History doomed to repeat itself? Project promises digital cartridge for film SLRs
posted Tuesday, August 20, 2013 at 9:44 AM EST
Readers who've been interested in digital photography since the early days will no doubt remember its most infamous piece of vaporware: the Silicon Film EFS-1 Electronic Film System. There was certainly plenty of appeal to the design, launched way back in 1998. In an era where even the most affordable digital SLRs cost upwards of US$3,000, Silicon Film promised photographers the ability to capture digital images with the standard film SLR they already owned. The idea's popularity notwithstanding, though, the EFS-1 failed to reach the market.
Today, we hear news of the DigiPod, a project of James Jackson, a 58-year old ex-pro photographer in the United Kingdom. Jackson is aiming to create a very similar product to the long-abandoned Silicon Film cartridge, and currently attempting to raise money over at crowdfunding website IndieGoGo.
The news immediately set us to thinking about why the Silicon Film product failed, which of its main issues have been (or could be) solved with modern technology, and what new obstacles a digital film cartridge would face. Will the DigiPod come to fruition? The benefit of our hindsight doesn't give us great cause for optimism, we must admit.
A telling history lesson. First announced back in 1998 by Irvine Sensors Corp. subsidiary Imagek, the EFS-1 was a three-part system, consisting of the so-called (e)film cartridge, the (e)port reader that also doubled as a protective housing, and the (e)box converter which allowed copying of data to standard CompactFlash memory cards. (All named, of course, in the days when the e-prefix automatically multiplied your product's popularity by an order of magnitude. Apple had yet to try to convince anybody that the i-prefix was altogether more interesting!)
Although the lure of a drop-in digital replacement for your 35mm film was strong, the project faced huge issues that would eventually bring about its death. Perhaps the biggest problems were brought about by the 35mm film form factor, which was never intended to allow digital conversion. There were essentially two options: create a camera back which would replace the entire rear of the film camera, or create a cartridge which could fit entirely inside the camera body. Replacing the camera back would increase the cost significantly, because versions of the product would have to be designed and sold for specific camera models.
Compatibility. Silicon Film hence chose the latter route, but this brought its own challenges. The precise dimensions of the film path varied by camera body, and there would also be differences between the cameras' film transport and shutter mechanisms. That meant that although the cartridge wouldn't be specific just to one single camera model, its makers couldn't promise widespread compatibility, either. Testing would be required to confirm it would even fit, let alone work in any given camera. Out of the box, the company's potential audience was limited, as it promised compatibility with just six camera bodies: the Canon EOS 1N, EOS A2 and EOS A5 as well as the Nikon F5, N90 and F3.
The DigiPod will face exactly these same issues, but they could prove to be of even greater significance today. Fifteen years ago, film SLRs were still commonplace, and relatively few photographers were shooting with digital SLRs. Now, most of us have long since retired our film gear, and many have even moved away from mounts used in the film days, in favor of mirrorless cameras and mounts. The market for a drop-in digital film cartridge will be much smaller than it was in Silicon Film's day, and the need for broad compatibility so as to try and mop up as many photographers as possible will be even more key.
The IndieGoGo page makes the surprising claim that the DigiPod will work with "most, if not all 35mm SLR cameras and some ... non-SLR viewfinder 35mm cameras". Recognizing the difficulty of locating the sensor differently depending on the camera body design, Jackson does note that this is being taken into account. Somewhat alarmingly, though, it's suggested that the user will be required to adjust horizontal and vertical alignment themselves to get the sensor positioned appropriately.
Limited space. The packaging issues of a 35mm film body went on to be felt by Silicon Film elsewhere, too. Almost the entire product needed to be able to fit within the space occupied by a 35mm film canister, with a lengthy projection to allow for placement of the image sensor behind the camera's shutter. Such a design was likely to prove fragile, especially given that the sensor assembly would be placed in harm's way whenever the device was taken out of the camera. Worse, it left little space for removable storage -- or much in the way of internal storage -- or for a power source and the electronics required for the product to work, let alone for niceties like the RF shielding required to attain FCC approval, and provision for cooling.
Here, DigiPod may find advantages over the Silicon Film design. Storage capacities have increased to the point where we feel confident in stating that it would no longer prove an issue at all. The DigiPod will, apparently, have a MicroSD card slot, and a single high-capacity card should easily satisfy most photographers these days. If not, buy a second or third, and swap them. It's simply not a problem.
Although battery technology hasn't made a similarly huge leap forward in the last decade-and-a-half, modern electronics don't devour batteries with quite the same zeal that they once did. And in other respects, a modern design will also be easier to fit into the limited space open to the digital cartridge. It should generate less heat, to boot. But it's still faced with concerns about fragility -- and it's worth noting that what's likely the most expensive part of the product is precisely that which will be placed in harm's way outside of the cartridge.
Connectivity. In one area, there is the potential for a very, very big advantage for the DigiPod -- but early indications are that Jackson hasn't yet taken advantage of it -- or perhaps, is intentionally eschewing it for a more film-like (read: limited) user experience.
Since the entire Silicon Film EFS-1 cartridge had to sit inside the camera body, there was no way to make adjustments to the settings between shots, let alone to review your photos post-capture. Fifteen years later, though, almost all of us carry smartphones, and most include low-power Wi-Fi and Bluetooth radios which would seem ideally suited to providing settings adjustment and image review without the need to remove the DigiPod from the camera.
