I have two sets of soft boxes. One has incandescent bulbs (3,200k) and the other uses compact fluorescent bulbs (CFL @ 5,500K).

I'm trying to sort out some of my understandings regarding the light types and their qualities / attributes.

The basis of my question centers around this piece of information: I've heard many times from others "I look horrible when seeing my reflection in the mirror when in a room that's lit with fluorescent lights" and I myself have experienced the same thing. The skin looks harsh and it seems every blemish / bump, bruise, discoloration of the skin, etc. is accentuated.

Now, my question is: Is this due to the TYPE of lighting (Fluorescent vs Incandescent), or is this a function of the lighting temperature typically found in Fluorescent as opposed to Incandescent?

If its merely a function of the lighting temperature, then technically you should be able to get two different types of bulbs at the same temperature and get the exact same quality of lighting with the same attributes, correct?

Additionally, I have heard (if I recall correctly) that fluorescent lighting lacks the magenta color? Is this true? If so, how does this manifest itself when it comes to lighting quality?

  • To be clear, I'm not suggesting that I use the two different sets together. I was merely illustrating the point that I see different results when using the different sets (as I would expect with different color temperatures), but wondering if there is more to the result of the light beyond JUST color temperatures given that each set is a different type of lighting (CFL vs Incandescent). Sorry if this was not clearly stated. May 28, 2013 at 15:17
  • 2
    All lights lack the magenta colour in the spectrum - it doesn't exist. See richannel.org/colour-mixing-and-the-mystery-of-magenta :) May 28, 2013 at 20:28
  • This is very similar to Best fluorescent bulb color temperature for shooting people and interviews?; although you're coming at the question from a slightly different angle, I could pretty much cut and paste my answer from that one.
    – mattdm
    May 28, 2013 at 21:10
  • @mattdm, I loved your response on the other post. That answers quite a bit and distills the information very well. For me to put a "cap" on what I'm trying to get answered here, I would need to flip my question around a bit: Why would anyone want to shoot with florescent? What are some good use cases? May 28, 2013 at 21:26
  • 1
    I think it boils down to: they're relatively cheap, they run cool, they're energy efficient, and they are good enough for many situations.
    – mattdm
    May 29, 2013 at 2:27

5 Answers 5


There are two things at play. One is the spectrum and the other is the shape of the light-source.

If you place both types of light in a soft-box, you will be wipe out the difference in shape which will make the harsh look and appearance of imperfections, bumps, wrinkles match more closely.

The spectrum of an incandescent bulb is more more uniform and evenly distributed than that of a fluorescent. This means that all colors are present and so a subject can reflect its true colors. However, the color-temperature adds a bias which you can shift by correcting for white-balance. This happens since an incandescent bulb produces more light which we consider warm. Different tints painted on the bulb often shift this towards cooler color to produce cool-white or daylight bulbs.

The spectrum of a fluorescent bulb is highly irregular and some colors are indeed missing or very low, depending on the exact type of fluorescent. When it happens, the missing colors cannot be reflected back by the subject and no amount of white-balance correction can compensate.

Since fluorescent are more and more used, there are now many types of lights which offer improved coverage of the color-spectrum. You can even buy full-spectrum lights that output colors covering a very similar spectrum as the sun, including UV.


Now, my question is: Is this due to the TYPE of lighting (Florescent vs Incandescent), or is this a function of the lighting temperature typically found in Florescent as opposed to Incandescent?

The effects have little to do with colour temperature, except in extreme cases you can always calibrate images for colour temperature provided it's reasonably consistent across all lightsources.

Instead it is to do with the TYPE of lighting, as you suggest but also, when it comes to fluorescents, what the light was designed to do. You standard office strip light is designed to be cheap to run. These tubes often have very poor spectrum missing a lot of red which renders skintones with an unhealthy looking green cast. This is why people say "I look horrible when seeing my reflection in the mirror when in a room that's lit with florescent lights".

But there are other bulbs manufactured for different intents, including those designed for as natural a spectrum as possible. You'd have to expect that lights sold for use in a soft box would be of this type, or at least a lot better than your average office strip light.

