D1 Noise Hoax Exposed

[smack]

Finally an intelligent test that tells the true story. Check out this web site:
http://www.robgalbraith.com/diginews/index2.html

 

[Erik13455]

YUP...I liked it...I would like to believe it, BUT...

I have read EVERY posting on this forum. And I have looked at EVERY picture.
There are at least TWO folks in the news business who HAVE the camera and
who have complained. And then three or four others..some WITH posted samples.
So..several folks who have all managed to make the SAME mistake with this camera
(of course, all of THEM may be influenced by Mr. Linnemann's alert).

"Occam's razor" says the simplest solution that fits the facts is generally
the solution to a problem. Not everyone uses the SAME camera. The variations
in yield of usable, high quality CCDs--WHICH NIKON DOESN'T MAKE--is the
"simplest solution."

Now we have Rob Galbraith with yet another sample of the D-1 and a convincing
explanation for other folks' observations. To believe Mr. Galbraith, I have
to discount almost everyone else's data. OKAY...I can do that IF:

(1) Everyone with 'blue noise' recants based on (this) new information.
and/or

(2) Nikon provides a public explanation of the phenomenon -- presumeably supporting Mr. Galbraith's argument.

YUP...I liked the article...I just am not completely convinced that it's the last word.

--------------------------------------

Finally an intelligent test that tells the true story. Check out this web
site:
http://www.robgalbraith.com/diginews/index2.html

 

[Peter iNova]

Mr. Galbraith's own images show the phenomenon. 950's don't.

Rob's "Sunset" image has it if you try to pull the silouette detail out of those condos:

down load it from



) (remove the trailing parenthesis, it's an attempt to NOT get the huge picture to load here)

My own 950 doesn't show anything of the sort when pushed to the degree that the D1's do. Here's a before and after:



In Photoshop's "Levels" control, simply drag the rightmost (highlight) slider to the left until it reads 0, 1.0, 31 and you will have this:



Suddenly the shot made at ISO 320 is boinked up to ISO 2gazillion. No streaks, no flaws no bands no D1-like artifacts.

If I can get this kind of performance from a $1000 camera, the question stands unanswered: Why can't I get similar performance (exposure only, of course) out of the D1? Can anybody? Is it a batch of bad imager chips? A software bug? A post sensor processing routine? A thing so onerous that Nikon can't speak to it? What is happening to the shadow detail of these cameras and why?

-iNova

I have read EVERY posting on this forum. And I have looked at EVERY picture.
There are at least TWO folks in the news business who HAVE the camera and
who have complained. And then three or four others..some WITH posted
samples.
So..several folks who have all managed to make the SAME mistake with this
camera
(of course, all of THEM may be influenced by Mr. Linnemann's alert).

"Occam's razor" says the simplest solution that fits the facts is generally
the solution to a problem. Not everyone uses the SAME camera. The variations
in yield of usable, high quality CCDs--WHICH NIKON DOESN'T MAKE--is the
"simplest solution."

Now we have Rob Galbraith with yet another sample of the D-1 and a
convincing
explanation for other folks' observations. To believe Mr. Galbraith, I have
to discount almost everyone else's data. OKAY...I can do that IF:

(1) Everyone with 'blue noise' recants based on (this) new information.
and/or
(2) Nikon provides a public explanation of the phenomenon -- presumeably
supporting Mr. Galbraith's argument.

YUP...I liked the article...I just am not completely convinced that it's
the last word.

--------------------------------------

Finally an intelligent test that tells the true story. Check out this web
site:
http://www.robgalbraith.com/diginews/index2.html

 

[Rob Galbraith]

Hi All,

I appreciate the test I performed being called intelligent, but I would have to disagree. Actually, I wouldn't call the test I performed much less flawed

than the test at http://www.pressphoto.dk . It was merely meant to illustrate, for the benefit of those who have looked at http://www.pressphoto.dk and thought that

their D1 images at any ISO, of actual subjects doing actual things, were actually going to look like the images in that test, that they needn't lose sleep.

