Thursday, December 31, 2009
2009 is fast coming to an end and I wanted to get at least one more post out the chute before 2010 arrives. Today's post is a bit of a departure from the more pedantic nature of past postings. I thought it about time to start showing what can be done with some studio equipment in real world use. Keep in mind that by my way of thinking, a studio is any space where I can control the lighting to make photographs. While that includes the use of natural light, here in the greater Seattle area we have learned not to depend on the sun, so artificial lighting is more a way of life than a convenience.
How I Started with Off-camera Flash
Saturday, November 28, 2009
The last post dealt with color shift from light modifiers. This prompted me to conduct yet one more similar experiment, only this time using light bounced from my collection of various reflectors. suspicion that some of my reflectors are causing a color shift was fueled by the findings from testing umbrellas and softboxes. Since reflectors get used primarily for filling shadows and often in conjunction with a fill light, the color shift is often not particularly noticeable and therefore easy to overlook. However, when using a single light source and a reflector for fill, the color shift could be very dramatic, so if there is a problem with any reflector, it needs to be identified and eliminated.
Wednesday, November 11, 2009
Color Shift from Light Modifiers
Why it pays to buy quality
In this posting we are going to explore the difference between some light modifiers of various styles and brands. This is not a review or an endorsement for any particular product. Instead, we are going to take a look at how different manufacturers light modifiers affect the color temperature of light. This important aspect of light modifiers seldom gets any kind of attention, yet it is just as important as any other characteristic, especially when mixing light modifiers. My goal is to point out just how dramatic the differences can be. There also appears to be an implication that the savings from buying cheap Chinese products, does come at a price.
Monday, November 9, 2009
Now, let's get back to business. Today we are finishing up by setting a manual white balance for a raw file using Bibble 4.10 Pro.
Tuesday, June 30, 2009
The Manual Controls
Most raw processing software offer at least two controls for altering the white balance manually. One is Temperature and the other is Tint. Temperature is used to shift the color from warm to cool (essentially from red to blue). Usually, the control is a slider that you slide left to cool down a warm photo, and right to warm up a cool photo. The control displays an approximate color temperature, which should match the temperature of the light source once a proper white balance is achieved.
From previous postings we know that incandescent lights emit a very warm light, so for a photo taken with incandescent light we would slide the temperature control to the left to lower the temperature. Shade produces a cool, high temperature light, so for a photo taken in shade we would slide the control to the right to raise the temperature.
The other control you will find is for Tint. This control shifts the color between green and cyan and is used to level the RGB values after the color temperature has been set. This is needed because the variability in spectral content between light sources.
So let's take a look at how we can set the white balance manually using DPP. In this first image we are looking at the shot of the white balance card from earlier in this series. The light source was tungsten, so the appearance is heavily biased toward red (the color temperature is very warm).
We begin the process of finding the white balance by first setting the adjustment mode (circled at the upper right of the image) to “Color Temperature” as shown in the next photo.
Next, we move the temperature slider until the white patch appears close to white. Measurements are taken by simply moving the cursor over the white patch. Note that the measurements will vary a little from spot to spot, but one quickly gets an idea of just how close to a neutral white the color is getting. Remember that neutral will be when all three of the RGB values are identical (or in this case, close).
I was able to get a reading of 157, 148, 149 by only adjusting the color temperature slider. This is not bad at all, but why stop there when we can shoot for something better? With DPP we can fine tune the color by pressing the “Tune” button.
The “White balance fine adjustment” window pops open and in it we can see a color wheel that covers the rainbow around its perimeter, with a graduated fading to white toward the center. In the center is a a small movable marker. Moving the marker toward any color will add that color to the image. The closer it is moved toward the perimeter, then the more of that color gets added. Note that the colors for opposing sides of the color wheel are opposite colors. This gives you a visualization of what is happening with your image as you move the marker. Moving it toward red, adds red and subtracts cyan. Moving it toward blue, adds blue and subtracts yellow, and so on.
