The new D2 will use the D1X body with a 7.8 megapixel Foveon Sensor.
This will equal the resolution of three times the Megapixel count or 23.4 Megapixel. The frame rate will be close to 15 frames per sec. This is all I can say at this point. Because of NDA I cannot reveal my source.
I'll probably hate myself in the morning for replying to this, but...
I doubt that this mysterious 7.8MP Foveon sensor exists, NDA or not. Here's why:
1) It would be far and away larger than any Foveon sensor to date. Their biggest, baddest, product right now in the X3 Pro 10M , which is 2268 x 1512 (3.4MP), and has a frame rate of 2 FPS. The link you refer to says the new camera will fire at 15 FPS. Perhaps this could be possible at reduced resolution using VPS, but I really doubt it a full res.
2) Foveon does not rate their chips in MP, and there is a reason. They want to play on the total number of photo sensors so the total pixel count does not appear to be small. Example: 7.8 million photo sensors would be a 2.6MP sensor (~1974 x ~1316). A 7.8MP sensor would be 23.4 million photo sensors, and this is most likely how Foveon would advertise it.
3) One of the largest design goals Foveon has right now is overcoming the X3's low usable ISO rating before noise becomes a problem. I believe the 10M can provide a max ISO rating of 400-600 before noise is an issue. I could be wrong.
IMO, it's doubtful Nikon would introduce a top of the line DSLR with a limited ISO range.
4) I have heard no scuttlebutt about such a sensor except over the past few days as part of the new D2. It's pretty difficult to keep such a secret through both the design and manufacturing stages.
To be offered in the D2, it would have to be ready for prime time. That means Foveon has finished the design, National Semiconductor fabed a few pieces for testing, Foveon and National Semiconductor go back and forth until chips are stable and meet all design specs, National Semiconductor gets the go ahead and ramps up for production, and then produces an initial production run.
Now Nikon is told about the chip and decides to test it in a design. They go through design cycles with several sensors and then decide on the Foveon. Nikon approves the final design and places an order for enough units to produce a small run for show display and initial reviews. Nikon then gears up for production to be ready in time for timely delivery after the product announcement.
IMO, this entire process would have to be accomplished within the past few months and without any leaks from Foveon, National Semiconductor, or Nikon employees. The reason for just a few months is based upon the item below, and 3-4 months is just not enough time to accomplish all of this IMO.
5) I spoke with Eric Zarakov (Vice President of Marketing for Foveon) a few months back and asked many questions regarding the X3 technology and what market it was targeting. It was made very clear that Foveon is primarily targeting the consumer product groups such as pocket digicams, and camcorders (ie, they want to sell millions verses thousands of chips). I also inquired about pending X3 products, and although I was not told any specifics, I got the distinct impression that current projects were for smaller sensor designs aimed at the consumer market. He did suggest many types of designs were in the early stages of development, although none of them were currently ready for production.
I hope I don't end up with egg on my face within the next day or so, but in all the time I have taken such stands I have only been wrong once that I can remember, so I guess twice wouldn't be so bad. I could live with that