Especially since it was causing me to lose sleep by being wakened at 6:30 AM to talk to a distraught Director of Photography!

Peter Inova sent me a helpful note this evening in response to what I wrote. In reading it I became concerned that my central message was being
lost. Let me distill the 1000+ words on my site down to three points:

• The D1 will deliver images that appear to be noise free when shot at low ISOs, if the images are properly exposed, as they should be in the hands

of a professional photographer. That is, the noise is THERE, but it isn't visible.

• The D1 gets noisy at higher ISOs. That is, the noise that's always there becomes visible. It crosses the visible threshold and begins to intrude upon

image content.

• On balance, the D1's overall noise it is subjectively no more or less severe than a DCS 520/D2000/DCS 620 image at higher ISOs when

comparable images are toned for printing.

I repeat, I know the streaks are there, as I can do the same things as everyone else in Photoshop to bring them out. Peter, you use the example of the

sunset shot on my site to make your point. That illustrates my point too. If the shadows are cranked in Photoshop then the streaks appear. But the

picture was shot to be a silhouette, and prints without any hint of the streaking when the shadows are kept as dark as I wanted them to be to keep the
mood in the photo.

The title of the previous message is D1 Noise Hoax Isn't Fake. Let me be clear that I was not trying to say that the D1 doesn't plop noise into the

images it produces. It does, more than I expected given Nikon's marketing of the camera prior to its release. I was above all trying to introduce an

idea that seemed altogether lost in the D1 noise discussion: does the noise intrude into real photos printed on real paper and viewed by real

people? That's what matters most to me as a photographer. In that context, I've been really happy with the low-ISO D1 images I've shot over the past
several

 

[Erik13455]

Thank you Mr. Galbraith!

So there you have it from the horse's mouth, as it were. The Nikon DOES "plop noise into the picture." More than Nikon's "marketing hype" led any of us to believe. If one reads their brochures on-line, Nikon makes a big deal out of the camera's capabilities at very high ISO ratings. (I don't think that too many folks are planning to shoot at 1600 in bright sunlight! Do you?) So darkness is a given..and shouldn't surprise the manufacturer when folks start to complain about noise. The "massaging" with Photoshop is a given too, since it's much harder to get what one wants from photographs taken in low light (e.g. Mr. Linneman's shots at night)

Which leaves us with Nikon's response. Where is it?

And I really, really liked Mr. Galbraith's treatment of this whole subject.

Hi All,

I appreciate the test I performed being called intelligent, but I would
have to disagree. Actually, I wouldn't call the test I performed much
less flawed
than the test at http://www.pressphoto.dk . It was merely meant to illustrate,
for the benefit of those who have looked at http://www.pressphoto.dk and thought
that
their D1 images at any ISO, of actual subjects doing actual things, were
actually going to look like the images in that test, that they needn't
lose sleep.
Especially since it was causing me to lose sleep by being wakened at 6:30
AM to talk to a distraught Director of Photography!

Peter Inova sent me a helpful note this evening in response to what I
wrote. In reading it I became concerned that my central message was being
lost. Let me distill the 1000+ words on my site down to three points:

• The D1 will deliver images that appear to be noise free when shot at

low ISOs, if the images are properly exposed, as they should be in the
hands
of a professional photographer. That is, the noise is THERE, but it isn't
visible.

• The D1 gets noisy at higher ISOs. That is, the noise that's always

there becomes visible. It crosses the visible threshold and begins to
intrude upon
image content.

• On balance, the D1's overall noise it is subjectively no more or less

severe than a DCS 520/D2000/DCS 620 image at higher ISOs when
comparable images are toned for printing.

I repeat, I know the streaks are there, as I can do the same things as
everyone else in Photoshop to bring them out. Peter, you use the example
of the
sunset shot on my site to make your point. That illustrates my point too.
If the shadows are cranked in Photoshop then the streaks appear. But the
picture was shot to be a silhouette, and prints without any hint of the
streaking when the shadows are kept as dark as I wanted them to be to
keep the
mood in the photo.