  • Any high-quality fluorescent will list a CRI score (see What is CRI?), which for photography should be at least 90 and preferably mid-upper 90s.
    – mattdm
    Oct 28, 2013 at 18:12

As far as quality of image goes, if your light sources and your camera are balanced properly for the color, it isn't a problem. Mixing different temperature lights can be a problem due to color casts. Ie, you balance for one light and the other one casts a color.

The quality issue difference between CFL and Incandescents arises due to differing CRI or color rendering index. @Michael Clark touched on this by noting the parts of the spectrum being emitted or not.

The CRI rating of a CFL can be higher than an incandescent, however.

The reason why people say they look horrible under CFL is because of the slightly blue/green tint from poor quality CFL lights and because of the missing bands, effectively highlighting certain parts of the skin.

CFL bulbs and Incandescent bulbs of the same color temperature and the same CRI level will yield similar results.

Your best bet is to avoid mixing and matching, since as CFL and incandecant bulbs age, they shift in color temperature slightly, and will start resulting in color casts being generated or slight exposure differences in the case of CFL circuits starting to fail.

If you are stuck with a mixed set, your best bet is to take test shots and have gels handy to make one light source closer in color temperature to the other light source. Also, if you find that one light source seems to make things "look wrong", then use it to light a less significant portion of the scene.


It is a little of both. Most florescent bulbs are nowhere near full spectrum light. That is, they don't emit all of the wavelengths of visible light the way incandescent bulbs do. In the case of CFL bulbs tuned for 5,500K, they are usually much better than your run-of-the mill bulb made for an office fixture.

This video tutorial shows how to make your own lights using components, including tube type bulbs, from the local hardware store and how to shoot effectively with them. The results, for no more than he spent, are impressive. The biggest thing to keep in mind when using theses type of florescent bulbs is to use a shutter speed of no faster than 1/125 second (in areas where the electricity is 60Hz) to smooth out the flicker of such bulbs. As the alternating current cycles back and forth through the gas in the bulbs, both the intensity and the color of the light emitted shifts. By using a shutter speed long enough to capture at least half of a full cycle of a/c, the entire spectrum the bulb emits will be captured by the camera in a way that our eyes can not see it.

  • 1
    Just a note: the mains (wall) frequency rule of thumb is a little out of date now. While it's still true of magnetic ballast fixtures (standard T12-type), most modern compact bulbs and T8 fixtures use an electronic ballast operating at a higher frequency (usually in the 100 to 300Hz range) to eliminate visually-perceptible flicker, so you can usually get away with a higher shutter speed than 1/125s now (up to, say, your X-sync speed, but you'd need to test).
    – user2719
    May 28, 2013 at 19:20
  • Thanks for the update, Stan. Since the frequency of the a/c varies from country to country, as does the age of recently purchased ballasts, it is always a good idea to test each setup. By far most of my experience with light flicker is in sports stadiums where anything faster than 1/125 is still plagued by the flicker of the artificial lighting in all but one of the venues I frequent. Of course 1/125 is too slow for the action I'm shooting. At higher Tv the rolling shutter means 1/2 the frame is often caught with the lights peaking and the other half with the light bottoming out.
    – Michael C
    May 28, 2013 at 19:54

Light is a complex thing made of many different wavelengths of light. Color temperature is an oversimplified measure that describes the apparent color of "white" light. The actual frequencies that make up light can differ greatly though, even for light that is apparently similar in temperature.

The harshness of florescent light is both an artifact of how it is emitted (from a broader area), the temperature (the apparent color) and the light frequencies actually emitted (the actual frequencies of light produced).

Gels or filters can be used to match the color temperature of a florescent light and an incandescent light, but it will not replace the missing frequencies of light. Alternately, you could gel them both in such a way that only common frequencies of light are allowed through, but I'm not sure what the impact on image quality would be. Simply matching temperature will probably give the best result short of swaping out to use two similar light sources as the common pool of frequencies would probably be too much loss of light.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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