Does Fuji need Nikon? S4 with D80 body...

Thom Hogan said:

Joseph S. Wisniewski said:

Because Nikon hasn't made a single "niche" camera yet, and Fuji
hasn't made a single mainstream DSLR sensor yet.

Click to expand...

The underlying assumption here is that Fujfilm doesn't want to sell
a mainstream DSLR

Click to expand...

Oh, I believe that Fuji wants to sell a mainstream camera, they just can't figure out how to do it. They know how to make and market a niche camera: indeed, the S3's biggest niche is wedding shooters, a simple shift for Fuji's medium format marketing machine. A mainstream DSLR means they have to figure out a business model. How to fund really high $ R&D, how long to repay the investment, how much profit can they take from the camera, considering they don't have a lens line yet...

Thom Hogan said:

and Nikon doesn't want to sell a niche one. That
doesn't sound right to me (if the D2hs isn't a niche camera, I
don't know what one is, BTW).

Click to expand...

The word you used is "want". I bet Nikon really doesn't want to sell that particular niche camera. R&D was high, quantities are low, it's a hard sell. But it's one you have to sell. You were using Detroit analogies earlier, so I'll toss one back at you. The Mustang has a decades history of losing money for Ford. It's a rare year when sales offset engineering and marketing on that car. But the cost of not having something like Mustang in the line is a drop in the public's perception of the "coolness" of the Ford Motor Company, and sales drop across the entire line. This was made obvious when J. Mayes redesigned the Musting to look like a Toyota, and again when Jacques Nassar announced they were dropping the Mustang.

Thom Hogan said:

I think you're seeing marketing
making the best light of what they've got. Or marketing doing what
they know how to do.

Click to expand...

Exactly.

Thom Hogan said:

Thom Hogan said:

Because Fuji's claims for their 45 degree sensor are held by most
people outside this forum to be bordering on fraud, and Nikon
doesn't need the kind of trouble getting involved in that would
bring.

Click to expand...

That's a strong claim. If believed, then all sensor claims are
bordering on fraud. Do we really have 10.2 million luminosity
points in the D200? Nope.

Click to expand...

But we do have 10.2 million points of some sort. Fuji was marketing 3 million point cameras as 6 million points "output" for years.

Thom Hogan said:

And I believe that Nikon is pretty much
solely motivated on producing the highest IQ they can at a
competitive price that keeps or grows their share with an
acceptable GPM.

Click to expand...

Yup.

Thom Hogan said:

Thom Hogan said:

Because Sony has proved that their corporate management will not
let Sony's camera division coerce Sony's sensor division into
conduct that would give Sony cameras an unfair advantage over
Nikon, Canon, Casio, etc. while Fuji has no such established track
record.

Click to expand...

In business you're only as good as your last management change ; )
The sensor hardball hasn't started yet. It will.

Click to expand...

Interesting observation. I see it going entirely the other way, with Sony eventually spinning microelectronics off into a separate entity. At least I think they'll be smart enough to keep it named Sony something or other, and not make an "Infinion" or "Freescale" kind of muck up.

--
The Pistons led the NBA, and lost in the playoffs.
The Red Wings led the NHL, and lost in the playoffs.

It's up to the Tigers now...
Leading the league, and going all the way!

Ciao!

Joe

http://www.swissarmyfork.com

Zarathustra said:

When it comes to reverse engineering in camera related gear Sigma
is king of the hill.

Click to expand...

No. They may be kings of reverse engineering lenses, but their cameras look like a throwback to 1985. The Sigma SA-9 that their own SD9 and SD10 DSLRs, and the Kodak SLR/c, is functionally equivalent to the Nikon 8008. Single zone autofocus that is pretty slow in bright light and fails in low light. Simple 8 zone matrix metering, without spot meters.

Sigma did exactly the same thing that Fuji did. They got hold of a hot sensor, took a clunky film SLR, and added a lot of big stuff to it to make a large, awkward DSLR that was an ergonomic nightmare. And just like the Fuji users, Sigma's small cult talks about image quality to the exclusion of all else.

