Does Nikon listen?

Jim F. said:

I hope that if Nikon does go FF, they do it soon. Canon prices are
are too higt IMO ($8k for a 1DsMkII) and they need to come down.
They won't reduce prices without competition though

Click to expand...

You're assuming here that Nikon could produce a FF camera that is price competitive with Canon's two offerings. I don't think that's a safe assumption at all. A FF version of the D2X at US$10,000 wouldn't put any pressure on Canon's EOS-1Ds Mk II whatsoever. A FF version of the D200 at US$4500 wouldn't put much pressure on the EOS 5D.

It's possible to argue that Nikon should go ahead and build a US$10,000 FF D2X anyway, because some Nikon lens owners would pay it, if grudgingly. But that isn't a slam dunk argument in my mind (and here Thom and I may disagree). You could lose a lot of money designing and building a camera like that if you only ended up selling 5,000 or 10,000 of them over 2 or 3 years. And market failures, or obviously uncompetitive prices, detract from a brand's overall perceived "goodness", which might, at least in part, counteract whatever brand-image benefit you gained from matching the highest-end features of Canon.

And remember, too, that all business decisions, like investment decisions, must be made on a comparative ROI basis. Let's say it would cost Nikon $25 million dollars to develop a FF D2X and tool up to manufacture it. What would their return on that $25 million be? Is there some other product -- a $349 DSLR, the "D40", for example -- that would generate a better return on that $25 million investment?

And from what little evidence we have, that seems like the most plausible explanation for the way things are and have been on Nikon FF: Nikon can't find, or have built, a FF sensor that is both price and performance competitive at this point in time. They have, in oblique ways, essentially said as much. I think they are in something of a pickle on this issue, currently. Damned if they do, damned if they don't.

G Rash said:

Nothing objectionable on my monitor. Looks good to me.

Click to expand...

But that can't be. We've been told that Nikon's are useless at anything over 400 ASA. That picture must be polluted with noise, and somehow we missed it.

--

John in Minnetundra

--
Grimble Gromble

While we're comparing models in the top F line, any idea why Nikon remove the removable prism on F6, Thom? Modularity has been a unique feature on Nikon...
--
John

The F4 is the camera that started the slide from pro market dominance for Nikon. Canon had much better AF at the time (partly because of more ultrasonic motors in lenses) and that meant a lot of journalists and sports guys went to them.

While D200 is a fantastic camera (I love mine), capable of producing outstanding images, it's by no means perfect.

I own both a 5D and a D200. I prefer D200 as a camera - better build, faster, nicer "feel" in your hand.

But 5D is on another level as far as high ISO noise is concerned. ISO 3200 shots with 5D have about the same noise and actually a bit more detail than ISO 800 shots with D200. Couple that with a lack of fast wide angle lenses for DX, and you're getting FOUR STOPS advantage for low light wide angle shooting. Some of it comes from the FF factor (larger sensor = lower noise), some from Canon's better sensor technology, and a lot of it from the availability (or lack thereof) of fast wide lenses.

Four stops is huge. Enormous. For someone that wants to shoot low light wide angle, that's more than enough a reason to switch to Canon. Nikon HAS to do something about it.

I'm lucky enough to own both, but I would prefer much more if Nikon came out with a FF model. 17-35 2.8 and 28 1.4 would be awesome on FF. D200-like camera with a 5D-like sensor would be an instant hit.

Just to show I'm not trolling (or "full of it") here's a test I shot yesterday. D200 with a 50mm 1.8 lens at f/2.8, 5D with a 85mm 18 lens at 4.5 (same DOF). D200 at ISO800, 5D at ISO3200. 2/3 stops higher shutter speed with the 5D. Both shot in RAW, both converted using Adobe RAW. Zero sharpening, zero noise reduction. Same FOV, 5D shot resized to 10MP to match the D200.

Eamon Hickey said:

A FF version of the D2X at US$10,000
wouldn't put any pressure on Canon's EOS-1Ds Mk II whatsoever.

Click to expand...

Sure it would. It's not about cost, it's about where people came from to buy a 1DsII. Quite a few of those buyers were former Nikon users. Some were potential Nikon users but went the other way because they perceived that Canon had something that Nikon didn't. For what its worth, a FF D2x (i.e., use the same sensor technology) at US$10,000 would be a bargain, I think. Certainly well into the MF league at less than the MF price.

peripheralfocus said:

But that isn't a slam dunk argument in my mind
(and here Thom and I may disagree). You could lose a lot of money
designing and building a camera like that if you only ended up
selling 5,000 or 10,000 of them over 2 or 3 years.

