"GuessTheFormat is an experiment for photographers"

Discussion in 'This or That?' started by twic, Dec 10, 2013.

  1. twic

    twic Mu-43 Rookie

    Sep 15, 2013
    London, UK
    An acquaintance of mine has set up this experiment - see if you can guess the sensor format just by looking at a picture:


    Lo and behold, the results so far:


    Which tell us that when shown a picture taken with a four thirds size sensor (which could be µ4/3 or legacy 4/3, although the examples i saw were all µ4/3), and asked if it was taken with a four thirds size sensor or an APS-C sensor, people were right only 48% of the time. That is, mostly, people can't see a difference between µ4/3 and APS-C.

    I realise it's hardly a rigorous conclusion, but i'll take it! :thumbup:
  2. HappyFish

    HappyFish Mu-43 Top Veteran

    Sep 8, 2012
    Real Name:
    saw that some of the worst photos to compare ? and sized small etc not in print ? kinda IMHO complete worthless at best its a guessing game ?
  3. mattia

    mattia Mu-43 Hall of Famer

    May 3, 2012
    The Netherlands
    I'm getting about 65-70% of them right. The biggest giveaways (other than aspect ratio, which doesn't always apply in the examples shown) tend to depend on the subject matter chosen. Well-lit landscapes at small sizes? Tough call. Poor lighting and shallow DOF and wide-ish FOV? Usually a sign of a FF image. The compact camera shots are generally the easiest to filter out (seriously, show me an ultrawide shot and ask me whether's its MFT or compact? Other than corner performance, I'm not aware of any compact cameras wider than 24mm FF equivalent). The only ones that are relatively easy to distinguish vis APS-C vs. MFT are of subject that more or less require good tracking AF (fast-moving birds in flight).

    For over half of the pictures, almost any of the systems will give you perfectly great, almost indistinguishable performance. Certainly at web sizes.
  4. carpandean

    carpandean Mu-43 Top Veteran

    Oct 29, 2010
    Western NY
    It's kinda like showing the answers from a bunch of people's math tests and asking which ones are from 6th graders, college students, working professionals or PhDs (in mathematics.) Only the questions are 95% addition and subtractions, with a little bit of multiplication and division thrown in.

    We all know that, especially when displayed on small computer screens, even camera phones take good enough photos in most situations. It's only in those fringe areas (low light, high DR, action, shallow DOF, etc.) that the larger formats give you an advantage, and even there, technology eventually provides much of that, too (for example, compare the noise of an E-M1 to that of an original 5D at the same measured ISO.)

    We also know that a good photographer with a camera phone will take better pictures than a bad one with a FF camera. For example, a good photographer knows how to frame a shot to get shallow DOF with a compact; it's just easier (i.e., there are more options) with a FF camera.

    Not to say that many wouldn't be surprised by these results. A lot of DSLR purchasers with kit lenses stapled to the front might be surprised by photos shot by somebody who know that they are doing with a compact camera. Learning how to compose is far more important than the technology. I've got the technology level that I want (other than tracking focus), so now I'm focusing on composition more.
  5. jamespetts

    jamespetts Mu-43 Top Veteran

    May 21, 2011
    London, England
    How are people supposed to tell when the images are so small and full size is not available?
  6. kevinparis

    kevinparis Cantankerous Scotsman

    Feb 12, 2010
    Gent, Belgium

    by looking ? :)

    maybe because the vast majority of photographs today are not viewed at full size or even printed out.

    As far as I can see the site is not pitching its self as some definitive guide to which camera is 'best', but a light hearted exercise in perception that perhaps for some might help define what they actually need in a camera for their own photography.

    I looked at around 100 shots on it.... got about 65% right... usually got the FF and compacts right... though not always..t though dont think I ever guessed a ff when it was compact or vice versa . APC v 4/3 was a closer call.

  7. jloden

    jloden Mu-43 Hall of Famer

    May 15, 2012
    Hunterdon County, NJ
    Real Name:
    Fun idea, though I'd love to see larger sizes available too (though I do get the point, that most images probably aren't any bigger than that on the web) :smile:

    I haven't tried yet, but I highly doubt I could tell the difference on my laptop from an image that size.