The title of the previous message is D1 Noise Hoax Isn't Fake. Let me be
clear that I was not trying to say that the D1 doesn't plop noise into the
images it produces. It does, more than I expected given Nikon's marketing
of the camera prior to its release. I was above all trying to introduce an
idea that seemed altogether lost in the D1 noise discussion: does the
noise intrude into real photos printed on real paper and viewed by real
people? That's what matters most to me as a photographer. In that
context, I've been really happy with the low-ISO D1 images I've shot over
the past
several

 

[lloyd chambers]

This afternoon I shot with a D1 at Keeble & Shucat Photography in Palo Alto, CA. They have a demo unit there now (not for sale).

I can say with 100% certainty that the D1 DOES SUFFER from a horizontal "banding" noise pattern EVEN AT ISO 200. I see it in every photo I took which has darker areas. It's plainly visible viewing just the unaltered file--no Photoshop Levels or any adjustment are necessary to see the problem. I printed one example and it's also visible in the print (Epson Stylus Photo 700).

I used the D1 in Program mode, ISO 200, JPEG Fine with a 35-70/f2.8 zoom. Then I shot 55 pictures over about 20 minutes.

Though I think the casual viewer would probably not notice the problem, it's plainly visible. Otherwise, the photos are stunning.

Lloyd

Finally an intelligent test that tells the true story. Check out this web
site:
http://www.robgalbraith.com/diginews/index2.html

 

[Francis]

Thank you Mr. Galbraith!

So there you have it from the horse's mouth, as it were. The Nikon DOES
"plop noise into the picture." More than Nikon's "marketing hype" led any
of us to believe. If one reads their brochures on-line, Nikon makes a big
deal out of the camera's capabilities at very high ISO ratings. (I don't
think that too many folks are planning to shoot at 1600 in bright
sunlight! Do you?) So darkness is a given..and shouldn't surprise the
manufacturer when folks start to complain about noise. The "massaging"
with Photoshop is a given too, since it's much harder to get what one
wants from photographs taken in low light (e.g. Mr. Linneman's shots at
night)

Eric, here is a strange answer directly from Nikon Japan:

Dear Sir,

Thank you for your E-mail and the information of the
website which we should have checked out:

Unfortunately, we cannot cancel the noise of the CCD at
the moment.
However, we think that D1 in this condition is still useful
for many customers. Furthermore, the data includs which
we cannot know the test condition.
Generally, if you set up long shutter speed, the noise
increases.

Anyway, we suggest that you will check the other various
information and consider if you obtain one or not.

Please accept our sincere apologies for any inconvenience
caused by this problem.

Sincerely yours,
Nikon Corp. spokesman

What do you make of this?

Francis

 

[smack]

Are people really reading Mr. Galbraith's article or are they reading what they want too believe. The bottom line is that the $5000 D1 displays the same amount noise as the
$15000 Kodak DCS.

So there you have it from the horse's mouth, as it were. The Nikon DOES
"plop noise into the picture." More than Nikon's "marketing hype" led any
of us to believe. If one reads their brochures on-line, Nikon makes a big
deal out of the camera's capabilities at very high ISO ratings. (I don't
think that too many folks are planning to shoot at 1600 in bright
sunlight! Do you?) So darkness is a given..and shouldn't surprise the
manufacturer when folks start to complain about noise. The "massaging"
with Photoshop is a given too, since it's much harder to get what one
wants from photographs taken in low light (e.g. Mr. Linneman's shots at
night)

Which leaves us with Nikon's response. Where is it?

And I really, really liked Mr. Galbraith's treatment of this whole subject.