Zarathustra said:

What if Sigma can clone a D80 body for Fuji?
If Sigma can't do it, I really doubt Fuji will do it better alone.

Click to expand...

Well, whoever does it, the design and manufacturer of a D80 or D200 quality body is going to be a job running into 10's of millions of dollars. So it's not going to happen, unless Fuji upper management sees a billion in potential sales to pay for such expensive R&D.

If Fuji does make the decision to begin a $50M R&D effort, I can see them wanting to keep it in house, rather than sending $50M to Sigma.

--
The Pistons led the NBA, and lost in the playoffs.
The Red Wings led the NHL, and lost in the playoffs.

It's up to the Tigers now...
Leading the league, and going all the way!

Ciao!

Joe

http://www.swissarmyfork.com

Thom Hogan said:

Not in the sense you're thinking. I have no problems with Nikon's
UI (or Fujifilm's for the most part). The problem I see is a
business case one. What you propose is only one small step removed
from slapping a logo on someone else's product. Look what that did
to Detroit...it simply isn't a long term viable business strategy.

Click to expand...

But it's one being embraced more and more by the camera companies. How much of Nikon's and Canon's lower end lens lines are being manufactured for them by Tamron. What about Pentax and Tokina, or Pentax and Samsung? And how many point and shoots are being manufactured for Nikon, Canon, Oly, Kodak, etc.

Thom Hogan said:

My feeling is that Fujifilm either needs to put up or shut up at
this point. As in put in the R&D to be competitive or leave the
DSLR market and just sell sensors to others that want them.

Click to expand...

That's assuming someone actually does "want them".

--
The Pistons led the NBA, and lost in the playoffs.
The Red Wings led the NHL, and lost in the playoffs.

It's up to the Tigers now...
Leading the league, and going all the way!

Ciao!

Joe

http://www.swissarmyfork.com

Joseph S. Wisniewski said:

The "demosaic ASIC" almost certainly doesn't exist. As I mentioned
in another post, it's not needed with Nikon's choice of processor.

Click to expand...

Exactly what embedded SPARC do you think can demosaic, apply other processing (sharpening, etc) and do a jpeg conversion in 0.2s? (e.g. sustain a rate of 5FPS.) That's 50 million pixels per second. Even if you are running a 500 MHz SPARC (iffy with the power/heat budget), thats only 10 ops per pixel. Even if they are doing the simple 8 neighbor algorthm you suggest than means that are doing at least 8x operations per pixel in only 10 cycles even if you are just adding them together. That's not the kind of pipeline you find in an general purpose CPU

Thom's right: they are most likely using a SPARC core in an ASIC with some dedicated DSP-like hardware. This type of processing is ideally suited for dedicated parallel pipelines and poorly suited to a general purpose CPU like SPARC.

--
Erik

Zarathustra said:

Apparently Fuji also claims that by giving the photodiodes a
octagonal shape the can increase the fill factor, meaning larger
sensitive area, hence lower noise.

Click to expand...

I believe that's the primary reason: a somewhat larger photosensitive area than other CCD technologies, all else being equal. Thus the SN of photons to underlying electron noise is higher.

--
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

Joseph S. Wisniewski said:

Oh, I believe that Fuji wants to sell a mainstream camera, they
just can't figure out how to do it.

Click to expand...

If that's true, then Nikon would be out of their minds to help them do it, wouldn't they? Essentially, there's an Ogden Nash-like "can't get there from here" aspect to what I think you're suggesting.

Joseph S Wisniewski said:

The word you used is "want". I bet Nikon really doesn't want to
sell that particular niche camera.

Click to expand...

Actually, unless something drastic has changed since the last time I talked to a Nikon executive directly, I'll take that bet. Beyond the basic "make as much money in the camera business as you can," there's still plenty of ego about making the "right" products, regardless of volume. Things like the FM3a don't appear because they're going to goose the numbers, they appear because there's pride in the details.