Since the RGB reading of 157, 148, 149 indicates a bit too much red, I moved the marker just a little towards cyan. Unfortunately, Canon did not think to provide a way to measure the effect on the image while the White balance fine adjustment window is open, so to test the impact of the adjustment, this window must be closed so a measurement can be made. Clicking “OK” will close the window without loosing the adjustment setting.
I performed a series of adjustments by opening and closing the White balance fine adjustment window until getting the near perfect white balance seen in the next image. The reading of (155, 156, 157) is very good, so I leave it at that.
One of the advantages of using raw files is that you can set the white balance for all files shot under the same lighting. To make this easy, Canon have supplied a way to save the image settings (which Canon calls a “recipe”) into a file. Select the Edit menu and then Save recipe in file... as shown. You can give the recipe any name you like.
Since I have a photo that was taken under the same lighting as the white patch, I want to apply my saved recipe to it. I open the photo in DPP and then to make use of the recipe, select the Edit menu and then Read and paste recipe from file...
The file has inherited the saved settings and the result is a perfectly white balanced image.
If there are several for which you want to apply the same recipe, this can be done from DPP's main window without opening a raw file. Simply select the images (holding down the control key lets you select multiple individual images) and then access the Edit menu and then Read and paste recipe from file... and select the previously saved file. All of the selected files will inherit the saved settings.
I can see that this post is getting kind of lengthy, so I'm going to break it apart. Next time I will cover manual white balance using Bibble Pro. Until then...
Continue on to Temperature of Light - part 9 (Manual White Balance with Raw Files using Bibble 4.10 Pro)
Saturday, June 6, 2009
Welcome to another installment of Studiography! It has been a few days since you last heard from me, sorry for the delay. This was supposed to be the last installment for the series on the temperature of light, and I think it is about time to wrap things up, but after writing the material for this post on using the automatic method for obtaining a custom white balance, I realized that I also need to discuss the manual white balance controls, so I will finish up with one final post after this. If you are new to the blog, please feel free to peruse the earlier postings. I suggest going back to the beginning of the series on “Light: In the Beginning” and read forward from there. The first post is a bit dry, but they do get better after that :-) All of the archived postings contain a link to the next post, so you can easily read them in sequence.
Raw Files (background info)
This post involves working with raw files, but at this time we will only be looking at white balance. For anyone unfamiliar with the term raw file, it is a file composed of data read directly from the camera's sensor. Most digital cameras use a monochromatic sensor, so in order to create RGB (Red, Green, Blue) color channels, the sensor is fitted with a color filter array composed of red, green and blue filters arranged in a symmetric pattern with one filter element for each photo site on the sensor. A program called a raw converter is used to interpolate a color and intensity for each photo site by using readings from surrounding pixels to create a jpeg or tiff file.
A study of the color filter array will yield a better understanding of what a raw file contains. There are several good articles already written on this subject and it is far beyond the scope of this posting, so rather than replicate what already exists in abundance, here are some links to follow for more information. DPReview, Wikipedia, Harvard, and John Savord's excellent writeup. An Internet search will produce a proliferation of similar reports.
All DSLR cameras and many digicams can output raw files either along side or in place of jpeg files. These files contain all 100% of the data from the sensor with no loss from compression or any other processing. In order to make use of these files, they must be processed by a raw converter. The camera manufacturer usually includes raw converter software as part of the package, but many people prefer the performance and/or features found in third party software, such as Adobe Lightroom, Adobe Camera Raw, Bibble, Capture One, et al.
For a number of years there has been a debate about the benefits of raw -Vs- jpeg files from the camera. At the end of the day one ends up with a jpeg file either way. However, the proponents of raw say that raw offers much more flexibility for processing, especially for white balance options. The jpeg shooters argue that in-camera jpegs are high quality and with features like picture styles, eliminate the need for shooting raw. In addition, in-camera jpeg files save a lot of processing time. For the record, I have been shooting raw for years and from my experiences, am thoroughly convinced that raw is the better work flow for me. That is all I'm going to say on the subject for now.