Hi All,

I appreciate the test I performed being called intelligent, but I would
have to disagree. Actually, I wouldn't call the test I performed much
less flawed
than the test at http://www.pressphoto.dk . It was merely meant to illustrate,
for the benefit of those who have looked at http://www.pressphoto.dk and thought
that
their D1 images at any ISO, of actual subjects doing actual things, were
actually going to look like the images in that test, that they needn't
lose sleep.
Especially since it was causing me to lose sleep by being wakened at 6:30
AM to talk to a distraught Director of Photography!

Peter Inova sent me a helpful note this evening in response to what I
wrote. In reading it I became concerned that my central message was being
lost. Let me distill the 1000+ words on my site down to three points:

• The D1 will deliver images that appear to be noise free when shot at

low ISOs, if the images are properly exposed, as they should be in the
hands
of a professional photographer. That is, the noise is THERE, but it isn't
visible.

• The D1 gets noisy at higher ISOs. That is, the noise that's always

there becomes visible. It crosses the visible threshold and begins to
intrude upon
image content.

• On balance, the D1's overall noise it is subjectively no more or less

severe than a DCS 520/D2000/DCS 620 image at higher ISOs when
comparable images are toned for printing.

I repeat, I know the streaks are there, as I can do the same things as
everyone else in Photoshop to bring them out. Peter, you use the example
of the
sunset shot on my site to make your point. That illustrates my point too.
If the shadows are cranked in Photoshop then the streaks appear. But the
picture was shot to be a silhouette, and prints without any hint of the
streaking when the shadows are kept as dark as I wanted them to be to
keep the
mood in the photo.

The title of the previous message is D1 Noise Hoax Isn't Fake. Let me be
clear that I was not trying to say that the D1 doesn't plop noise into the
images it produces. It does, more than I expected given Nikon's marketing
of the camera prior to its release. I was above all trying to introduce an
idea that seemed altogether lost in the D1 noise discussion: does the
noise intrude into real photos printed on real paper and viewed by real
people? That's what matters most to me as a photographer. In that
context, I've been really happy with the low-ISO D1 images I've shot over
the past
several

 

[Phil Askey]

This afternoon I shot with a D1 at Keeble & Shucat Photography in Palo
Alto, CA. They have a demo unit there now (not for sale).

I can say with 100% certainty that the D1 DOES SUFFER from a horizontal
"banding" noise pattern EVEN AT ISO 200. I see it in every photo I took
which has darker areas. It's plainly visible viewing just the unaltered
file--no Photoshop Levels or any adjustment are necessary to see the
problem. I printed one example and it's also visible in the print (Epson
Stylus Photo 700).

I used the D1 in Program mode, ISO 200, JPEG Fine with a 35-70/f2.8 zoom.
Then I shot 55 pictures over about 20 minutes.

Though I think the casual viewer would probably not notice the problem,
it's plainly visible. Otherwise, the photos are stunning.

Can you post the images along with full exposure information please.

 

[Dave11861]

• On balance, the D1's overall noise it is subjectively no more or less severe than a DCS 520/D2000/DCS 620 image at higher ISOs when

comparable images are toned for printing.> > >

MY RESPONSE:

I think the above statement (admitedly taken out of context of his entire article) illustrates an important point. When an entire image (especially one in low light or where much shadow detail resides) taken at higher ISO's is toned for printing, the "imbedded" noise is not seen or obtrusive and compares favorably with other high end professional digital cameras. Occasionally though (maybe more so) printed or web image "levels" need to be increased (sometimes considerably) to bring out shadow detail or save an otherwise "lost image" and it is in these cases that the lines or streaks in the D1 are brought more into prominance, at least to some degree. It is under these circumstancs that some may be concerned. It relevancy or lack thereof will depend on the individual and his/hers intended use and manipuation of the images.

Dave

 

[Bill Harris]

This afternoon I shot with a D1 at Keeble & Shucat Photography in Palo
Alto, CA. They have a demo unit there now (not for sale).

