Imaging Beyond Pixels: Where In The World Is Foveon?

This blog post references my feature article 'Imaging Beyond Pixels: Enhancing The Elementary Picture' in EDN's March 15, 2007 edition.

Astute readers may notice that Foveon wasn't covered much in my print article. Trust me, it wasn't for lack of trying. I attempted to contact the company on a number of occasions, both via email and telephone answering machine messages, with no response. I even pondered parking out in the company's lobby, but ultimately decided against this extreme measure; to do so, I decided, would give Foveon an unfair advantage over other companies ultimately mentioned (and not) in my writeup.

The company's silence unfortunately wasn't terribly surprising; Matt Miller had similar frustrations when attempting to contact them last year. But nonetheless it was disappointing, particularly because I felt that the article's premise of 'going beyond brute-force megapixel escalation' was well suited to the company's strategy and products. While X3 imagers don't capture as much image resolution as competitors' CMOS and CCD devices, they do include three photosensors (respectively tuned to the red, green and blue portions of the visible light spectrum) per image pixel, versus relying on Bayer filtering coupled with post-capture interpolation of pixels' missing spectrum segments.

In the absence of a direct statement from Foveon, what can we indirectly deduce about the company's current status and future fortunes from past history and partners' present accomplishments? Not much, unfortunately, and what we do know doesn't paint a particularly pretty picture. Through February of this year, Sigma had introduced two generations of DSLRs (the SD9 and subsequent SD10, the latter with improved low light performance and other minor tweaks) based on Foveon's 10.2 Msensor (3.4 Mpixel) CMOS imager. Neither product achieved a notable market share; out of fairness to Foveon, I should point out that the SD9 also marked Sigma's first entry into the DSLR business (closely following in the footsteps of the company's first silver halide-based SLRs, the equally underwhelming SA7 and SA9), whereas competitors such as Canon, Nikon, Olympus and Pentax were already in full market stride at the time of the SD9's unveiling.

Back in February of 2004, Polaroid announced its intentions to productize the x530 point-and-shoot digicam, based on a 4.5 Msensor (1.5 Mpixel) Foveon X3 imager. This camera was notable, among other factors, for its ability to output already-processed JPEG files; the SD9 and SD10 supported only the consumer-unfriendly (i.e. requiring post-processing in a PC) RAW format. Unfortunately, exhibiting a production pushout pattern that will be familiar to Foveon observers, the originally announced June shipping date came and went. The x530 finally entered production in February of 2005, and ceased production soon afterwards, a premature end-of-life that came complete with a 'technical issue'-induced recall.

Sigma's SD14 DSLR, based on a 14.1 Msensor (4.7 Mpixel) Foveon X3 imager, was announced at the September 2006 Photokina show. Like the ill-fated x530, it supports both RAW and JPEG file format options, and production was initially slated for November. Subsequently pushed out to early December, it was further delayed at that time due to an unspecified 'hardware issue', and the SD14 finally went into production on March 6, according to company representatives I spoke with at last week's PMA show. Sigma also plans a compact point-and-shoot based on the 14.1 Msensor X3 imager, called the DP1; unfortunately, a company spokesperson wouldn't give me anything more firm than an 'April or May' shipping estimate at PMA. Frankly, the fixed focal length is a showstopper in my mind, and I haven't even seen the pricetag yet.

Continue reading with 'Where In The World Is Foveon: The Silicon Eye'….