D700/D300 cost

[jack_e]

This is just a question for thought.

 

It is my understanding that one of the cost prohibitive factors of producing sensors comes from the number of

sensors that fail the quality check with respect to the number of bad pixels. This being the case the greater

the number of pixels the greater the chances that you will have bad pixels on any one sensor and more sensors you

have to trash. That is one of the reasons why the high mega pixel cameras cost more. That means the high pixel

density FF sensor cameras will be significantly more expensive, e.g. the Cannon EOS-1Ds-Mark III, and probably

the D3X or what ever it will be called.

 

If this is true (please correct me if my logic is in error), shouldn’t the low pixel density FF 12 mp cameras be

very comparable in cost to manufacture as the DX 12 mp cameras? The number of pixel sensor sites will be the

same. Is the D700 which is has very similar components in most ways really more expensive to manufacture than

the D300. Or is the price difference more about marketing and demand?

[hugh_look]

Partially true, but the costs of any camera are far more than just the cost of components and assembly. The D700,

and indeed the whole FX system, will have to repay high costs for research and development, design, prototyping,

tooling costs and launch marketing costs. In the case of high-end cameras, these costs must be offset against a

smaller number of units sold as it's inherently a device with a smaller market.

 

Cost engineers and marketers then have to balance off the return on investment profile against the price that the

market will bear - which is also conditioned by what the competition does. This is a dynamic balance that changes

all the time. We can expect to see the price of the D700 (and perhaps other Nikon cameras using the basic FX

componentry) to fall in time (as the D300 has done) as the competition ups its own game with new products and/or

more aggressive pricing.

[Ed_Ingold]

The physical size of the sensor seems to have more impact on manufacturing cost than the number of pixels. With increasing size, you get fewer arrays per slice, and a greater chance of unconformities in the substrate.

[jack_e]

Hugh, I am curious how much research dollars went into the D700. It is my understanding the sensor and most of the internal working are right out the D3 and the body is basically a modified D300. I have not had my hands on a D700 yet so I may be incorrect. As far as prototyping, tooling and marketing goes, I would think that the cost would be very similar if not a little less since the actual manufacturing process would probably be similar (less cost in designing the process, since they have lessons learned and may even be able to modify the current structure instead of starting from scratch).

[jack_e]

Oh and one other thing. If the cost of the camera was more comparable you would possible see a higher number of units sold, especially since everyone seems to think that FF is superior. I still like DX for the higer pixel density/resolution, but that is a discussion that I know has been beaten to death :)

[mark_dalrymple1]

It's also possible that Nikon was planning on the D700 from the start, and is spreading the costs of R&D over both body

lines.

[jack_e]

That thought occured to me also

[dan_brown4]

No, it's not just the number of pixels, although that is a factor. The biggest issue is the size of the device. The probability of a flaw

squares to the length. In fact, in the mid 1990's many thought a full frame sensor would always be cost prohibitive in mass sales

products.

[chris_maryan]

The cost of any microchip is dominated by two things:

- physical size (full frame or DX)

- feature size (pixel density)

But which one is dominant varies a lot from one design to another, so there's no judging what costs more for Nikon.

 

Moreover, the price of a camera has relatively little to do with the price of the chip. It's part of the marketing equation, charge as much as you can for a given market segment (pro, amateur, etc.), while covering your production costs (chip costs + R&D), and competing with whoever else sells cameras in that range (Canon). So the chip cost is a relatively small part of it all, unless you have a very small market and have to cover chip+R&D costs for a small production run (medium format digital and other more specialized applications).

 

My understanding is that a typical D200 grade Nikon camera sells something like 70,000 units a year, which is well inside what is usually considered economical for volume chip production (10k+ total units). So I suspect the other marketing factors dominate.

[bgelfand]

Cost is only one factor that determines retail price. Mostly retail price is determined by Supply and Demand. Nikon's market research show they could get $3000 for the D700, in the U.S., and sell "X" units, so that is the price point, now. And they are selling out.

 

We will have to see what Canon comes out with as a 5D successor and at what price point. If it is close in both function and price, the D700 will stay at $3000. If the Canon has the same performance, but say a $2000 price, the D700 will drop.

[fran__ois_p._weill]

IMHO, Brooks is spot on the problem,

 

We should not forget manufacturing a camera and selling it is a business and as such their owners search for the

maximum profit to be extracted from what they produce.

 

The only link between production cost and retail price is the economic law the production cost should in no

occasion be superior or equal to the retail price as you won't make any profit.

 

However it seems to me a new evolution is manifesting itself in technology: during the first part of the digital

evolution the technical difficulties to produce large sensors which induced a high rejection rate were the main

limiting factors, first to the production of full sized sensor, then to its massive diffusion due to the cost of

each acceptable unit. This was the decisive factor for the widespread use of APS sized sensors on DSLR's (and

even smaller ones on compact cameras). The main way to improve sensors was to increase the pixel density while

the engineers tried and successed to fight the loss in dynamic range and the noise at high ISO by any mean. This

period will probably stay in the fledgling history of digital cameras as the "Pixel counters era" :-) ...