I can say with 100% certainty that the D1 DOES SUFFER from a horizontal
"banding" noise pattern EVEN AT ISO 200. I see it in every photo I took
which has darker areas. It's plainly visible viewing just the unaltered
file--no Photoshop Levels or any adjustment are necessary to see the
problem. I printed one example and it's also visible in the print (Epson
Stylus Photo 700).

I used the D1 in Program mode, ISO 200, JPEG Fine with a 35-70/f2.8 zoom.
Then I shot 55 pictures over about 20 minutes.

Though I think the casual viewer would probably not notice the problem,
it's plainly visible. Otherwise, the photos are stunning.

Can you post the images along with full exposure information please.

I certainly agree that the banding does exist at ISO 800 & 1600, but I have yet to see it at 200 or 400 using any of the four available quality modes!

Concerning's Rob comments about the banding I believe that he is understating a serious flaw with the camera that Nikon MUST adressess. Even if Kristians test is flawed or not, is almost irrelavent! There are currently enough of us that have D1's and can produce this banding problem time and time again to show that there is a problem with the camera.

I have contacted and spoke with Frank from Nikon Technical support and he has confirmed that Nikon is aware of the problem and the the factory is looking into a fix, but could not comment on when one would be available.

I love the camera and firmly belive that this problem will either be fixed or minimized by Nikon, but as Rob states "The noise test at http://www.pressphoto.dk is but one example of information on the 'Net that has needlessly thrown both colleagues and clients into fits. To the nervous photo editor quoted above I offered the following thoughts:" is NOT AT ALL FAIR! Is it possible that Rob has become a marketing spokesman for Nikon?

If anything lets give credit where credit is disserved - lets remember that it was Kristain that pointed out this banding issues first and it was Kristian who pointed out the Frequently Asked Questions Web Site.

Bill

 

[Ed]

Are people really reading Mr. Galbraith's article or are they reading
what they want too believe. The bottom line is that the $5000 D1 displays
the same amount noise as the
$15000 Kodak DCS.

Lets set the record straight.

First, the Kodak DCS 620 isn't 15,000. I know I bought one this week, Its 7800 after the 500 Kodak rebate.

Second, pixel for pixel, I don't believe that the Kodak has the same amount of noise, my personal experience is that the D1 has more then the Kodak, but even if they have the same amount pixel for pixel, the noise created in a banding fashion as the D1 creates is almost impossible to get rid of while the noise created by the Kodak can be almost eliminated or greatly reduced with Camera Bits AP Filter.

I had been putting off buying a new Digital camera. After shooting with the NC2000 for a year and a half, I knew it was time, but I waited because I wanted to try the D1.

I had the opportunity to buy a D1 this week and I decided to buy a Kodak 620 instead. The banding problem I had experienced (not the stuff I saw on the net but with real images that were shot in the world I shoot in) occured in images with iso's as low as 400. As a photojournalist, I shoot lots of 400iso and an equal amount of 800 and I prefer to shoot available light whenever possible.

One other factor that made me decide to spend 2800 bucks extra was, when reviewing images on a D1, if something happens in front of you (as it will in the news business) you have to switch modes before you are ready to shoot, with the Kodak, you are always ready to shoot, just press the button.

The D1 is probably still available at the dealer, if anyone is looking, e-mail me. And if you don't shoot much in low light its a GREAT CAMERA. I wish I could have saved the 2800 bucks and bought one, its lighter and much more responsive then the 620.

Ed

 

[Erik13455]

Francis asks "What do I make of this?"

I am fascinated by the wording of his answer. From the syntax of his reply and the anonymous responder's title--"Nikon Corp., spokesman" (in the USA, we'd get at least a first name), this IS a Japanese individual who sent you this note. So it doesn't surprise me that the wording is indirect.

"Unfortunately, we cannot cancel the noise of the CCD AT THIS MOMENT."

Aha...it IS noise in the CCD -- which Nikon doesn't make, remember, and their deal with the manufacturer precludes saying too much about it. Now he has confirmed that Nikon is aware of the problem and is trying to fix it. (Their reputation requires that they do that, I think).