Click to expand...

This would argue that Acura shouldn't have done the NSX, or Mercedes the Maybach, or Sony their very high-end TVs and projectors, etc. That's not quite how it works, and even looking at Nikon's history you can see that's not how it works. Nikon high-end development has always been about exposing new technologies that trickle down to the rest of the line in subsequent models. For an engineering-oriented firm, this is almost a must. Raise the bar. Raise it again. Raise it still higher.

The real question is just how stretched the Nikon engineering teams are. If they're so stretched that the next high-end system is compromised, I'd take that as a bad sign.

peripheralfocus said:

And market
failures, or obviously uncompetitive prices, detract from a brand's
overall perceived "goodness", which might, at least in part,
counteract whatever brand-image benefit you gained from matching
the highest-end features of Canon.

Click to expand...

Actually, one could argue that the 1DsII is just that for Canon. It certainly hasn't sold in high volume. It's by far the most uncompetitively priced DSLR on the market. And, no, it doesn't detract from the brand's perception at all, does it?

peripheralfocus said:

And remember, too, that all business decisions, like investment
decisions, must be made on a comparative ROI basis.

Click to expand...

Well, I'm one that argues that case a lot. But I also allow for some "brand building" and "distinction" in that. A company that prides themselves on being a leader has to lead every now and then.

peripheralfocus said:

Let's say it
would cost Nikon $25 million dollars to develop a FF D2X and tool
up to manufacture it.

Click to expand...

If we just take a D2x and add FF, it won't take US$25 million. It wouldn't take US$1 million is my guess. So what would that extra money be used for? New technologies, which, as I said, you eventually trickle down to your other products.

peripheralfocus said:

Nikon can't find, or have built, a FF sensor that is both price
and performance competitive at this point in time.

Click to expand...

Well, there may be some truth in that statement. Sony, perhaps, doesn't want to set up anything in their factories to do the multi-shot necessary for FF. Which would leave Nikon with LBCAST, Kodak, and Dalsa as their primary options. Each of those have problems as a candidate, though if we really have a 10mp LBCAST APS sensor coming that's competitive, as many of us suggest, then there isn't an issue any more, as that gets you to a 22mp FF with the same properties. The only real remaining issue is then with edges.

--
Thom Hogan
author, Nikon Field Guide & Nikon Flash Guide
editor, Nikon DSLR Report
author, Complete Guides: D50, D70, D100, D200, D1 series, D2h, D2x, S2 Pro
http://www.bythom.com

Just a quick and dirty Std Dev shows the D200 to be at 17.3 and the 5D to be 21.2 in one of the noisier areas you show. While both numbers are high, that is a measurable difference and in some areas the difference is higher. Moreover, there are also differences between the two cameras in color saturation and contrast that appear to be ISO caused. This just goes to show that people don't evaluate noise the same.

I personally wouldn't have said those examples look the same. There are things I like about both of them, but neither are they equal.

--
Thom Hogan
author, Nikon Field Guide & Nikon Flash Guide
editor, Nikon DSLR Report
author, Complete Guides: D50, D70, D100, D200, D1 series, D2h, D2x, S2 Pro
http://www.bythom.com

reck said:

D70/D70s complaint:
No vertical grip.
D80: Check.

Click to expand...

http://nikonusa.com/template.php?cat=1&grp=258&productNr=25345

--
There's no place like 127.0.0.1!

I'm not saying they are identical, but they look about the same to me, in terms of noise.

In other words, D200 at ISO 1600 looks a lot noisier than both, at ISO 400 looks a lot less noisy, 5D at ISO 6400 is a lot noisier, at ISO 800 is a lot less noisy than either of the two shown.

The 5D shot is a bit lighter overall, meaning that the RAW shot was pushed slightly more than necessary to match the D200 exposure (it seems that Canon underestimates their ISO rating by a bit more than 1/3 stop), exaggerating the shadow noise.

Also, the 5D shot has more detail, which means you can apply more noise reduction without losing the detail.

Anyway, this was just for illustration purposes. Take it for what it's worth. For me, it means that 5D outperforms the D200 by about 2 stops in high ISO noise, and I adjust my shooting and choose my tools accordingly.