 

Nowadays, it seems the production engineers have found some way(s) to produce bigger sensor more economically and

to cut more and more the rejection rate of these bigger slice of silicium. It is IMHO clear this trend to return

to the tradtional formats of film photography has more immediate promises for performance increase (by using

bigger pixels the noise reduction problem and the increase in dynamic range seems easier, less costly in R&D, and

faster) while the increase in pixel density without reducing or precluding the increase in performance at high

ISO and maintaining or increasing the dynalmic range at the same time has taken the back seat.

 

As there are a lot of good reasons (including user's preference) to return to the old nominal formats, we can

only speculate on the extent of this trend.

 

But it looks like we are approaching the moment a given technology is mature enough to reach the point its

progress become slower and more difficult to obtain (so will cost more for each positive step) with the increase

in pixel density while maintaining or increasing the resistance to noise and dynamic range on the present sensor

technology.

 

I'm convinced this is not the case for the increase in size of the sensors. Recently even the compact cameras

were touched by this trend with the Sigma DP-1 (it has an APS-C sensor), in the medium format world, the increase

of pixel density is still gaining momentum but the demand for high ISO performance is less important than with a

small format camera and the prices are steadily decreasing. In the small format DSLR world, the best compromise

for the state of the art has been reached by Nikon with the D3 and then the D700 while the very high pixel count

Canon 1Ds Mk.III which cost a lot more is nowhere approaching the high ISO performance of the D3/D700 couple

while it remains unconvincing and rather uneconomical for medium format users who can find a better defintion for

relatively few more $ in buying a MF body for which most have already compatible lenses. A reason why I doubt

Nikon will ever issue a "D3x" in the 20mpx + category to complete the range of pro DSLR's of this generation.

 

I think Nikon engineers will wait for their next generation of DSLR where the progress made will allow for a

better defintion without sacrificing the high ISO performance and dynamic range to introduce a new sensor with a

higher pixel density. There is also another reason of economic nature: producing a unique model for a given

target will allow for economy of scale and consequently a decrease in price while maintaining the profit margin

per unit and broadening the potential market. Remember that Nikon is already using the same sensor unit on the D3

and the D700, which is a smart move as to cost amortizement.

 

The immediate future seems to me very predictable. the next generation of Nikon DSLR's will probably eliminate DX

format from the pro and semi-pro range, while the (I will call them D4 and D800 for sake of facility of reading)

new high end models D4 and D800 will share the same new sensor (may be 24mpx) with the same high ISO performance

the D3/D700 have (may be better but I doubt that). The DX lens range will stay as it is with (almost if at all)

no new lens. The next high end amateur grade body will use the D300 DX sensor and electronics. The FX lens range

will be developed probably with "revised" prime designs. Unless an unpredictable technological breakthrough

appears, the following generation will completely eliminate DX format to allow the concentration on the

production of FX lenses only (again economy of scale). The prices will probably decrease in the meantime (even in

constant money) to stabilize at the face value of the old film camera bodies for each level of performance. After

the total elimination of the DX range, we will probably see a concentration on fewer amateur DSLR models, may be

only one (the 21st century Nikkormat) with a sensor and electronics which will probably come nearer and nearer in

performance to the high end models but with a cropped menu and more automatism and probably no

retro-compatibility with manual lenses. The delay before obsolescence of a camera body will probably increase as

times goes by and the technology will reach its maturity.

 

FPW

[hugh_look]

I'm not defending Nikon's pricing, just noting what goes into it. The tooling costs will not be fully shared - the D700 is made in Japan, the D300 in Thailand. As for R&D, any product that demands new approaches to integrating existing componentry still carries a hefty cost - maybe more D than R, but still high.

 

However, as Brooks says and I said earlier, in the end it's competition at the price/features/value/reputation point that will have as much effect on the price as anything.

 

In terms of costs amortised across a given sales volume, then the company will always be faced with the decision about the right price for an optimal return: it's a black art with a lot written about it. Typically, the introduction of an innovation (such as FX for Nikon - and yes, Canon got there first but it's still an innovation for Nikon itself) will allow a period of high-margin selling to early adopters when you squeeze what you can out of the people who are prepared to pay a premium to have access to the innovation early, followed by a tricky decision about when to drop the price to get the next segment in. In practice, many firms wait for the competition to decide for them. So if Canon introduce an upgraded 5D in the autumn and at a price below the D700 then Nikon will have to think hard, and make a strategic decision about how many people might jump ship to Canon when it comes time for them to upgrade. Not easy.

 

Nikon also needs to keep the price at a level where the D700 does not compete too much with other Nikon products - i.e. doesn't cannibalise either D300 or D3 sales - too low and it could harm both, which would be unpleasant.

 

Sometimes you do get the "unbelievable value for money" launch where a category-killer sweeps everything before it on the basis of features/price ratio: the problem is, when the competition enters with its me-too product you often then have nowhere to go other than to eat into already thin margins.