Is there any online resource that provides camera specs that adhere to ISO standards?

ISO has published many standards for camera specs:

Of course there are many others like CIE...

But I've never seen those mentioned in tech specs. Especially not by the manufacturers. ISO sensitivity is always rounded, dynamic range never officially mentioned...

Does that mean that anyone who wants standardized specs should go ahead and benchmark their own camera using expensive tools?

  • I'd have to guess that your average consumer would lose interest reading any one of those standards, so perhaps it's simply that the marketing departments don't see the need to publish their cameras compliance statements? I'm pretty sure the engineers designing/making the cameras would be at least aiming for some of these though.
    – HamishKL
    Jan 27 '16 at 23:01

No one publishes standardized camera specs online based strictly on ISO standards.

The basic problem is that many of these standards were originally written to be applied to categories specific to the nature of photographic film. Even though they have been updated to apply to digital imaging, there are enough differences between how film images are captured and developed and how digital images are captured and developed to make at least some of these same categories less meaningful in terms of the resulting viewable image that can be produced by a camera with a specific set of these technical specifications.

Depending on the demosaicing algorithms used, the same raw data can be developed differently and thus yield widely varying measurements from the same image file in terms of these categories originally created to describe film characteristics: Tonal Response, Resolution, Dynamic Range, Signal-to-Noise Ratio, Sensitivity, etc.

The measurement of a film's sensitivity, for example, was based on the shape and size of the crystals in the film's emulsion and how much they reacted to a specified amount of light. Changing the development time or the concentration of the developing chemicals didn't change the original chemical composition of the emulsion or how that emulsion reacted to a specified amount of light when first exposed, it just changed the subsequent chemical reaction applied to that emulsion after the film had been exposed.

With digital imaging many, but certainly not all, of theses categories are more dependent upon the processing applied to the raw data coming off the sensor than to the hardware that captures the raw data. And since the raw data may be offloaded from the camera and be processed by a near infinite variety of systems and applications, the measurements provided by ISO standards such as those mentioned in the question are less meaningful, in terms of the capturing hardware, in the digital environment than they once were when applied to the variety of film available. Film had to be selected prior to the capture of an image. With digital imaging, processes that don't even exist at the time the raw data is captured may later be applied to that data.

There are two interrelated characteristics mentioned in the question that are heavily dependent upon hardware: Dynamic Range and Signal-to-Noise Ratio. Among those listed in the question, these two categories also happen to be the ones most referenced by both manufacturers and well known independent testing labs. But even there, most photographers are much more interested in the artistic possibilities hardware with a specific set of specifications will allow them to exercise than they are the exact method and standards used to obtain comparative measurements between two particular camera models they may be considering to use to capture images. Both the marketing departments of camera makers and the independent testing labs are going to direct their efforts at those most likely to be interested in the practical usage of the cameras because that's where they both stand to gain the most - either in terms of buyers and units sold or in terms of visitors to and ad revenue from their web sites.

Does that mean that anyone who wants standardized specs should go ahead and benchmark their own camera using expensive tools?

Pretty much. Because there is no realistic economic incentive for anyone else to do it for you.

  • I'm sorry to say this but this doesn't really answer the question. Furthermore I'm afraid you're wrong. The list of standards above are all written from scratch for digital imaging. And if you read some of the introductions (in the free previews) - and I forgive you if you didn't - you would find out that one of the reasons to make new standards (they where all born around 2000) was that the norms for film just didn't apply any more.
    – Duvrai
    Jan 28 '16 at 6:46
  • They be totally rewritten for digital imaging (and the answer acknowledges such), but the categories they attempt to quantify are based on film technology. And as also stated in the answer, the processing pipeline (external to the camera) is such a large piece of the end result as to render measurements of strictly the camera itself as far from indicating the final results possible with a particular piece of hardware.
    – Michael C
    Jan 28 '16 at 9:04
  • Ok thanks for the added first and last phrases. It now is an answer ;-)
    – Duvrai
    Jan 28 '16 at 9:15
  • Again, they have nothing to do with film and are perfectly useful for digital photography: OECF only applies to sensors and their resulting raw output, it immediately tells you how tones are distributed (log), how far you can push your exposure in post and precisely how much detail there is in any tone range, accompanied and rendered even more useful by "ISO DSC dynamic range" and the new definition of SNR, two simple numbers and purely digital specifications. The new sensitivity standard (luckily) provides comparability to old ISO 5800 film speed standard, but totally redefines it for DSCs.
    – Duvrai
    Jan 28 '16 at 10:06
  • Again, the disconnect is that no one views raw data without running it through a processing pipeline first. You can measure all of the sensor output you wish, but it doesn't really translate directly to what the eyes can see when they view an image. Sure the engineers and designers creating the cameras pay attention to these things because the sensor's capabilities do have an effect on what a final image can look like. But that doesn't mean the consumer market (and even more so the professionals) for which the cameras are ultimately intended are obsessed with such metrics. Not at all.
    – Michael C
    Jan 28 '16 at 23:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.