Do your tests in broad daylight and the results will be very close. The 5D is overrated. Please stop quoting from Canon press releases.

YellowBullet said:

While D200 is a fantastic camera (I love mine), capable of
producing outstanding images, it's by no means perfect.

I own both a 5D and a D200. I prefer D200 as a camera - better
build, faster, nicer "feel" in your hand.

But 5D is on another level as far as high ISO noise is concerned.
ISO 3200 shots with 5D have about the same noise and actually a bit
more detail than ISO 800 shots with D200. Couple that with a lack
of fast wide angle lenses for DX, and you're getting FOUR STOPS
advantage for low light wide angle shooting. Some of it comes from
the FF factor (larger sensor = lower noise), some from Canon's
better sensor technology, and a lot of it from the availability (or
lack thereof) of fast wide lenses.

Four stops is huge. Enormous. For someone that wants to shoot low
light wide angle, that's more than enough a reason to switch to
Canon. Nikon HAS to do something about it.

I'm lucky enough to own both, but I would prefer much more if Nikon
came out with a FF model. 17-35 2.8 and 28 1.4 would be awesome on
FF. D200-like camera with a 5D-like sensor would be an instant hit.

Just to show I'm not trolling (or "full of it") here's a test I
shot yesterday. D200 with a 50mm 1.8 lens at f/2.8, 5D with a 85mm
18 lens at 4.5 (same DOF). D200 at ISO800, 5D at ISO3200. 2/3 stops
higher shutter speed with the 5D. Both shot in RAW, both converted
using Adobe RAW. Zero sharpening, zero noise reduction. Same FOV,
5D shot resized to 10MP to match the D200.

Click to expand...

Prints from D200 look better at 3200. Period.

Canon reinvented plastic.

Thibault D said:

Per Inge Oestmoen said:

Potentially more noise at high ISO,

This is rather more than a potential.

Click to expand...

It's not that large. Pictures up to 800 ISO won't be affected as
much on print. Looking at photos cropped to 100% on a computer
screen is causing a lot of lack of perspective! Even at 3200, the
pictures from a D200 give reasonnably good results on print with
some post-processing. And a large number of photographs never go
beyond 800 ISO anyway...
In film, people used to be able to work with a media that was very
degraded beyond 400 ISO. 3200 ISO was very tricky to get good
results with - and only in black&white. Yet, people managed to take
great pictures...
Some people never need to go beyond 100 ISO : some work in studio,
some just use a flash (and the Nikon flash system is one of the
best), some use a tripod... In other activities, high ISO noise is
not that annoying. Performance at ISO at 1600 and beyond is
important mostly for a niche market.

Besides, for Canon, how much is large pixel size and how much is
clever in-camera post-processing?
Fuji has shown that heavy amount of post-processing can make even a
tiny sensor look decent as far as noise is concerned...

Thibault D said:

Smaller photosites cannot give as good a dynamic range or as
favorable a signal/noise ratio. That is the rationale for "Full
Frame."

Click to expand...

Really?
Last I checked, cameras like the Fuji S3 dwarfed the 5D as far as
dynamic was concerned. Yet, the S3 uses a DX sensor...
And the difference between a regular DX and FF sensor is really
small compared to the difference between a regular FF sensor and a
Fuji high dynamic DX sensor.
This shows that the solution to get more dynamic is not geometric
(increase sensor size) but technological (upgrade the sensor
technology)...

Click to expand...

Thom Hogan said:

Thom Hogan said:

Nikon can't find, or have built, a FF sensor that is both price
and performance competitive at this point in time.

Click to expand...

Well, there may be some truth in that statement. Sony, perhaps,
doesn't want to set up anything in their factories to do the
multi-shot necessary for FF. Which would leave Nikon with LBCAST,
Kodak, and Dalsa as their primary options. Each of those have
problems as a candidate, though if we really have a 10mp LBCAST APS
sensor coming that's competitive, as many of us suggest, then there
isn't an issue any more, as that gets you to a 22mp FF with the
same properties. The only real remaining issue is then with edges.

Click to expand...

Are you saying that a 10 mp LBCAST APS ( C or H ?)
would be comparative to a 22mp FF CMOS or CCD ?

f128 said:

Do your tests in broad daylight and the results will be very close.
The 5D is overrated. Please stop quoting from Canon press releases.

Click to expand...

Who cares about high ISO performance in broad daylight?

