The Digital SLR

Several factors delayed the move to SLR technology for digital photography. First, the early sensors were much lower megapixel resolutions, far too low to begin to compete with film photography. Photo hobbyists saw few advantages to moving to a low resolution system since the reason they used an SLR was high image quality. It wasn't until resolutions reached the 2 to 4 megapixel range that there was any interest in a digital SLR.

Second, the early sensors were very small, developed primarily for video usage. Some tried to develop proprietary systems based on a smaller sensor size with more compact lenses, but nothing caught on in the industry. The cost of early sensors was also astronomical, making the early digital SLR cameras only useful for production, high volume photography where the cost could be justified.

Third, with sensor resolution and size in constant development it was much easier to design a point-and-shoot camera around each new sensor generation. Until sensor size stabilized, a digital SLR that used existing lens systems was not practical.

APS (Advanced Photo System)

35mm first appeared on the scene in the 1930s and the film format simply took 35mm motion picture film and spooled it into a light-tight canister. By the 1960s, with point-and-shoot and developments in SLR technology, 35mm had become king of the film formats. Even as film manufacturers tried to introduce other film formats, 35mm continued to grow and prosper. With the very early developments in digital photography, Kodak, Fuji and others saw the handwriting on the wall for film photography. One of the ongoing complaints about 35mm film, however, has been that the 2:3 image format required image loss in almost every standard print size. 8x10 and 5x7 are neither a 2:3 ratio and both required cropping of the 35mm negative.


In 1996 a new APS (Advanced Photo System) initiative by Kodak, Fujifilm, Minolta, Nikon, Canon and others was introduced to save film by standardizing on new ratios and adding new "computer-like" data storage capabilities in the taking and processing of images. APS included a new film size - 30.2mm x 16.7mm - that could be printed full-frame (H or HDTV format), or use standard crops of 25.1mm x 16.7mm (C or Classic 2:3 format) or 9.5mm x 30.2mm (P or Panorama). Actually, the image size that was always shot was the 30.2mm x 16.7mm, and the other sizes were just standard crops.


The industry was confident they could sell APS, which in the most common C view was only about 55% of the already small 35mm size. Some manufacturers like Minolta developed new APS lens lines with smaller cameras and lenses, and Canon and Nikon developed APS camera bodies that could mount their regular lenses - and a few custom APS lenses.


In the end, APS failed in the film market, but it is important in our discussion of Digital Photography because it was also a standard for digital development. At the time of the APS initiatives, it seemed reasonable to aim for development of digital sensors for that same APS format, so lenses for both systems were interchangeable. The industry aimed for the APS C size sensor, which would be about 16.7mm x 25.1mm the same ratio (2:3) as the classic 35mm format. This allowed existing 35mm lenses to be used, with a lens factor of 1.5 to 1.6, and the development of new lenses with a smaller image circle.

What do we mean by Lens Factor? Why does it matter?

The Lens Factor

APS really doesn't matter in the film market, except as a brief and intense flurry of activity that has produced a huge number of APS cameras in every imaginable format. Some of the high-end models are particularly interesting and will become collectibles. APS cameras like the tiny Minolta Vectis S-1 SLR system, the Canon IX, and Nikon Pronea are remarkably well-built and intelligent designs for the APS format.


The APS may be discontinued and dying in the film market, but it is alive and well in today's Digital SLR market. Today Sony, Nikon, Canon, Pentax, Minolta, Samsung and every other production SLR on the market (except Olympus) uses a sensor that is approximately APS C size. This matters because the lenses designed for 35mm SLR photography do not behave the same when used to capture images with the smaller APS sensor.

Why the Digital SLR? Understanding 35mm and Digital Lenses
Comments Locked

81 Comments

View All Comments

  • ic144 - Monday, September 25, 2006 - link

    why don't we all just head over to http://www.dpreview.com">www.dpreview.com?
  • silver - Monday, September 25, 2006 - link

    Some people appreciate diversity in opinion. Take me for instance. I would prefer to read three reviews on a motherboard and come to my own consensus prior to actually purchasing said motherboard.
  • tsapiano - Monday, September 25, 2006 - link

    Overall a very good article, however there is one major points that I'd like to address here. The conclusion page (under 'Lens Confusion') makes the following statement:

    quote:

    ...eventually name lenses by their real APS C/DX focal lengths once 35mm faded away


