San Juan Capistrano, CA-based Semiconductor Technology Associates (STA) has designed the world’s highest-resolution digital camera chip, capable of holding an image composed of more than 111 million pixels. By comparison, the best consumer cameras take shots of 12 to 16 million pixels, and an average computer monitor offers about one million pixels.
The imaging chip, which is a charge-coupled device (CCD), was designed for use in telescope cameras that map stars and ever-moving objects in the solar system, says Richard Bredthauer, STA’s president. But this large-scale chip – it measures four inches square – could be useful in more fields than just astronomy, he says, including high-resolution microscopic images of proteins, military surveillance applications, and even civilian mapping projects that require detailed aerial photography.
Currently, most large-scale digital photographs are either taken by arrays of smaller CCDs connected together, or created by stitching together hundreds of images. With arrays, the quality of each part of the picture can vary because each CCD might have been manufactured under slightly different circumstances. Additionally, assembling an array of multiple CCDs can be expensive and complicated. With the stitching method, the lighting in the pictures changes over time, since it can take from minutes to hours to collect enough pictures for a large-scale panorama. The 100-megapixel CCD (a mega pixel is one million pixels) doesn’t have these drawbacks, and could also potentially be cheaper to manufacture, since a single chip could give the same resolution as many arrayed chips, says Robert Groulx, CCD product manager at Dalsa Semiconductor, the Ontario-based foundry where the chip was manufactured.
The reason such a large CCD was achievable, says Bredthauer, is in part due to the same advances in semiconductor fabrication that have crammed more transistors into microprocessors and memory chips, such as in photolithography, the technology used to pattern smaller and smaller transistors on a chip. A CCD device is made using the same processing steps and materials – silicon, silicon dioxide, and aluminum – as microprocessors and flash memory, differing only in the design of the circuitry on the chip. “CCD imagery has been following on the coattails of [the semiconductor] industry,” Bredthauer says.
Perhaps the key improvements in semiconductor fabrication that have made possible this giant imaging chip are the quality of materials used and the cleanliness of fabrication facilities. If the “clean” rooms aren’t sufficiently clean, dust and other particulate matter can contaminate the chemical layers used to build the CCD, says Barry Burke, a senior staff member at MIT’s Lincoln Laboratory. Not only can defects create defunct pixels – ones that are permanently white or black, he says, they can also create electrical shorts in the circuitry, rendering portions, if not all of the device, defective. “To get a perfect device of that size is a major challenge,” he says. Hence, advances in air-filtering technology in clean rooms have enabled larger defect-free devices, Burke says.
Design considerations are also important in making such a large chip, says Bredthauer. The pixels of a CCD collect photons that are instantly converted to an electrical charge proportional to the intensity of light. In order for the charge to actually produce an image, it is passed through the chip to circuitry lying on the periphery of the device, where it is amplified and converted into a voltage that’s used to record the final image. A large CCD runs the risk of producing “traffic jams,” where the electric charge can’t quickly and effectively make it to the output circuitry. To solve this problem, Bredthauer designed the output ports to simultaneously extract charge from different regions of the chip, thereby keeping the charge flowing smoothly and allowing for quick image rendering.
Currently, these mega-chips are priced at a breath-taking $80,000-100,000 each, and are custom manufactured for each specific (mostly astronomical) application. The steep cost is one reason such CCDs are not likely to show up in consumer products soon. And, for the average image-taker, current resolutions in the tens of megapixels is sufficient. “From the consumer’s point of view, it seems like there’s almost enough pixels to do what people want to do – to look at pictures and share them,” says Michael Cohen, senior researcher in Microsoft’s interactive visual media group and graphics group.
However, the chance to take extremely high-quality digital pictures, which can produce three-foot by three-foot posters without losing quality, is alluring to photography buffs outside scientific and surveillance fields. Since the announcement of their record-breaking chip, Bredthauer says the photography blogging community has been writing about using the chips in professional cameras that traditionally use film that’s four inches by five inches. He’s received inquiries about their 100-megapixel chip from private parties. “That’s high-end photography,” he says.
DeepMind’s cofounder: Generative AI is just a phase. What’s next is interactive AI.
“This is a profound moment in the history of technology,” says Mustafa Suleyman.
What to know about this autumn’s covid vaccines
New variants will pose a challenge, but early signs suggest the shots will still boost antibody responses.
Human-plus-AI solutions mitigate security threats
With the right human oversight, emerging technologies like artificial intelligence can help keep business and customer data secure
Next slide, please: A brief history of the corporate presentation
From million-dollar slide shows to Steve Jobs’s introduction of the iPhone, a bit of show business never hurt plain old business.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.