Joseph S Wisniewski said:

You were
using Detroit analogies earlier, so I'll toss one back at you. The
Mustang has a decades history of losing money for Ford. It's a rare
year when sales offset engineering and marketing on that car. But
the cost of not having something like Mustang in the line is a drop
in the public's perception of the "coolness" of the Ford Motor
Company, and sales drop across the entire line.

Click to expand...

Exactly my point, actually. I believe YOU were the one that said Nikon didn't want to sell niche cameras. I believe they do. But it has to be the right niche, right camera.

Joseph S Wisniewski said:

But we do have 10.2 million points of some sort. Fuji was marketing
3 million point cameras as 6 million points "output" for years.

Click to expand...

Sorry, but Fujifilm cameras were actually putting out 6 million points of something (and now 12). Interpolation happens in both cases. In one case you're interpolating two sets of data from a third (Bayer), in the other you're interpolating three sets of data from three (and Fujifilm is somehow getting an advantage there that others can't, so they know something about that interpolation that's better than a bicubic enlargement).

Joseph S Wisniewski said:

Joseph S Wisniewski said:

In business you're only as good as your last management change ; )
The sensor hardball hasn't started yet. It will.

Click to expand...

Interesting observation. I see it going entirely the other way,
with Sony eventually spinning microelectronics off into a separate
entity.

Click to expand...

That's a distinct possibility. I don't know the internal politics at Sony well enough to make any predictions on that, though.

--
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

Joseph S. Wisniewski said:

Starting with the assumption that the "demosaic ASIC" even exists.
I dispute that.

Click to expand...

You might want to study the insides of a Nikon DSLR and try to assign functions. The chip in question has Nikon printed on it and in brochures is referred to as an "image processing ASIC" when it is identified. Did you really think that Nikon didn't have a Digic?

Joseph S Wisniewski said:

Joseph S Wisniewski said:

If they abandon their four button/LCD UI, sure. If they don't,
they've got way too many buttons on the camera.

Click to expand...

Do they? It works fine for Nikon.

Click to expand...

I was referring to adding Nikon's button inventory to Fujifilm's. I've already got--let's see what's handy, hmm, my D2x--thirteen buttons on the back of my camera, I don't need or want four more.

Joseph S Wisniewski said:

Joseph S Wisniewski said:

What's the business case for Fujifilm? None.

Click to expand...

The sensor division gets to sell tens of thousands of profitable
sensors. The camera division still makes some profit on the
cameras, because the Fuji niche markets, like wedding shooters, are
still willing to pay more for Fuji color and dynamic range.

Click to expand...

And the business case for selling sensors to someone like Nikon and not making cameras, I think, makes more money. As in millions of profitable sensors without any camera R&D.

You're also making an assumption that I don't know to be true. The way Fujifilm has been moving the remaining S3 inventory, I'm not sure they're making a profit on it. That's another problem with the current approach: they can get a premium early in the camera's production (assuming it has something state-of-the-art, like DR was for the S3 Pro), but that degrades very rapidly because they're behind camera-wise.

Joseph S Wisniewski said:

But Nikon is still selling the bodies, just to Fuji, not to the
normal channel.

Click to expand...

As long as the body sale is incremental to what else Nikon is doing, I don't think they have a problem with it. If such sales started to erode Nikon's DSLR sales to customers, Nikon would have a problem, since they'd be getting less GPM.

--
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

Joseph S. Wisniewski said:

But it's one being embraced more and more by the camera companies.
How much of Nikon's and Canon's lower end lens lines are being
manufactured for them by Tamron.

Click to expand...

There's a little bit of difference there, in that I don't believe that Tamron is designing all those products.

Joseph S Wisniewski said:

What about Pentax and Tokina, or
Pentax and Samsung? And how many point and shoots are being
manufactured for Nikon, Canon, Oly, Kodak, etc.