My guess is that Nikon is "in a flap" over this thing, and that they are wondering how to deal with the problem and do "damage control" at the same time.

My guess is that there are significant variations in quality of the CCD which "pushes the envelope" of the technology.

And my guess is that Phil Askey based his review on a (pre-released)version of the camera with one of the GOOD (or better) CCDs (perhaps the "pick of the litter" so to speak?

Erik

Thank you Mr. Galbraith!

So there you have it from the horse's mouth, as it were. The Nikon DOES
"plop noise into the picture." More than Nikon's "marketing hype" led any
of us to believe. If one reads their brochures on-line, Nikon makes a big
deal out of the camera's capabilities at very high ISO ratings. (I don't
think that too many folks are planning to shoot at 1600 in bright
sunlight! Do you?) So darkness is a given..and shouldn't surprise the
manufacturer when folks start to complain about noise. The "massaging"
with Photoshop is a given too, since it's much harder to get what one
wants from photographs taken in low light (e.g. Mr. Linneman's shots at
night)

Eric, here is a strange answer directly from Nikon Japan:

Dear Sir,

Thank you for your E-mail and the information of the
website which we should have checked out:

Unfortunately, we cannot cancel the noise of the CCD at
the moment.
However, we think that D1 in this condition is still useful
for many customers. Furthermore, the data includs which
we cannot know the test condition.
Generally, if you set up long shutter speed, the noise
increases.

Anyway, we suggest that you will check the other various
information and consider if you obtain one or not.

Please accept our sincere apologies for any inconvenience
caused by this problem.

Sincerely yours,
Nikon Corp. spokesman

What do you make of this?

Francis

 

[Peter iNova]

Francis asks "What do I make of this?"

I am fascinated by the wording of his answer. From the syntax of his
reply and the anonymous responder's title--"Nikon Corp., spokesman" (in
the USA, we'd get at least a first name), this IS a Japanese individual
who sent you this note. So it doesn't surprise me that the wording is
indirect.

"Unfortunately, we cannot cancel the noise of the CCD AT THIS MOMENT."

Aha...it IS noise in the CCD -- which Nikon doesn't make, remember, and
their deal with the manufacturer precludes saying too much about it. Now
he has confirmed that Nikon is aware of the problem and is trying to fix
it. (Their reputation requires that they do that, I think).

My guess is that Nikon is "in a flap" over this thing, and that they are
wondering how to deal with the problem and do "damage control" at the
same time.

My guess is that there are significant variations in quality of the CCD
which "pushes the envelope" of the technology.

And my guess is that Phil Askey based his review on a
(pre-released)version of the camera with one of the GOOD (or better) CCDs
(perhaps the "pick of the litter" so to speak?

Erik

The examples from Phil Askey's gallery show the bright side of the force. As Galbraith and others have noted, for 99+% of your images, the streaks are completely below image data level numbers. I tried, unsuccessfully, to elicit them from the darkest of Phil's shots and there was so much dynamic range to the images (a good thing) that the shadows were full of data way above noise and it would have taken much more extreme images to coax them to show.

You can also trim off the noise almost completely by simply lifting the left-most control in the histogram of the Photoshop Levels window by 4 or 8 units depending on the ISO you shot at.

I still think that there may be a relatively automatic way of trimming the noise from a master using an exotic reverse/mask technique in Photoshop by combining a lenscap-on "flaw frame" shot at the same ISO and exposure time as an extreme frame to be "rescued". The goal of that technique would be to produce an image that THEN could be lifted to ISO 2.6 bazillion without seeing the streaks. (Somebody post a URL to a pair of shots that fit this bill and I'll be glad to play around with them. ISO 400 would make an interesting test.)

For most images, it simply isn't a problem. For extreme images it can be.

I'm now tracking down the unexpected benefit to all this, the fact that the 950's can be "push processed" into some pretty interesting realms of "available darkness". The footbridge shots I posted earlier on this thread has me thinking that shooting for ISO 1600 or 3200 -like results isn't as silly as I might have supposed a week ago.