    Focal length is a physical property of the lens itself - it is not affected by the size of the sensor. While the field of view (and hence behaviour) of a given focal length changes based on the size of the imaging plane that doesn't change the fact that the focal length remains the same. As this article notes, each format (8x10", 4x5", 6cm, 35mm, APS, etc.) uses different focal lengths for different purposes so the 35mm vs. 24mm (APS/DX) change is no different ;)

    'Equivalent' focal lengths are useful as a transitional aide, but if the 35mm format fades away, then why would we want to base things on being equivalent to an obsolete format? For a new user comming into the DSLR market, it would seem just as easy to teach them that a '35mm lens is a normal' rather than trying to explain how focal lengths were used on a format that they aren't using ;) Naturally, for users comming from the 35mm space it might be easier to use the equivalent numbers - however that is only a problem durring the transition period.

    Conversely, if the market moves toward a tiered system with professional equipment using 35mm sensors and consumer equipment uses APS-C then using 'equivalent' focal lengths will simply create more confusion. For instance, if an APS camera user were to buy a professional lens (that will be labeled with the real focal length) their consumer equipment (which is labeled with an 'equivalent' focal length) will be marked differently. IMHO simply telling users to multiply all focal lengths by 1.5x is much easier than explaining that a professional 100mm lens produces the same FoV as a consumer 150mm lens ;)
  • Wesley Fink - Monday, September 25, 2006 - link

    I basically agree with you. My only concern here is that the "digital" lenses designed for the DX format will not provide a full image on a 35mm camera. You also quickly find that the 28mm to 80mm lens is NOT a wide-angle to short tele when mountd on a digital SLR. Then there is the exotic sounding 18-55mm kit lens that behaves on a digital SLR like a 28mm-85mm lens and that most definitely will NOT operate as a full-frame 18mm on a 35mm camera.
  • tsapiano - Monday, September 25, 2006 - link

    quote:

    My only concern here is that the "digital" lenses designed for the DX format will not provide a full image on a 35mm camera.


    True, but if you label a 12-24mm DX lens an 18-36mm 'equivalent' lens and put it on the shelf beside a 17-35mm FF lens you're bound to confuse people even more. The 17-35 will also mount on an APS camera, but it will provide the eqivalent of a 25-50mm lens - however with the differential labeling, it may appear that it would be wider than the DX lens. If you just stick with the real focal length of the lens, then you give people a better idea of their relative reach.

    quote:

    Then there is the exotic sounding 18-55mm kit lens that behaves on a digital SLR like a 28mm-85mm lens and that most definitely will NOT operate as a full-frame 18mm on a 35mm camera.


    But that's the fundamental problem - the only reason we relate 18mm as an ultra-wide focal length is that we're still used to shooting with 35mm. People with this level of experience generally can figure this stuff out relatively easilly - the people who have trouble with it are generally those just starting out. For someone who has never used 35mm, however, 18mm doesn't really mean much other than what they are told - simply telling them that 18mm is a moderate wide angle is much simpler than trying to explain 'equivalence' to a format they've never used (and likely never will) ;)

    As for mounting an 18-55mm lens on a full-frame camera, you'd still get the same FoV as a full-frame 18mm lens it'd just be that the edges would have extreme vinetting so the image would be difficult to make use of. Either way, it's easier to explain 'don't put DX/EF-S/etc. lenses on a FF camera' than it is to explain why the APS 18-36mm lens is much wider than the FF 17-35mm lens ;)


    Regardless, I certainly understand what you are saying ;) The caveat is that the vendors are kind of stuck between a rock and a hard place. They could rebrand APS lenses with equivalent focal lengths, however (a) that creates confusion with full-frame lenses also in their lineup (which for most vendors is the vast majority of their lines) and (b) it means that they'll be stuck using this 'equivalent' nomenclature long after most consumers forget what 35mm was ;) If they stay with the real focal lengths, it creates some confusion in the short term, but it allows them to use a standard designation across their entire lineup and will eventually become a non-issue once people get used to it.

    Purhaps the easier solution would be to simply place FoV angles alongside the focal length designations on the lens barrel. The concept of a 45 horizontal degree field of view would likely mean a whole lot more to the average consumer than saying a 28mm lens on an APS-C body ;)
  • silver - Monday, September 25, 2006 - link

    I never did understand why manufacturers didn't simply go with a magnification factor. A normal lens would be a 1X which would be the equivalent of the diagonal of the film/CCD/CMOS. A 0.5X would be half the diagonal and so forth. The only issue I can see would be in considering the perspective of the subject to the surroundings in the image. Not a huge concern when using focal lengths in the 0.5X ~ 4.0X range though admitedly there are a huge number of cameras with lenses that zoom well beyond those parameters.
  • Visual - Tuesday, September 26, 2006 - link

    well obviously its because the manufacturers have no clue on what camera you'll strap the lenses and what their film/sensor diagonal will be.
  • silver - Tuesday, September 26, 2006 - link

    In that case they really need to hire better engineers if they don't know what size the CCD's in their own cameras are !