Custom White Balance (Intro)
So, let's get down to business. Virtually all raw converters offer a custom white balance tool. For the most part, all you need to do is select the white balance tool (eyedropper) and click on something that has neutral color content for a reference. This reference can be gray or near white, but cannot be black or pure white (0,0,0 or 255, 255, 255). For the examples in this posting, I am using the white patch from a QP Card.
You will need one image of the reference and will need the exposure to be reasonably close to correct and certainly not overexposed. If you have too hot of an exposure and the white reference has one or more blown color channels, the software cannot accurately determine its color. Once you click the tool on the neutral reference, the software will establish the white balance and you are done. It is very much like setting a custom white balance in the camera. For every shot taken under the same light source, you can use the same white balance settings or simply copy the white balance to those photos. This is one of the powerful features available to the raw shooter.
Custom White Balance Using Adobe Camera Raw (ACR)
As you can see from the image above, the light source is very warm and the camera's setting is not even close to correct. To set the white balance, simply click on the White Balance tool at the upper left, then click the eyedropper anywhere on the white patch.
After balancing, you can see that the white patch is now a good representation of white. I took a color reading in the center of the patch. ACR is showing RGB values (1) of (154, 153, 151), which is very good. The “White Balance” control (2) now shows “Custom” and the “Temperature” display (3) is at 2600 degrees K. ACR does not show the tint on this screen.
Custom White Balance Using Bibble 4.10
To set the white balance, simply click on the White Balance tool at the right, then click the eyedropper anywhere on the white patch.
After balancing, you can see that the white patch is now a good representation of white. I took a color reading in the center of the patch. Bibble is showing RGB values (1) of (185, 193, 188), which is good. The “White Balance” control (2) now shows “Click White”, the “Temp” display (3) is at 2762 degrees K, and the “Tint” control is at -34, which adds a little bit of green.
Custom White Balance Using Canon Digital Photo Professional (DPP)To set the white balance, simply click on the White Balance tool at the right, then click the eyedropper anywhere on the white patch.After balancing, you can see that the white patch is now a good representation of white. I took a color reading in the center of the patch. DPP is showing RGB values (1) of (147, 152, 156), which is acceptable. The “White Balance” control (2) now shows “Click White Balance”. Oddly, there is no temperature display for DPP.
It is now past my bed time, so I'm saving the part on manual white balance for next time. Until then...
Saturday, May 30, 2009
What you need
To set a custom white balance you will need a known white surface and of course your camera. Best to use a calibrated white balance system. There are several to choose from ranging from around $10 U.S. and up. I suggest something like the QP Card, Opteka Premium Reference Color & White Balance Card Set, or the Digital Grey Kard.In a pinch, a piece of white paper will do, but try to choose one that appears less bright white because the bright white papers typically use optical brighteners that cause a blue color cast when exposed to ultraviolet light. The brighteners fool the eye into thinking it is a brighter white by adding a tiny blue bias. This works because of the way our eyes interpret color. However, your custom white balance will compensate and make things come out with a bit of a red/yellow cast. Usually this is minimal and so using a piece of paper with optical brighteners is still preferable to using nothing.
Steps for Obtaining a Custom White Balance
- Take a picture of the white patch (or white paper) making sure that it covers the center area of the frame. For a Canon 450D, it “should fill the spot metering circle.” (pg 90 of
- From the menu system (shown below), select “Custom WB”.
Select the photo of the white patch and then press the “Set” button. The camera will ask if you want to use white balance data from this image for the custom white balance. Press the “Set” button for OK.
The camera then displays a reminder to set the white balance to custom. Oddly enough, though there is plenty of screen space, instead of spelling out the word custom, they simply display the icon for custom white balance.
Press the “WB” button on the back to set the white balance type. Use the right/left arrow keys to select Custom (shown below), then press the set button.
You are now ready to shoot. If the display is active, it should show the custom white balance icon below the ISO setting.