Click to expand...

"Slap a label on it" is different than true "OEMing." And the more commodity like a product is, the more likely that minor changes and brand can impart different sales. But as you move up in price and complexity, such co-producing strategies get very problematic.

I also wonder whether some of the companies aren't revisiting their strategies of sharing. A couple of higher end features that have popped up across a broader range of camera companies than you'd expect are not the result of R&D but of cross patent licenses that leaked technology between maker and makee.

Joseph S Wisniewski said:

Joseph S Wisniewski said:

My feeling is that Fujifilm either needs to put up or shut up at
this point. As in put in the R&D to be competitive or leave the
DSLR market and just sell sensors to others that want them.

Click to expand...

That's assuming someone actually does "want them".

Click to expand...

But YOU must want them, because you're advocating sticking them into a body that otherwise has no Fujifilm technology in it! If customers demand something, what comany in its right mind would deny that? I think the SuperCCD has proven itself both in the digicam and DSLR market as having highly desirable features. I think that if Fufifilm wanted to, it could sell the sensor. The questions are two: do they want to, and would they sell it at a competitive price? Yes to both those, and then any DSLR from them is executing the wrong strategy, IMHO.

--
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

Joseph S. Wisniewski said:

Well, whoever does it, the design and manufacturer of a D80 or D200
quality body is going to be a job running into 10's of millions of
dollars. So it's not going to happen, unless Fuji upper management
sees a billion in potential sales to pay for such expensive R&D.

If Fuji does make the decision to begin a $50M R&D effort, I can
see them wanting to keep it in house, rather than sending $50M to
Sigma.

Click to expand...

I don't think they're going to make a D80 or D200 quality body. It would be much better for them to build a whole body themselves and upsize the sensors they use for SLR-like cameras. They should make an EVIL. All they'll have to steal from Nikon is the mount and possibly the interface for I-TTL flash.

They would have a camera with bottom-of-the-barrel performance in some aspects, like auto-focus and speed. OTOH they would have movie-mode, best dynamic range, a lighter camera than the rest, absolutely no camera shake due to mirror slap, easy low-light composing and all the other goodies that the EVIL (or ELVIS) brings.

They could give the camera a DVI or VGA connector so you could use an external monitor (a big one) for composing and focusing. I'd bet studio photographers would love that.

A camera like that could fill several niches. Nikon wouldn't mind too much - it sells Nikon lenses, without really killing the market for the D80 or D200.

"The D80 is camera-wise not that great, especially the new shutter leaves a lot to be desired."

On what are you basing that? The camera is not even out there in the real world yet as far as I am aware? And for once its a proper shutter - doesn't switch to becoming "electronic" at high speeds when the blades won't run fast enough.

I am possibly missing something - tell me more

Thom Hogan said:

Thom Hogan said:

Thom Hogan said:

My feeling is that Fujifilm either needs to put up or shut up at
this point. As in put in the R&D to be competitive or leave the
DSLR market and just sell sensors to others that want them.

Click to expand...

That's assuming someone actually does "want them".

Click to expand...

But YOU must want them, because you're advocating sticking them
into a body that otherwise has no Fujifilm technology in it! If
customers demand something, what comany in its right mind would
deny that? I think the SuperCCD has proven itself both in the
digicam and DSLR market as having highly desirable features.

Click to expand...

Yes. I want it. I'd love to have an Olympus C7070 with the sensor that the FujiFilm F30 has. And I'd love to see the long overdue E-3 (or whatever it might be called) show up with a 10-12Mp SuperCCD. That is just the type of thing a 4/3 camera could use to equalize the moderate sensor size difference betwee 4/3 and APS-C.

Of course, they could probably sell a lot more sensors if they could get Nikon as a customer instead.

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

Jay Turberville said:

Yes. I want it. I'd love to have an Olympus C7070 with the sensor
that the FujiFilm F30 has. And I'd love to see the long overdue E-3
(or whatever it might be called) show up with a 10-12Mp SuperCCD.
That is just the type of thing a 4/3 camera could use to equalize
the moderate sensor size difference betwee 4/3 and APS-C.