Other artifacts show up and in the example given, the ISO has jumped from 320 to 2560 (he said, estimating from the 8X increase) and is better looking in some ways than would have jumped out from a longer exposure (I think it was 1/4) where blown pixels start showing.

-iNova

 

[marcus13412]

Thank you so much!!!!! Here is what Nikon has said concerning this overpriced camera using a low-cost (Might as well say "in their circles" CHEAP CCD to justify 5000.00. THEY said basically, deal with it SUCKERS! This camera with flaw is a DAYLIGHT CAMERA!!!!!!!! Face it. And I guess until Nikon wants to admit more which will further ruin the already strained feelings with those that have purchased this promising yet in still ill-produced wondernag that can't be upgraded. BTW is Nikon going to forced to fix the CCD's already in the hands of the CHICKHENHEADS that had to have it NOW in spite of us more level-headed that said 'before I plop 7700.00 plus for this set-up, let me do some checking to be sure of my investment.' See, I too am on the waiting lists at 2 camera stores for the D1--- but not for long. I have decided to wait this out for a while and felt all along that this camera was just a trial balloon for another D (something) in the not-so-distant future. Nikon could rival how the initial D1 was a huge success and here is the next D(something) to fix what should have been the original in the first place!

Maybe you Hypes & Chickhenheads will listen and heed a voice crying in the wilderness on the web!
Marcus

Thank you Mr. Galbraith!

So there you have it from the horse's mouth, as it were. The Nikon DOES
"plop noise into the picture." More than Nikon's "marketing hype" led any
of us to believe. If one reads their brochures on-line, Nikon makes a big
deal out of the camera's capabilities at very high ISO ratings. (I don't
think that too many folks are planning to shoot at 1600 in bright
sunlight! Do you?) So darkness is a given..and shouldn't surprise the
manufacturer when folks start to complain about noise. The "massaging"
with Photoshop is a given too, since it's much harder to get what one
wants from photographs taken in low light (e.g. Mr. Linneman's shots at
night)

Eric, here is a strange answer directly from Nikon Japan:

Dear Sir,

Thank you for your E-mail and the information of the
website which we should have checked out:

Unfortunately, we cannot cancel the noise of the CCD at
the moment.
However, we think that D1 in this condition is still useful
for many customers. Furthermore, the data includs which
we cannot know the test condition.
Generally, if you set up long shutter speed, the noise
increases.

Anyway, we suggest that you will check the other various
information and consider if you obtain one or not.

Please accept our sincere apologies for any inconvenience
caused by this problem.

Sincerely yours,
Nikon Corp. spokesman

What do you make of this?

Francis

 

[Steven13520]

One other factor that made me decide to spend 2800 bucks extra was, when
reviewing images on a D1, if something happens in front of you (as it
will in the news business) you have to switch modes before you are ready
to shoot, with the Kodak, you are always ready to shoot, just press the
button.

Ed

Ed if thats the reason why you spent 2800 bucks more you could have saved lots of money by buying the D1. The D1 shoots immediatly in shooting mode while reviewing the last shot and any previous shots.No need to switch the dial to playback and back to shooting mode. Again some wrong information Just press the button and at 4.5 frames per second with instant response you'll think you have a film camera, unlike the relativly very slow 620 which shoots like a digital camera.

 

[lloyd chambers]

I don't have anywhere to post them, but I've emailed one unaltered sample to you, Phil.

This afternoon I shot with a D1 at Keeble & Shucat Photography in Palo
Alto, CA. They have a demo unit there now (not for sale).

I can say with 100% certainty that the D1 DOES SUFFER from a horizontal
"banding" noise pattern EVEN AT ISO 200. I see it in every photo I took
which has darker areas. It's plainly visible viewing just the unaltered
file--no Photoshop Levels or any adjustment are necessary to see the
problem. I printed one example and it's also visible in the print (Epson
Stylus Photo 700).

