Focus on the future

Everyone must have tried it at some time or another. The old trick of gluing consecutive holiday photos together to produce a panoramic view of the vista from your hotel room might be good enough for amateurs, but for serious professional photographers, a medium-format panoramic camera is a must.

For years medium-format, fixed-lens cameras with lenses 6cm high x 17cm wide have been used to take single shots of landscapes to produce a true panoramic image without the curving ‘fisheye’ effect caused by other, rotating, panoramic lenses. Until now, for technical reasons, this photographic niche has remained a strictly film-based, non-digital affair.

However, Seitz — a small, family-run business in Switzerland — is preparing to shunt the medium-format panoramic industry out of the analogue age with the launch of the first digital panoramic 6x17 camera early this year. Its name, the Seitz 6x17, is unlikely to win many awards for originality, but this camera’s rather unassuming name belies its power.

Each 6x17 digital photo produces 160 million pixels, amounting to 1-2Gb of data a shot. Designed specifically for professional photographers at the high end of the market, the 6x17’s unique attributes mean that, just like its film-based forebears, it is ideally suited to landscape, architecture and group photography.
‘This is the first time a digital camera has gone beyond 6x6cm. Whereas others might have a resolution of 36mm x 24mm, ours will be 60mm x 170mm. It’s a big difference,’ claimed Urs Krebs, the camera’s project manager.

To produce the image, a specially-developed sensor is used to scan the back of the huge, large-format lens. Seitz decided the sensors that were being used for their existing range of rotating ‘Roundshot’ cameras were not up to the job of scanning a 6x17 lens and producing the definition required.

‘Previously our cameras used a Sony scanning sensor, but we were never really happy with the performance as it was slow and quite insensitive to light,’ said Krebs. ‘While it might have been cheaper to go for another off-the-shelf product there was really nothing available that fitted our needs.’

Seitz decided that to create a digital 6x17 camera with the required clarity of image, the only option was to develop a new scanning sensor from scratch.

Over the last two years it has worked closely with sensor and image specialist Dalsa to develop a custom-built scanning sensor — the first of its kind — specifically for photography. The sensor is a unique version of a time-delay integration (TDI) sensor, more commonly found in industry or in high-end scientific instruments aboard satellites. The resulting scanner is more than 100 times faster than any previous digital scanner, according to Krebs.

The sensor needed to be more than just extremely fast. It also had to have high resolution and perfect image quality. The reason Seitz went for an optimised TDI sensor is that it allows different parts of the sensor to be used for different jobs. One particular area is used to increase its sensitivity without any additional background “noise” interfering with the quality of the image. For Seitz, this was a major issue.

‘To increase sensitivity with normal digital cameras you increase the “gain” and make the capture more intense. This amplifies the signal, but along with more sensitivity you get what we call wild pixels, particularly in the darker parts of the image where the shade is wrong,’ said Krebs. ‘We don’t get that in our camera as instead of increasing the overall gain, we just use one section of the sensor to increase the sensitivity.’

After the image has been captured the next problem is to store it, a technical obstacle that Seitz developers were not sure could be overcome. Each 160m pixel image produced by the camera results in up to 2Gb of data that must be transferred and stored.

Krebs said when the project began, the processing capability of existing PCs was not fast enough to deal with such large chunks of data. Advances in processors and other hardware have now made Seitz’s 6x17 dream possible.

‘Actually what we can capture is still short of peak performance,’ admitted Krebs. ‘But this is something for the future. You could say that we have developed the sensor of the future but it’s a matter of time until computer technology catches up.’