NikonWeb.com

Interview with Kiyoshige Shibazaki, Nikon General Manager, 1st Development Department, Development Management Department, Imaging Company (whew!):

"I guess that it's now safe to reveal that the D1 image sensor, with specifications noting a pixel count of 2.7-million pixels, actually had a pixel count of 10.8-million pixels.

The technical reason for an actual pixel count four times greater than that indicated publicly lies in the need to achieve high sensitivity and a good signal-to-noise ratio.

Unlike current cameras, for which final pixel counts account for individual pixels, we had to include multiple pixels in each pixel unit with the D1. In short, our development of an image sensor with so many pixels at such an early stage in the history of digital cameras indicates the importance placed on SLR camera development at Nikon."

http://imaging.nikon.com/products/imagi ... /index.htm

Jarle

Its a very interesting interview, I suggest all who are interested in digital camera history read it.
from what I understand SHIBAZAKI-SAN is claiming that each of the D1's Pixels consist of 4 smaller pick-ups housed within the sensor housing, hence the claim of 10.8megapixels. Its like Sigma claiming their Sd-10s' Fevron 3.5 megapixel image sensor is 10.5 megapixels because each pixel consists of 3 sensors for different wavelengths or kinda like Fuji claiming their 3 megapixel S2 was a 6 megapixel equivalant because of the way the pick-ups were positioned within the sensor array.
I am not an engineer but from what I understand it doesn't matter if you have 1,2,3 or 4 pick-ups housed within the same pixel because ultimatley they are reporting data as a single pixel...I like to hear Stan's comments on this issue....What is the definition of a pixel??
I commend Nikon in witholding such claims until D1 became obsolete.
Kiu

The actual outcome of this interview is indeed amazing! I do not understand why Nikon has not made public this information at the presentation of the D1 mid 1999. Given the strange and absurd pixel-race among/between various sensor and camera producers (and journalists) it would be interesting to hear what the real sensors are alike. Nikon only used the LBcast-sensor once; why? I know a bit about the structure of the CMOS-sensor and the CCD-sensor, but why didn't Nikon convinced the world what their sensors really are. Is it Nikon's inability to execute their Consumer Relations Management (anounced 2 years ago) or are they just technicians without a decent knowledge about PR?
My love for my D1 is increasing.............
warmest regards from Holland.
Nico

Hi,

This was all speculated about 'in the day' after a few of us took a look at the sensor itself under microscopes. It looked like there were far more photosites incorporated into the structure than there were pixels.

The sensor itself is a Sony device, and it made a lot of sense to me that Sony would have leveraged their video cam sensor base technology (which used a far smaller photosite size than Nikon was specifying for the D1) when it came to producing the physically larger sensor for the D1. So, the thought at the time was that they were probably using four photosites ganged together to make up each pixel.

Not that any of us knew for sure, mind you. Neither Nikon nor Sony were saying anything about it.

Not that it matters, anyway. Like the Foveon scheme, what matters is *not* the photosite count, but the picture element count - what we see on our PC screens post-processing of the sensor data (regardless if it's us doing the PP, or the camera doing it).

So, coming out me and stating that the D1 series has a 10 MP imager, when that's not what one gets in the photo itself would simply lead to flame wars - just like the ones that popped up when some folks tried to use the 3x math on the Foveon device.

What *is* interesting is how, by using a different mapping scheme and color filter layer, they made the 4 photosite per pixel (PPP) device into a 2 PPP one to make the D1x/D1h pair out of the original D1 and extend the life of the basic design.

*That* was the clever part.

Also, we have to keep in mind that there was no pixel race at the time the D1 came out. The only other player at that point was Kodak, with the 2MP 520 and the 6MP 560 (which Canon was peddling as the D2000 and D6000), and they weren't even in the same market segment given that the 520 was $10k and the 560 was $25k, while the D1 was $5k.

Oh, and, yes, Kodak also had the Nikon based variants, the 620 and 660 at the time, but they also were not competition to the D1. Besides Nikon got some bucks from the sale of each of those models to boot, given that Kodak had to get the F5 chassis from Nikon.

There was no real competition to the D1 until Canon came out with the D30 some, what?, two years later. Even then, we had 2.75MP vs. 3MP., which is hardly a pixel race is it?

So, I can see why Nikon would not want to mention the 4 photosite per pixel trick at the time, when it would have just led to confusion about how digital realted to analog - as if there weren't enough confusion as it was.

Stan

_________________
Amateur Photographer
Professional Electronics Development Engineer