    Note that this all points out the need for some standardization as well. Standardization is in the best intrest of the consumer as it opens up competition. This is why PC sell 99:1 versus Mac. Mac has completely alienated a lot of their user base by persistently changing their architecture. Something they seem determined to keep doing.

    Lastly, the mangnification would be indicated based on the lenses targetd system. A lens designed for use on a Canon full-frame system would simply need to be stamped differently for use on a Nikon. The lenses aren't interchangeable unless one uses a lens mount such as the Tamron Adaptall that has seen very modest use in the past on 35mm cameras.
  • wgoldfarb - Monday, September 25, 2006 - link

    Sorry for the very long post, but photography is a subject that is close to me and I was left with some very mixed feelings after reading your article. I have been a "prosumer" photographer for over 25 years, with some professional work thrown-in a long time ago. I just switched to 100% digital not too long ago, so naturally I was excited to hear AT was starting a digital photography section. As a newbie when it comes to computers and overclodcking I have come to admire your incredibly thorough reviews and articles for computer hardware, and was looking forward to the same level of depth and knowledge in photography. However, I must admit that after reading this article I was somewhat disappointed.

    Overall the article is excellent, as is usual with AT. But in the article you also send some mixed messages, and say some things that affect your credibility.

    You start by saying that most computer enthusiasts don't know enough to understand the "jargon" used by traditional review sites. However, you are also focusing on digital SLRs which can cost $500 and more once you include the body, lens and other required accesories. Anyone seriously considering such an SLR (instead of a cheaper, simpler point-and-shoot) almost certainly already knows enough about photography to understand the "jargon", or at least can easily learn it with some research. This is a double standard. You say most photography sites use language that is gibberish to a non-photographer. Guess what: your own reviews of hardware are gibberish to most computer newbies, and even to some more knowledgeable users. Yet this is precisely why you offer great reviews. Assuming some prior knowledge on the part of your reader allows you to focus on incredibly detailed, in-depth reviews of hardware. That is, to me, one of your greatest values. I can learn the basics anywhere, but only here can I find such outstanding in-depth reviews or a community of incredibly knowledgeable users. Yet for photography you are suggesting you will target a comparably less informed user. I think this underestimates the knowledge of your community. In addition, catering to such an audience will limit the technical depth of your reviews. I'd much rather see you target a more informed user (maybe provide some guides to allow newbies to learn the jargon?) and concentrate on the same level of technical depth you currently have in hardware, along with jargon, gibberish and all. Anyone can provide a basic review of equipment, be it computer hardware or digital SLRs. Very few people can provide truly in-depth, reliable, technical reviews, and I think this is where AT can stand out from the rest.

    You also make some comments that could affect your credibility. As an example, on page 1 you say "In the end taking a digital photo is still basically dependent on the same set of 'rules' as taking a film image, as the only real difference in digital and film is what happens after the image is captured". This is not accurate. As a very important example, the nature of digital sensors has a significant impact of how you expose an image for digital capture. Yes, you could just point and shoot with a digital camera, but you would not be getting the most out of your image or equipment. Given the nature of digital sensors, with digital cameras you need to "expose to the right" to maximize the amount of detail you can capture (i.e. you need to overexpose your image somewhat). You also need to think carefully about the format you will use to store the image on the camera (which may impact quality and definitely impacts the amount of post-processing you may need), and must make careful decisions on white balance before taking the picture. There is an inherent difference between film and digital sensors, and they do affect how you take pictures even before you release the shutter. Look at it this way: if a newbie was about to purchase some DDR2 1066 RAM only to permanently run it at DDR2 553 speeds you would almost certainly suggest that he/she learn enough to overclock his system and leverage the RAM to its fullest, or else you would suggest that he/she not waste the money on such expensive memory. It is no different with digital cameras. If someone pays upwards of $500 for an SLR ,he/she should learn enough to use it to its full potential, and not just use it as a film camera or a point-and-shoot digital.

