frame 35mm dSLR, there is no crop factor. The 25 micrometer should
compared to the 7.3 micrometer of the 1Ds mk II.

I don't think your computation makes much sense: for an 1.6 crop sensor,
you have to reduce the CoC with same factor.

The pixel equivalent for full-frame 35mm is easy to compute:
36/0.025 * 24/0.025 = 1382400 or 1.4 Mpixel.

Which makes sense because a CoC of 25 micrometer is already much too big
for good quality enlargements of 35mm film.

>This also tells us that neyond 6 Mpx, the megapixel game is yielding
>diminshing returns.

Of course not. The CoC has nothing to do with image quality. Image quality
is ultimately limited by diffraction (assuming perfect lenses).

If there is nothing beyond 6 Mpixel, then large format much have been a
complete waste of time for many years.


--
That was it. Done. The faulty Monk was turned out into the desert where it
could believe what it liked, including the idea that it had been hard done
by. It was allowed to keep its horse, since horses were so cheap to make.
-- Douglas Adams in Dirk Gently's Holistic Detective Agency