That's it for now. In the next posting I will demonstrate how to set a custom white balance using raw processing software. I will use my favorite, Bibble 4.10, but might also take a stab at it with Canon DPP and Adobe Camera Raw.
Monday, May 25, 2009
My purpose in writing this posting, is to convince you to use a custom white balance. If you are using jpeg files from your camera, it is the only way to ensure you get correct color rendering. Some time ago I gave a white balance card to a friend and taught him how to use it with his digicam. A few months later I heard from his daughter about how much his photos have improved since he started using the “thingy” that I gave him. Why is that and why couldn't the camera just get it right? Read on to find out.
What is a Custom White Balance?
A custom white balance calibrates the camera's color rendering to the actual light being used. This works regardless of the light's temperature and it works reliably. For most cameras, you take a picture of a known white or near white surface and then program the camera to calibrate its color rendering to that white.
Why a Custom White Balance?
The short answer is because digital camera auto white balance is very unreliable. Take a look at the above images. All of these were taken under the same incandescent lighting, but using different white balance settings as indicated. The one on the bottom closely matches the actual magazine.
Why not use Auto White Balance?
Auto white balance may seem a simple function, yet few if any cameras get it right much of the time.
Take a look at the photo above. It was taken with incandescent light and the camera was set for auto white balance. This magazine cover has a good B&W print that should have been sufficient for the camera to get a decent color balance. However, as is the case for many cameras (especially Canon DSLRs), the auto white balance fell far short of the mark.
The lower photo shows the camera's display, including the RGB and luminance histograms. Notice how red dominates. The camera has chosen a color temperature that is not even close to incandescent.
Auto white balance attempts to figure out the color temperature of the light being used by searching out the brightest thing in the image. It assumes that this is a white surface and so does a color balance based on that. It sounds good on paper, but in practical use it leaves much to be desired.
The problem lies with not always having a true white reference. What if there is nothing white or even close to white in the scene? In this case the camera makes a best guess at what might be close to white and calibrates to that. In reality it does not usually get it right and sometimes it doesn't even come close. There is also the effect of color shift. Under incandescent light, something with a blueish hue might appear as white to the camera, in which case it will make a bad assumption about color. In the case above I cannot imagine what went wrong, but it certainly is not right.
Why not use Presets?
Presets for daylight and incandescent usually get you close, but quite often, not often not close enough, and sometimes they are just plain wrong. The image above was taken with the incandescent preset. The improvement in color compared to auto white balance is obvious. However, it still has a reddish hue that is not found in the real life subject. The RGB histograms from the camera show the improvement, but also show the lack of proper white balance. Because the magazine cover is essentially black and white, all three histograms should look the same.
If without a way or time to calibrate and under known lighting, I would use a preset rather than trusting auto. However, what if I'm in a room that is lit with Reveal incandescent bulbs? From part 4 of this series we know that the Reveal incandescent bulb will produce a light that has more blue than a standard incandescent bulb. If my camera is set to incandescent white balance, then I will get a blue cast in my images.
Fluorescent bulbs are even more problematic, as they have much variability from based on the manufacturer and type. They also shift color with use. Some cameras have two or more fluorescent settings, but that cannot possibly cover all of the variations one might encounter in the real world. Even if there were a preset, how would you know which one to use? When possible (as it usually is), it is always better to get a custom white balance. Daylight varies by time of day and atmospheric conditions.
What About Using a Custom White Balance?
Glad you asked because if you are going to let the camera generate jpeg files for you, this is the best way to go.
The photo above was taken after performing a custom white balance. As you can see, the image correctly depicts the black & white photograph on the cover of the magazine. A look at the RGB histograms in the next photo down confirms an excellent white balance. The simple step of performing a custom white balance has produced a near perfect rendering of the subject, enough said.
In the next posting I will demonstrate how to set an in-camera custom white balance using a Canon 450D. This tutorial should make a good primer even for non-Canon users, so until then...