Unfortunately, in noting its planned ISO sensitivity range of 24 to 3200 equivalents, Jackson states that ISO sensitivity adjustments must be "set manually before the Digipod is put in the camera". This suggests that just as with its precursor, the Digipod will provide no way to adjust setup remotely, let alone to review and delete images, or offload them wirelessly for social sharing and so forth. Indeed, the only connectivity noted on the IndieGoGo page is standard Micro USB, which will prove to be of little use in the field. (Not many smart devices can even act as a USB host, and so couldn't even be used to adjust settings via a USB tether.)
(Much) more limited than an entry-level DSLR. And that brings up another point: Anything the camera itself wasn't designed to do, the Silicon Film (e)film cartridge likewise couldn't offer, and nor can the DigiPod as currently designed. Video, for example, is essentially impossible. Even if you were to hold the mirror up for the duration of a clip, you have no way to view your framing. Whenever the sensor is exposed, the viewfinder is dark.
Even if Jackson was to redesign the DigiPod to allow video capture with the mirror up using a Bluetooth or Wi-Fi radio for a remote live view feed, things we take for granted even in entry-level DSLRs such as autofocus would be impossible. The DigiPod can't communicate with a film body that was never designed for it, let alone have any way to control focus. In fact, there's not even a way for it to signal to the camera body that it can't keep up with burst shooting. Shoot faster than the device can record, or exceed its buffer depth, and the camera's controls will provide no indication of this.
A very early concept. Another reason that the EFS-1 stumbled is because its small sensor meant an extreme 2.8x focal length crop, and this was coupled with extremely low 1.3 megapixel resolution. With enough production volume, there would be little difficulty procuring a high-res APS-C or even full-frame sensor with high resolution, and at a price Silicon Film could never have dreamed of. (Although it's unlikely a sensor anywhere near as large as full-frame could be made to fit within a camera designed for film.) And with a smaller-than-35mm sensor, you'll need to modify your own viewfinder to provide framing guidelines.
The specific chip intended isn't clear, but the IndieGoGo page does say it will be a CMOS type. In his introductory video, Mr. Jackson states first that he intends to use a sensor with dimensions of approximately 10 x 8mm, or perhaps a square sensor. He goes on to suggest that sensor size depends on the number of people supporting the project financially. Then, an update later in the video promises a 12 x 8mm sensor. If the size isn't increased significantly, any of these are going to yield a much more extreme crop even than the Silicon Film product. (And an even narrower viewfinder view.) Wide-angle shooting will be a physical impossibility, and the relatively tiny sensor -- less than one-tenth the area of a 35mm film frame -- will mean image quality far below what a modern interchangeable-lens camera can provide.
And not only does the current development model use an even smaller 1/2.5"-type sensor, but it appears not to interface with the camera in any way at all, beyond simply peering through its lens with the film door held open. In his introductory video, Mr. Jackson can be seen to press a shutter button on the digital camera board -- either a development unit or the guts of an existing camera -- to take a picture. A prototype plastic housing is shown for the cartridge, but with no electronics mounted inside. It is clearly very early days yet for this idea, and little of the development work seems to have gone beyond the drawing board.
Certainly achievable. You could be forgiven for thinking we believe commercializing a product such as the Silicon Film EFS-1 -- and by association, the DigiPod -- to be an impossibility. That sentiment would be somewhat wide of the mark. In fact, we know with certainty that the (e)film cartridge was more than just hype, because we handled a working unit ourselves, and were provided with copies of its output. Making an allowance for its low sensor resolution, image quality actually wasn't that bad for the time, with reasonable color rendition and sharpness. Certainly nowhere near what even a smartphone camera can manage these days, but for the time, likely acceptable to many photographers.
With unlimited funding and in a market that wasn't moving forwards as rapidly as that at the start of the noughties, we have little doubt that Silicon Film could have brought the EFS-1 to market, and sold them by the pallet-load. But therein lies the problem: the money wasn't unlimited, and the market wasn't standing still.
A dose of reality. And so we come to the DigiPod's biggest challenge: it is no longer relevant to the great majority of photographers. Silicon Film came close -- very close indeed -- to actually shipping their product some three years after it was first announced, but already in that time it had become an irrelevance. Affordable, high-quality digital SLRs with almost none of the limitations of the EFS-1 system were becoming widely available, and although some photographers still looked wistfully at their film SLRs, truth be told the market had already moved on. Unable to pass European certification testing before the money ran out, the company folded without shipping a single unit.
The closest a commercial product has come since to fulfilling the EFS-1's promise was Leica's Digital Modul R, a digital back system for the Leica R9 film SLR -- and even with one company designing both camera body and digital back, it still shipped a year late at a price that ensured limited adoption.
Who is the target customer? In the present day, relatively few of us would surrender our modern cameras for our film SLRs. The modern cameras are infinitely more capable, whether at the entry level or for pros -- and at price tags that are probably lower than we'd have paid for the equivalent film gear back in the day.
The real reason some of us still lust over our film SLRs these days has less to do with the hardware of the camera, and more to do with the recording medium. Some of us long for the way in which our subjects were rendered on film, for the way in which shooting film shaped our shooting style, and yes, even for the time we spent in our darkrooms with the smell of chemicals in the air and the excitement of photos soon to be seen.
The Digipod returns little of that experience to your film SLR, and yet brings you few of the advantages of a modern interchangeable-lens digital camera. (And it comes accompanied by some very awkward disadvantages.) We wouldn't go as far as to claim it will never reach the market, but even should it do so it strikes us as an idea whose time has long since come and gone.
More details can be found on the IndieGoGo website.