I use high ISO only when I have to - and that's definitely not broad daylight.

5D overrated? Perhaps so. There's quite a few things I would like improved with it. But high ISO performance is definitely not overrated.

And no, I'm not quoting Canon. In fact, I'm not a big fan of Canon as a company. They've been riding to coattails of their (admittedly superior) sensor technology for a long time now. They need to redesign their wideangle lenses for FF. They need a body that can match D200 in ruggedness and build quality in a compact body (not the 1D series).

I own both the D200 and a 5D. I use them both. I did a test to see whether there really is a truth to the whole high ISO hoopla. Found out that there is. I adjust my shooting and the choice of tools accordingly.

You can ignore what you see or you can adjust your shooting to it. Your choice.

Many daylight tests show this is true. Canon needs to redisn WA lenses. Nikon is wiser. They didn't fall into the abyss.

YellowBullet said:

f128 said:

Do your tests in broad daylight and the results will be very close.
The 5D is overrated. Please stop quoting from Canon press releases.

Click to expand...

Who cares about high ISO performance in broad daylight?

I use high ISO only when I have to - and that's definitely not
broad daylight.

5D overrated? Perhaps so. There's quite a few things I would like
improved with it. But high ISO performance is definitely not
overrated.

And no, I'm not quoting Canon. In fact, I'm not a big fan of Canon
as a company. They've been riding to coattails of their (admittedly
superior) sensor technology for a long time now. They need to
redesign their wideangle lenses for FF. They need a body that can
match D200 in ruggedness and build quality in a compact body (not
the 1D series).

I own both the D200 and a 5D. I use them both. I did a test to see
whether there really is a truth to the whole high ISO hoopla. Found
out that there is. I adjust my shooting and the choice of tools
accordingly.

You can ignore what you see or you can adjust your shooting to it.
Your choice.

Click to expand...

Obviously, Nikon sells mostly cameras and film scanners while Canon sells cameras, printers, scanners, copiers...

A fairer fight : http://www.googlefight.com/index.php?lang=en_GB&word1=Canon+5D&word2=Nikon+D200

f128 said:

Many daylight tests show this is true.

Click to expand...

What is true? That high ISO is a lot better in daylight? We all knew that.

But nobody shoots high ISO in daylight. When it REALLY matters is low, artificial light.

f128 said:

Canon needs to redisn WA
lenses.

Click to expand...

I completely agree with that. 16-35, 17-40 and 24 1.4 need better extreme corner performance.

f128 said:

Nikon is wiser.

Click to expand...

I completely disagree with that. Nikon doesn't even HAVE fast wide angle lenses (good or bad) for DX. The fastest wide you can go with Nikon is 14mm 2.8 (I don't count the fisheye). That's TWO STOPS slower than Canon 24 1.4. Soft extreme corners or not, two stops will make or break your shot every time you need that speed.

Thom Hogan said:

Eamon Hickey said:

A FF version of the D2X at US$10,000
wouldn't put any pressure on Canon's EOS-1Ds Mk II whatsoever.

Click to expand...

Sure it would. It's not about cost, it's about where people came
from to buy a 1DsII. Quite a few of those buyers were former Nikon
users.

Click to expand...

Well, as I said, certainly some who switched to Canon would not have if Nikon had a $10,000 FF D2X (I'm assuming similar resolution to whatever top-end Canon exists concurrently). But some would have looked at the price differential and seen it as confirmation that Canon has a technology/manufacturing advantage that Nikon continues to trail, and gone ahead and switched to the faster horse. What percentage would make which judgment is the crux of the question here, and I don't think it's that easy to answer.

Thom Hogan said:

For what its worth, a FF D2x (i.e., use the same sensor technology)
at US$10,000 would be a bargain, I think.

Click to expand...

I spoke loosely. A $10,000 D2X at 20+ megapixels would indeed be a bargain -- then Nikon has a performance advantage to go along with their higher price. If that's possible, they should definitely do it. But my hypothetical presupposes that Nikon is offering the same performance for a higher price, not more performance for a higher price -- i.e. a 16-megapixel FF D2X for 15-20% more than the 1Ds Mk II. Or a 22mp D2X-FF for 15-20% more than the 22mp EOS-1Ds Mk III ArcTan(2/x + y).

Thom Hogan said:

This would argue that Acura shouldn't have done the NSX, or
Mercedes the Maybach, or Sony their very high-end TVs and
projectors, etc.