Tuesday, May 19, 2009
So far in this series we have looked at what color temperature is, how color temperature is determined by spectral content, and the temperatures to expect from some common light sources. This posting gets more practical with a little experiment to show the effect of color temperature on what we get from the camera.
For this experiment, I have made five sets of photos, all with the camera set for daylight color balance. This tells the camera that we are using sunlight, so it adjusts the color balance accordingly. Using the same setting for all of the photos allows us to see the differences between light sources. The setup includes a calibrated color check chart so we can see the effect on color rendering. It also includes a QP Card, which has patches for near black, middle gray, and near white. The white patch gives a good visual representation of the light's color. The middle gray patch will be used for making RGB color balance measurements to assess each light sources color rendering.
Bear in mind that the interpretation of spectral content for sunlight is controlled by the camera manufacturer, Canon. The following images are from in-camera jpeg files. The camera is a Canon 450D. I set the camera's preset to “Faithful”, which sets all of the color and contrast controls to zero.
As you probably already know, the Seattle area has a reputation for having the most consecutive overcast days per year compared to any other state in the union. Few places can claim more cloudy days than us. We have some of the happiest slugs on earth... but I digress. What I meant to say is that by chance, it was a near cloudless day, so I was able to get outside and make the sunlight photo in bright, 11:00 A.M. sun. Sometimes things just work out :-)
Lightsource: Sunlight at 11:00 A.M. on a near cloudless day.The colors in the chart are clearly discernible and the white balance card shows what appear to be neutral white and gray patches with no obvious color cast. Using the color picker to take a measurement on the middle gray patch, I get (96, 97, 99), which is close to a perfectly neutral gray.
Our other clue to good color balance is seen in the Red, Green, and Blue histograms.
The between the Red, Green, and Blue histograms. The three histograms show the same basic distribution of pixels at about the same luminance. This is good enough to use with no additional white-balance adjustment.
Lightsource: 150W Halogen Incandescent light bulb (modeling lamp from a monolight).I didn't have any regular incandescent bulbs on hand, so I used the modeling lamp from one of my studio lights. It is no different from any normal halogen bulb except that it has an additional layer of glass to make it safer (JDD style). The spectral output is virtually the same as for an ordinary incandescent light bulb. They both burn a filament to create light. The only difference is that for one, the filament burns in a vacuum and for the other it burns in halogen gas.
Back to our analysis. Notice that the white balance card shows a distinct reddish tint and the colors in the chart are badly skewed toward red. I don't need to make a measurement to know that the color is off considerably. Nonetheless, using the color picker to take a measurement on the middle gray patch, I get (140, 80, 46), which confirms the red bias that our eyes see. This reddish look is the same that we get when we forget to set the white balance on our digital cameras when taking photos indoors. When this happens, the photographer's phrase for describing this effect is that “it was an artistic choice.”
The Red, Green, and Blue histograms tell the story. The reds are about twice the brightness of the greens and nearly four times the brightness of the blues. The spectral content of this light is not even close to sunlight.
Lightsource: 100W Reveal Incandescent light bulb (special coatings to raise color temperature)I had some “Reveal” incandescent bulbs available, so thought I would take a look to see if “Reveal's unique neodymium glass filters out dull, yellow rays unlike regular soft white bulbs...”
Visually you can see that the although the colors are still substantially skewed, reds do not dominate quite as badly compared to the unfiltered halogen bulb. The histograms indeed show a slight improvement in that the blue is now a bit stronger relative to the green and red. The difference is enough to be notice by the eye, but for photographic purposes, the difference is marginal at best. This light source is still a distant match for daylight and the bulbs still dissipate the majority of the power used as heat.
The gray patch of the white balance card measures (131, 75, 52), which confirms a strong red bias with slightly more blue content compared to the halogen lamp.