Click to expand...

And of course that's been my point for a while now. The current Fujifilm strategy in digital is losing market share, losing money, and has resulted in a lot of interesting promise (due to the sensors) that gets undelivered in cameras (F30, S3 Pro were at or near the head of the IQ class in their category when introduced, but have significant "camera shortcomings"). As proof of concept for the sensor capabilities, Fujifilm has done a decent job. As usable all-around cameras that are competitive, they fail. Put any of the Fujifilm sensors in current state-of-the-art body designs and we get better than what Fujifilm has delivered. I think that was Joe's point, too, though we got sidetracked into the specifics of Fujifilm attempting to do that with someone else's body.

The debate internally (and externally as seen in these threads) has to be:

A. Execute better, state-of-the-art cameras with our state-of-the-art sensors.
B. Get out of the camera business.

Fujifilm appears to be going to attempt A without the level of engineering commitment you see in Canon, Nikon, Sony, Olympus, et.al. Won't fly, IMHO. Given the recent corporate statements out of Fujifilm, B is a distinct possibility if A fails, but it doesn't currently appear that they would add "and sell sensors to others" to B. One of the reasons I'm so vocal in this issue is that I believe Fujifilm is setting themselves up for failure in photographic digital imaging no matter which decision they make. I think that would be a shame. As you and others have indicated: the Fujifilm sensor is gold. Arguably capable of key things that photographers want (DR, pleasing color as opposed to accurate, well-tuned AA, low noise, etc.). My personal take is that they should choose one of these:

A. Devote adequate R&D resources to a full line of digicam and DSLR bodies with state-of-the-art technology (AF, metering, IS, etc.). They are closer to this with the digicams than they are with the DSLRs. And I think to do it right, they'd need to start at the right spot: license the F6 and produce a FF SuperCCD DSLR. From there, work downwards, but always execute at the higher level than existing product.*

B. Match Sony's prices in sensors and license to one and all while exiting the camera business.

• Fujifilm isn't going to win market share wars in the near future. Trying to produce 2m DSLRs from their current position is near impossible. Thus, they need to start with Niche Killers and then push that technology downwards. FWIW, that's exactly what Nikon did with the D1.

--
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

Thom Hogan said:

B. Match Sony's prices in sensors and license to one and all while exiting the camera business.

Click to expand...

The problem with this strategy....is the opressive capital costs involved in moving from a second or even third tier semiconductor manufacturing company into the forefront of manufacturing. The capital investment required to really throw a hat into that arena is significantly higher than the cost of building the camera themselves.

Perhaps they can partner with a current market leader but they have shown little tendancy to partner with anyone.

Does Nikon need Fuji?

The answer is Yes!

Walter S Matthews said:

The problem with this strategy....is the opressive capital costs
involved in moving from a second or even third tier semiconductor
manufacturing company into the forefront of manufacturing. The
capital investment required to really throw a hat into that arena
is significantly higher than the cost of building the camera
themselves.

Click to expand...

I'd say that ANY real strategy for remaining a player in digital imaging for photography would require significant capital investment. To think otherwise is silly. If Fujifilm isn't interested in investing in photography, then it should get out, period.

But that introduces yet another problem for Fujifilm: one of the reasons why they have proprietary sensors is that they're used in other Fujifilm businesses. The SuperCCD appears in some of their medical equipment, for example. Thus, they can't simply stop making SuperCCDs or doing R&D in that area without impacting other businesses they're in. I was taught that when you have unique underlying technology you need to spread and maximize it across products and segments or you're not getting all the bang for the buck your shareholders should be demanding. Fujifilm seems to be in some sort of middle ground.

--
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

mmmmmmmm said:

Does Nikon need Fuji?

The answer is Yes!

Click to expand...

Olympus needs them more than Nikon does.

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