I used the D1 in Program mode, ISO 200, JPEG Fine with a 35-70/f2.8 zoom.
Then I shot 55 pictures over about 20 minutes.

Though I think the casual viewer would probably not notice the problem,
it's plainly visible. Otherwise, the photos are stunning.

Can you post the images along with full exposure information please.

 

[lloyd chambers]

My two cents:

The D1 has considerably less noise than my noisy C2500-L. But the type of noise in the

D1 is very disturbing because it forms horizontal streaks. Random noise is annoying,

non-random noise is a show-stopper. I see that as the crux of the issue, NOT how much

total noise there is. In my daylight shooting, I see horizontal streaks at ISO 200 in

shadows (I've sent Phil a sample). I was going to buy a D1, but this kind of noise is
just not acceptable.

I don't understand why Rob Galbraith, who is obviously very well informed, seems focused

on the quantitative aspect of the noise rather than the disturbing qualititative aspect.

Lloyd

Finally an intelligent test that tells the true story. Check out this web
site:
http://www.robgalbraith.com/diginews/index2

 

[Ed]

One other factor that made me decide to spend 2800 bucks extra was, when
reviewing images on a D1, if something happens in front of you (as it
will in the news business) you have to switch modes before you are ready
to shoot, with the Kodak, you are always ready to shoot, just press the
button.

Ed

Ed if thats the reason why you spent 2800 bucks more you could have saved
lots of money by buying the D1. The D1 shoots immediatly in shooting mode
while reviewing the last shot and any previous shots. No need to switch
the dial to playback and back to shooting mode. Again some wrong
information Just press the button and at 4.5 frames per second with
instant response you'll think you have a film camera, unlike the
relativly very slow 620 which shoots like a digital camera.

NO the reason was QUALITY, and from a quality perspective the 620 is a hands down winner. To me, my images are worth the extra $2800. Why do you think I bought the 620? My D1 is still sitting at the dealer! I could have bought one. Where I work the editors are very happy I didn't buy the D1. I suppose time will tell.

How long have you been shooting digital? I've been shooting with the NC2000 for almost 2 years and I know what to expect from digital. I know what I have to shoot day in and day out.

You should have seen the pictures that were shot at Madison Square Garden with the D1 - They are on the wire, just look for the banding lines and you'll know which ones they were. Given that I shoot at the garden 50 times a year - the D1 just isn't for me. Shea Stadium and Yankee Stadium didn't look much better.

You sound like you want to be RIGHT. I think that everyone has to make a decision based on the type of work they do. I know the work I do every day, I'm often in situations where I'm not in control of the light. Often its not ideal and lots of shadow.

Lets do this, If you have the time make a few 800ISO photos - under lets say a single light source, tungsten, and post them. I'll do the same and post them. Let people judge for themselves.

If you keep saying that there isn't a problem or that its acceptiable for a professional camera to have a bandng problem - how do you expect Nikon to address it?

Please try not to drag this to a personal level. I haven't questioned your judgement or motives. My motives are simple - to share the information I've seen with people on this site... And as for my digital experience and judgement, you know the people I work for, ask them about my knowledge of digital cameras...

Ed

 

[Brent Keith13673]

I'm now tracking down the unexpected benefit to all this, the fact that
the 950's can be "push processed" into some pretty interesting realms of
"available darkness". The footbridge shots I posted earlier on this
thread has me thinking that shooting for ISO 1600 or 3200 -like results
isn't as silly as I might have supposed a week ago.

Other artifacts show up and in the example given, the ISO has jumped from
320 to 2560 (he said, estimating from the 8X increase) and is better
looking in some ways than would have jumped out from a longer exposure (I
think it was 1/4) where blown pixels start showing.

-iNova

Peter,

Are you suggesting to reach for the -2.0 ev control before increasing the ISO on the CP950? I wonder if this has merit on the D1 as well.

Brent