Definition of Macro Photography

Discussion in 'Open Discussion' started by Amin Sabet, Jun 6, 2010.

  1. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    I was reading some discussion about macro photography at GetDPI and thought it was an interesting topic for discussion. I would have commented about it in that GetDPI thread, but I didn't want to interrupt the flow of what has shaped up to be a nice image thread there.

    Macro photography has long been defined as photography where the optical magnification at the time of capture is 1:1 (1X) or greater. In other words, the size of the object in the image projected on the film/sensor should equal or exceed the true size of the object in order to refer to the image as "macro".

    I know that sometimes the term "macro" is used in other ways, but for the sake of discussion, let's accept that this is the established definition of macro photography. I have to wonder why that definition was chosen. Does it come from an era before a variety of format sizes were being used?


    Case 1: An 8x10 inch format image is captured at 1:1 (1X), and an 8x10 inch print is made from the resulting uncropped image. The size of an object in this print will match the true size of the object.

    Case 2: A 35mm frame image is captured at 1:1 (1X), and an 8x10 inch print is made from the resulting uncropped image. The size of an object in this print will be more than 7 times greater than the true size of the object.

    Case 3: A Micro Four Thirds image is captured at 1:1 (1X), and an 8x10 inch print is made from the resulting uncropped image. The size of an object in this print will be nearly 15 times greater than the true size of the object.

    If the point was to define a type of photography that shows us tiny things, why choose a term that fails to take format into account? It is analogous to defining telephoto photography as photography using a lens greater than 70mm, irrespective of format. It's a clear definition but not a very useful one. Does anyone know the history/circumstances/origin of this term?
  2. photoSmart42

    photoSmart42 Mu-43 Top Veteran

    Feb 12, 2010
    San Diego, CA
    At the risk of stepping with both feet into what normally is a contentious issue, I'll offer my point of view as best I can.

    1. On the definition of the word "macro", I think it's just from the Greek root that means "making big", which fits perfectly. In contrast, "micro", while meaning really small, in this case has the distinction of "making something really small big".

    2. I've actually seen definitions of "macro" photography being defined as anything between 1:2 to 5:1. I think it's probably something that developed in lens design as something more special than conventional viewing, and 1:2 was achievable in macro lenses without the use of extenders, 1:1 with extenders, so it makes sense to me to define "macro photography" as anything beyond 1:2, and not just 1:1.

    3. As for the format distinction, I think that's where most of the contention lies. I think we complicate matters if we throw in variables like sensor resolution and print size, and my belief is that in order to have a conversation rooted in theory, we need to keep some things constant. So in that vein of thought, I'll explore the scenario above:


    - We'll assume that we're dealing with film for a moment, and for the sake of discussion let's consider we're dealing with ASA100 film of various sizes - 8X10, 35mm, and half-frame (i.e. 4/3 or MFT). That way we have consistent grain size from photo to photo.

    - Let's also assume we use the same 8x10 object as a point of discussion, and use it to image across different film formats.

    Magnification and resulting image:

    - A 1:1 macro lens will indeed image the entire 8X10 object, so that's a given

    - A 1:1 macro lens for a 35mm camera will image a 36x24mm section of the 8X10 object

    - A 1:1 macro lens for a half-frame camera will image an 18x16mm section of the 8X10 object

    Print magnification:

    - If we print an 8X10 photo of the 8X10 format film, we get to see the entire object exactly as the 8X10 film sees it, life-size.

    - If we print an 8X10 photo of the 35mm film image, we get to see the 36x24mm section of the 8X10 object blown up, but we won't get to see more detail in that object because all we're doing is enlarging the grain size to the larger image. So while we've blown up the smaller image, we haven't actually magnified it to the point where it adds any more information about that object that we can't already see in the 8X10 image if we were to section off a 36x24mm piece of it and blow that up.

    - Same applies to the half-frame print, except we blow up the existing film grain even more without adding any additional information about that object.


    To me, the concept of magnification holds validity if we keep the other parameters constant (including, most importantly, the amount of visual information we can extract from that object), which brings up the dilemma of relating the magnification factor to the film grain size/sensor resolution. Do we discuss magnification as a result of the amount of information/detail we can actually extract from the image, or do we keep the discussion at the simple convention of relating the object size to the size of the imaging plane? What is the imaging plane in that case - the sensor or the individual pixel?

    If we have a 20MP FF sensor (5D2 for example), with a pixel size of 6.4 microns, and compare it to a 14MP 4/3 sensor (GH1 for example) with a pixel size of 4.3 microns, theoretically the imaging plane is that pixel. So while the 1:1 macro definition with respect to the sensor size is constant, the sensor with the smaller pixels theoretically has a higher magnification because it can extract more information per area than the sensor with the larger pixels (i.e. I can see more detail from the smaller pixel sensor than the larger pixel sensor if we blow up the two prints to the exact same DPI - NOT print size). We're assuming a noiseless environment here, which means we have sufficient light in either case to extract equal fidelity data from each pixel size.

