It seems to me that my costlier (Tamron 90 f/2.8 macro) lens gives better colors than my cheap Canon 18-55 IS kit lens.

So does lens matter for color quality? What physical processes are involved in that?

  • 1
    possible duplicate of How does a lens affect the saturation of a picture?
    – dpollitt
    Mar 6 '12 at 13:59
  • 3
    I will add two pictures taken with the two lenses, it can be interesting I think.
    – Paolo
    Mar 6 '12 at 14:47
  • Just as an aside, both X-Rite, with the ColorChecker Passport (and software), and Datacolor, with their similar SpyderCheckr system, can help you minimize the difference between lenses (and bodies, for that matter) for consistency across a shoot by creating profiles you can use in various RAW-processing applications.
    – user2719
    Mar 6 '12 at 19:59

Yes the lens design and construction can and does influence colour reproduction. Different glass formulas and different coatings transmit frequencies of light differently, and lens defects such as flare will likely affect the colour reproduction.

Whilst more vibrant colours are generally seen as "better" it's worth knowing that different manufacturers intentionally produce different renderings, sometimes according to cultural preferences. For example German lenses produced by Leica known for being cooler compared to Japanese designed lenses from Canon and Nikon.

  • So in a few word spending more money for a better lens, beside having sharer images, usually means having better colors?
    – Paolo
    Mar 6 '12 at 14:41
  • @Paolo generally yes, but beware that "better" can be subjective, some people prefer muted colours in their images.
    – Matt Grum
    Mar 6 '12 at 15:42
  • I hope I will manage to post a good example, but with better I mean more precise, not just more saturated.
    – Paolo
    Mar 6 '12 at 15:49
  • 1
    When I switched from Pentax to Canon I was surprised how muted the colours were. The kit lens was more muted then the Canon 50mm prime f1.4 but still the Comparable Pentax 50mm f1.4 produced much richer colors. I now have 2 Canon "L" series lenses and they yield most "natural" colours out of all the lenses I've had. Mar 6 '12 at 16:08
  • 2
    In a film world, the color characteristics of a lens really mattered. However, in the digital world of today, how important are the color characteristics of a lens? I would imagine that if one lens reproduced certain frequencies very poorly, such as greens, that it would matter...however I would think its generally not the case that lenses differ enough for it to matter in a digital world. That doesn't even account for differences in bayer color filters, and how those affect the color reproduction of digital images, or how the nature of those filters differ between manufacturers.
    – jrista
    Mar 7 '12 at 1:52

Here are two example pictures - same scene, same time of day, same camera (Sony with vivid profile), fixed white balance, two different lenses, that most would agree appear drastically different color-wise:

Agfa Color Ambion 35mm f4 Agfa Color Ambion 35mm f4

enter image description here MOG Oreston 50mm f1.9 (there was a 390nm UV filter, but no skylight filter or similar, attached)


So here is the mini test :)

Canon 400D camera. Both shot at f/5.6 4,0s iso 100 (just a lamp for light). Made camera white balance calibration.

Tamron 90:

Tamron 90

Canon 18-55:

Canon 18-55

So the amazing thing is that with the same exposure the 18-55 is ratehr darker.

Unfortunately the test is not quite resolutive, since there are too many variables to consider. And the test didn't cover them all.

For sure the bad exposition caused by Canon is affecting the color rendering, and this could be the effect I have experienced before.

Ok the two images above are too dark, so I take two more pics, with better exposure.

tamron 90:

enter image description here

canon 18-55:

enter image description here

Tamron 90mm has got a clear advantage, my guess is that reasons are:

  1. More accurate exposure just helps camera to better render colors
  2. canon 18-55 seems to have some subtle glare problem, making things a little less saturated

On the other hand I wouldn't really go looking at wavelength transfer function of glasses or similar physical effects.

  • 1
    Hi Paolo. This is not really an answer to your question, and would be better as reference material IN your question, so future answers can factor in this information. Could you move this into your question, and delete this answer? I'd like to provide an answer that demonstrates how little the two images really differ, and how even if they did, it probably wouldn't matter with digital photography. However I can only really do so if this is part of the question. Thanks.
    – jrista
    Mar 7 '12 at 1:56
  • Honestly I do not see why this isn't an answere. I'm showing the differences and explayining why
    – Paolo
    Mar 7 '12 at 8:42
  • 3
    I would attribute the changes to a difference in exposure, not a difference in color fidelity of the lenses. Its also very difficult to evaluate color differences with two photos that are taken at different angles and perspectives. With two properly taken photos, you could overlay one on top of the other in Photoshop, and set the upper layer to "Difference" blend. You would instantly be able to see any differences in color, which would be minor at best, and not a great indication of color fidelity differences.
    – jrista
    Mar 7 '12 at 17:33
  • 1
    To accurately evaluate the color rendition of a lens (which, in the case of digital, is often less about the lens and more about the sensor's color filters and picture styles), you need to use a properly calibrated color checker card (i.e. the X-Rite Color Checker), a properly calibrated and consistent light source (preferably a sunlight or daylight balanced bulb), and you need a camera mount setup that will allow truly identical framing of the subjects to ensure comparisons are indeed indicative of color rendition differences.
    – jrista
    Mar 7 '12 at 17:35
  • @jrista Thanks for you comments. Unforutately the two lenses haven't overlapping focal length, so even with the greatest care I guess the best one can do is to crop the 55mm shot in order to fit the resized 90mm shot... so I will held myself content with my poor men testing environment :D
    – Paolo
    Mar 9 '12 at 10:54

The 18-55 will not have very good light transmission as a kit lens, meaning more light is absorbed by the lens elements then in the Tamron lens hence being darker. For a full comparison you can always go to dxomark.

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.