Click to expand...

How do you measure how much the NSX has helped Acura against how much it may have hurt them (money invested, engineers diverted from other projects etc.)? It's clear that some companies do believe that there can be an overall net brand-image gain from building very low volume, very high-end showcase products from which they don't intend to derive any profit. But that doesn't mean it's a good idea; companies do ill-advised stuff all the time.

I said the argument in favor of doing so is not a slam dunk, and I remain comfortable with that statement. Does Toyota, easily the world's most successful car company, do something similar to the NSX (via Lexus, of course), for example? (I don't know the answer; I'm asking.)

Thom Hogan said:

That's not quite how it works, and even looking at
Nikon's history you can see that's not how it works.

Click to expand...

Nikon's history is replete with (sometimes odd, sometimes cutting edge) prestige products, yes (often ones that I love personally -- anyone for a brand new SP rangefinder made in 2005?). But it's even more replete with really serious business mistakes. Someday I'll write a book. So again, the fact that they have done it before doesn't mean it's a good idea.

But no mainstream Nikon flagship SLR has ever been made with the idea that breaking even on it, or losing money, is okay. They have all been built very much with the intention of making sackfulls of money on them.

Thom Hogan said:

Actually, one could argue that the 1DsII is just that for Canon. It
certainly hasn't sold in high volume.
It's by far the most
uncompetitively priced DSLR on the market. And, no, it doesn't
detract from the brand's perception at all, does it?

Click to expand...

The 1Ds Mk II is priced high, yes, but I don't think you can say its price is uncompetitive. There is no competition (and even when the Kodak FF DSLRs were available, there was a very significant performance difference). If Nikon had a FF DSLR of equal performance level that sold for $6,000, then the 1Ds Mk II price would be uncompetitive. But there is no such camera.

Thom Hogan said:

If we just take a D2x and add FF, it won't take US$25 million. It
wouldn't take US$1 million is my guess.

Click to expand...

Well, I don't have a guess as to how much it would cost to develop such a camera. My US$25 million was, again, just a hypothetical to illustrate the concept of comparative ROI. Obviously, the cheaper it is to develop a FF D2-type camera, the better your chances of making a return on any given volume of unit sales, and the greater the scales tilt in favor of doing it. But I don't know how cheap or expensive it really would be. Nikon does know, more or less.

Thom Hogan said:

though if we really have a 10mp LBCAST APS
sensor coming that's competitive, as many of us suggest, then there
isn't an issue any more, as that gets you to a 22mp FF with the
same properties.

Click to expand...

If, indeed. Much depends on Nikon's ability to get better performance out of their sensor photosites. If -- there's that "if" again -- they can leap ahead of Canon on this score (and a 10mp APS-C sensor with overall image quality competitive with the EOS-1D Mk II (not the "s") would be a significant leap ahead of Canon, at least currently), then I think there's no doubt that a 20+ megapixel FF version would follow soon.

It could all be happening behind the scenes right now, for all I know. I am under no Nikon NDAs right now, so I can speak freely of my ignorance about such a possibility!

f128 said:

Do your tests in broad daylight and the results will be very close.
The 5D is overrated. Please stop quoting from Canon press releases.

Click to expand...

??? He demonstrated what he believes, not quoted from a press release.

Shooting @3200 in daylight kind of defeats the object doesn't it? If you've got daylight then (most of the time) you don't need 3200. The reason that shooting high ISO in artificial light is such a harsh test is down to the white balance of the light. You get very little blue spectrum light and need to push the blue channel hard to correct the WB (hence lots of noise).

You may or may not need the advantage that the 5D gives you in this situation, but trying to claim it doesn't exist is rather burying your head in the sand.

Eamon said:

peripheralfocus said:

If we just take a D2x and add FF, it won't take US$25 million. It
wouldn't take US$1 million is my guess.

Click to expand...

Click to expand...

To me $1 million seems too conservative for an estimate: is not even the annual salary of 10 engineers put together. But Even $25 millions may be not be a bad deal. Selling a very high end camera projects an image of a very high-tech company - something akin to be the Rolls Royce in photography. Whether we thinik that selling a $9,000 camera justifies this perception is not relevant. The fact is that it is happening. It would take a lot of money spent in advertising to get the same result without a high-end FF camera in Nikon's current line-up, and even after spending all that money Nikon would still not achieve the same results.

--
Thierry