    Frankly I'm open to either convention, as long as the entire community agrees to the same convention. We can either talk magnification with respect to sensor size, or we can talk magnification with respect to pixel size. Personally I think the more accurate representation is the pixel size, but I suspect that's a convention change that will be next to impossible to change any time soon.

    Anyway, my $0.02. Cheers!

    • Like Like x 1
  3. Iconindustries

    Iconindustries Mu-43 Hall of Famer

    Gee Photosmart, you're a walking Encyclopaedia!! You must be a professor surely?
  4. bilzmale

    bilzmale Mu-43 All-Pro

    Seem to remember the term 'photomacrography' being used in film days - perhaps it was abbreviated to 'macro' giving us the commonly used term today.
    • Like Like x 1
  5. cosinaphile

    cosinaphile Mu-43 All-Pro

    Dec 26, 2009
    new york city
    i believe that the origin of the term macro as pertains to image size on a film or sensor surface is a truncation of the word macroscopic..... meaning large enough to be observed with the naked eye ..... then when ratios were considered between the image on film vs real world size of objects being photographed, it was only logical that the expression 1:1 macro would come into common usage

    macro contrasts with micro in microscopy which means magnifying the very tiny to make it observable . i believe the late great herbert keppeler pointed out more than once that nikons choice to call their macro lenses micro-nikkors was a misnomer and technically incorrect. i suppose that nikon considered their 1:1 lenses marvels of close up design and wished to tout this in no uncertain terms , while the rest of the industry contented themselves with macro , an altogether more sensible choice

    from what i know
    the earliest lens that refrenced itself as "macro" is a kern macro switar 50m 1.9 which could focus down to 8 inches or so
    but "macro" was being shot in the 19 c im sure even if it didnt call otself by that name
    • Like Like x 1
  6. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    Thanks, everyone. Dragos, I appreciate your thoughtful reply. I had thought about how pixel density would factor into these issues, but it was helpful for me to think about it from a standpoint of low speed film grain. Eg, if a 1:1 macro of an ant means that the ant is X grains long regardless of format size, then the possible detail shown in a given print would be the same regardless of the format. This assumes that in each format, there is a lens which can outresolve the film at the f-numbers used to make the image, which may not always hold true.

    I had been thinking of how this definition holds up in a typical internet forum, where the viewing is on a low resolution display and image dimensions are small. For example, looking at the GetDPI thread, the images shown are no more than 1024 pixels across. If Image A were captured at 1:1 using 35mm film and Image B also captured at 1:1 using half frame film, and both images were displayed uncropped at 1024px wide, Image B would show us more information about the tiniest objects in the frame. The same should hold true for equal sized, small prints, where the amount of spatial information in both captures exceeds the resolution of the printer.

    If one printed large enough that the resolution of the print exceeded the spatial information of the captures, they would show us an equal amount of information about those tiny details regardless of the difference in final magnification. However, large prints are not the typical presentation of images, especially in internet forums. Then again it is common for individuals to present magnified crops of the original capture, and in that sense, the potential for the capture to yield information about tiny objects in a subsequent magnification would be format independent.

    As stated earlier, all of this depends on the amount of detail being captured per unit area of sensor/film being the same, and I don't think this is the case with the varied format sizes we see with digital. You gave an example of a 14MP 4/3 sensor and a 20MP 35mm sensor, but we see photographs of tiny objects taken with everything from 15MP small sensor (eg, 1/1.8") to large format scanning backs. I think that from a practical standpoint, the amount of spatial detail we expect to see captured per unit of sensor area in a small sensor camera (eg 1/1.8") is going to exceed that captured per unit of sensor area in a large format camera.

    In the end, I guess I was just struck by how stringent the "criteria" were for that thread, and how roughly a certain early poster in that thread was handled for posting a small-sensor camera photo. If the cutoff is 1:1 in a large format forum, you might get some photos showing tiny details, but you're likely to get a lot of whole flower images. Applying the same cutoff in a Micro 4/3 forum is going to limit people to posting tiny and tinier.
  7. photoSmart42

    photoSmart42 Mu-43 Top Veteran

    Feb 12, 2010
    San Diego, CA
    I think the complexities of all these factors is the reason why people tend to simplify things and punt back to the convention of relating the object size to the image size, which makes sense for most photographers. I think for the most part it ends up being a wash because with larger sensors you don't need to blow up an image as much to get a good-sized print as you would a smaller sensor output.