    Having said all this, I think you could offer a great digital photography review section if you stick to what you do best: in-depth, reliable, highly technical reviews of equipment, with your own benchmarks and testing techniques. People can learn the basics anywhere, but here they would find great in-depth reviews not available elsewhere. You go into incredible detail, testing and benchmarking when evaluating the performance of a GPU, and certainly use lots of very technical "jargon" that allows you to offer outstanding reviews, without worrying about catering to less informed users. Those users (including me) usually do their basic research to learn the jargon before reading your reviews. Why would you not apply the same high standards when reviewing digital SLRs?

    Moving on to the type of tests, I would love to see extensive batteries of tests covering a wide range of factors, similar to what you now do with hardware. Conduct tests of cameras using identical images that we can compare across cameras and Brands. If testing an SLR body, use a reference high-quality lens so that lens quality is not a factor in the captured image, again allowing us to compare different camera bodies. Develop tests to measure noise at higher ISOs (a potential issue with smaller sensors, such as those in the 4/3 system pioneered by Olympus). Evaluate the time it takes for an image to be written to the camera's storage media at different formats, and how that impacts the ability to shoot continuous frames. If the camera has a built-in flash, evaluate the effectiveness and range of the flash. Look at the available lens offerings for a given body (which, as you correctly point out, can sometimes have a greater impact on image quality than the number of pixels on the sensor). Evaluate the quality of "kit lenses" (by comparing their images to those of the reference lens). Evaluate the in-camera software (e.g. how well does the auto white balance work under different lighting conditions? How well does the camera "process" the raw image to convert it to a jpg?). The list could go on and on, even without going beyond just cameras. There are numerous relevant factors to test in a digital camera, I just touched on a few. There is LOTS of room for thorough, in-depth, technical reviews that you would not be able to offer if you cater to a less informed crowd. And there certainly is room for a website such as AT to conduct these detailed, thorough reviews, using your experience with hardware testing to provide a level of thoroughness and detail that is simply not available elsewhere.

  • Wesley Fink - Monday, September 25, 2006 - link

    You bring up some interesting points, and they are points the staff at AT are still debating. You might find it interesting that almost all the AT Editors thought this introductory guide was possibly too technical. Several commented that they really had to pay attention to get the technical details I was discussing. This is from Editors who write those technical computer equipment reviews you enjoy here. I was worried it was too basic since I didn't even cover things like lens perspective and color balance, in the interest of keeping the size more managable and the article more readable.

    Our real issue with Digital Cameras is knowing which audience to address. We do understand there are readers that are vey knowledgable in digital photography, but there are also readers who know next to nothing about digital cameras. AT is known as a computer equipment review site, so if we sometimes go too far on the technical side of computer testing it is normally forgiven. However, based on our Editors and questions from readers, we really think we need to aim our Digital Camera reviews at a slightly lower common denominator.

    We will likely test and talk more about where various digital photo technology is useful and the best use for certain equipment and cameras than other photo reviews. This will likely be below what you want in the beginning, but if the demand is there we will move to ever more extensive benchmarking as the section evolves.

    As for your experience with the sensor in your camera, it appears to be camera/sensor specific. I could make the same argument for every film that exists - they all respond best to certain exposure techniques and those techniques differ depending on the film. However my personal experience has been the opposite of yours - blown highlight from overexposure are much more an issue than underexposure in digital, since I can usually correct shadow detail in Photoshop and I can't always recover blown highlights. Yes, white balance adjustments can be made on the fly with digital and ISO can be adjusted on the fly, which is very unlike film. However, white balance and ISO just work at a different point in digital, and they stil mean the same thing. No we don't have the added latitude of the print process as film does, but we have a similar and even greater latitiude with post processing in Photoshop or other image-editing software.

    While I do agree that anyone using only point and shoot is not getting all they can from the camera I also believe strongly that you can still get a lot of quality out of your digital camera on Auto when it is backed by some knowledge. I would also bet you that the great majority of our readers who use digital cameras generally use auto - even with SLRs - and only once in a while go off the program for special situations. I learned this the hard way trying to help our other "geek" editors who almost all own SLR cameras they bought themselves, but who really struggled to take decent photos for our reviews. Once they were given a few tips on lenses to use for available light and how to control depth of field their photos instantly improved = and they almost all still use auto.

    Your suggestions for tests and benchmarks are appreciated, and we will definitely consider those as we determine how we will move forward in reviewing Digital Photography equipment.

Log in

Don't have an account? Sign up now