    With my iPad display looking at 500px photos, I used to make a game of guessing formats. I was generally able to tell pretty reliably what was shot with FF versus 4/3 or a compact camera or in some cases an APS-C DSLR, but I can rarely tell on my macbook pro screen (pre-Retina) unless there's an obvious indicator. Barring DoF or the like, the giveaways are typically noise, clarity, and fine detail resolution.

    Obviously, if you're sharing photos in good light at low ISO at 800x600 the gap really narrows up :cool:
  8. jloden

    jloden Mu-43 Hall of Famer

    May 15, 2012
    Hunterdon County, NJ
    Real Name:
    Ok, so I tried it on the laptop now, 12 out of 15 correct after all. Two of the ones I guessed wrong were really heavily processed and could have been done with a potato for all you could tell, probably should have just skipped those :tongue:
  9. QualityBuiltIn

    QualityBuiltIn Mu-43 Veteran

    Jan 1, 2011
    Edinburgh, Scotland
    Wonderful exercise. Compact is pretty obvious, but APS v Full frame is beyond me. Would have been easier with the option to zoom.
  10. nstelemark

    nstelemark Mu-43 Hall of Famer

    May 28, 2013
    Halifax, Nova Scotia Canada
    Real Name:
    I did note that one of the "compact" cameras was the Canon G1X. The IQ is easily as good as m43 or APS-C, so that was a bit unfair.

    I did not do well on a lot of them, but there are a lot of variables, not to mention focus accuracy.
  11. T N Args

    T N Args Agent Photocateur

    Dec 3, 2013
    Adelaide, Australia
    Real Name:
    call me Arg
    Our own individual guessing scores are irrelevant. The whole point of an exercise like this is that some people are going to get a high score by pure luck. If you get a high (or low) score, don't feel good (or bad) about it.

    The relevant part is the Stats page. Wow! 4/3 vs full frame, 56% correct, is impressive for 4/3.
  12. jloden

    jloden Mu-43 Hall of Famer

    May 15, 2012
    Hunterdon County, NJ
    Real Name:
    Sure, the point is that overall, most people can't tell the difference at web resolutions.

    But the point for people of taking it and checking their own scores is because we're curious if *we* can tell the difference, even if we know most people can't. It's just interesting, and it contributes to the overall stats anyway :smile:
  13. kevinparis

    kevinparis Cantankerous Scotsman

    Feb 12, 2010
    Gent, Belgium
    my takeaway from the stats is that just under half the time people can't tell what camera at all has taken a picture

    given that probably most of those visiting the site have more than a passing interest in photography and photographic gear, it seems to me a pretty damning indictment of those who value gear over fundimental photographic skills

  14. tjdean01

    tjdean01 Mu-43 Top Veteran

    Feb 20, 2013
    As this gentleman said. I'd be better able to guess if I knew what aperture was used. You want to tell me your FF has better bokeh than my PM2 and a f0.95 Voitlander? You'll need to have at least $4000 to back up that claim. Similarly you can get some m4/3s quality shots from an LX7 or rx100 these days. And for those resized landscapes, an iPhone could shoot those!
  15. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    Copying my response from a thread discussing this at SC:

  16. Reflector

    Reflector Mu-43 Veteran

    Aug 31, 2013
    Please send your friend many thanks for dispelling the intense aura of pride-ignorance driven by photography gear and their own preconceptions.

    ...Funny how my friend with an APS-C XTi still crushes people rolling in M9s and D800Es. I'm glad to see this actually shows.
  17. Just Jim

    Just Jim Mu-43 Top Veteran

    Oct 20, 2011
    I got a lot of crazy heavily post processed stuff, like added in textures and what not... Did this the other day when I saw it on Reddit. Got 63% APS-C and 4/3 got me, unless they weren't cropped. The 4:3 can be a BIG giveaway. Compacts, man highlights just explode.
  18. kevinparis

    kevinparis Cantankerous Scotsman

    Feb 12, 2010
    Gent, Belgium
    it would be interesting to see this experiment done with price bands rather than sensor sizes

    say 500, 1000, 2000, 4000 dollars.