Lightsource: 23W Daylight Balanced Compact Fluorescent light bulb
I gave this a 10-minute warm-up period, as these bulbs shift their color temperature dramatically over the first few minutes of being powered on.The daylight balanced compact fluorescent has a completely different story to tell. As expected, the histogram shows a light that is much closer to what the camera expects for daylight. As you can see, it is a nicely close match to natural sunlight. Not only are we getting a white patch that looks white, the colors in the chart are vivid and easily discernible.
For the record, this is not one of the expensive bulbs that you get from the camera shop. This is an inexpensive CF bulb bought off of the Internet for around $5.50 each. If you need a decent continuous light source on the cheap, this is it.
Taking a reading of the middle gray patch I get (91, 96, 90). While not absolutely perfect, this is good enough to use without any additional white-balance adjustment.
Lightsource: 200WS Studio Flash set for f/5.6 Predictably, the daylight photo gives us a nice white that is very much like the daylight balanced fluorescent, except without the odd peaks. It is no surprise that a strobe (flash) is considered the best light source for photography. Not only does it provide a lot of lighting power for a small amount of energy, but it is very close to the spectrum of natural light. A flash can be mixed with sunlight with no corrective filters.
In this last image I have put the histograms side by side so you can more readily compare them. The RGB histograms really do tell the story.
In the next posting I will cover the various in-camera white balance settings and will compare and discuss the performance for each. It should be interesting, so stay tuned...
Sunday, May 17, 2009
Before getting into the experiment showing the effect of color temperature on a camera's output, I want to touch on histograms. If you are familiar with histograms, please feel free to skip this little side excursion. If you are not familiar with histograms, then you will need this information to better interpret what follows.
Take a look at the histograms in the image above. The black histogram on top represents only luminance. It tells us about the exposure, but contains nothing about color. Below it are three more histograms by color. These represent the luminance for each color channel. In this example, the peak that shows up in the middle indicates that middle gray is a dominant color. All three of the histograms have a similar shape, which tells us that there is a fairly even distribution of colors in the image.
In the next posting I will continue the series with a demonstration of what the camera sees when photographing under different light sources. Stay tuned...
Wednesday, May 13, 2009
The chart below shows light from beyond ultraviolet down through infrared. The visible spectrum is shown as a rainbow fountain fill from violet to red. This chart is only an approximation of the characteristics for these light sources. The colors shown are not a perfect representation, but are intended only to mark where visible light starts and stops. The numbers along the x-axis are the wavelength specified in microns and the y-axix is amplitude in unspecified units.
Looking at the chart, one of the first things to notice is that the sun's emissions (dotted black line) are strongest over the visible spectrum. This of course should be no surprise whether you are an evolutionist, a creationist, or anything in between. Our eyes are optimized for the light that is most prevalent on earth and this chart confirms this as fact.
The oldest continuous artificial light source of which I am aware, is the incandescent light bulb (dashed red line). Although it is a continuous spectrum light source, it is interesting to note just how little of this light's output falls in the visible spectrum. Its output curve is slanted heavily toward red and contains only a small amount of blue or violet. Even its strongest color within the visible spectrum falls far below its main concentration of output in the infrared. Most of the energy from an incandescent bulb simply gets expended as heat. This, along with a spectral content that is not a good match for sunlight makes the incandescent light source far from the ideal photographic light source. The amount of power required for good illumination will tax the wiring of most any building and the heat generated will cause our poor subject to perspire (or pass out).
For years photographers used incandescent lights with special coatings that improved the spectral content. Nonetheless, these lights require immense amounts of power to achieve reasonable lighting levels and they get extremely hot, to the point that many, myself included, consider them unsafe and a bad choice considering the available alternatives.
Daylight Balanced Fluorescent Light
The daylight balanced fluorescent bulb is one of the wonders of our modern technology and actually does a nice job of lighting. This continuous output light source outputs a continuous spectrum of light that closely mimics the visible spectral output of the sun. The intensity tends to be a bit low, but you can see from the chart that it has a basic curve that is similar to sunlight. The balance of emissions for a properly designed daylight bulb is close enough to a sunlight that you can use the sunlight setting on a digital camera's white balance with very good results. Note that very little emission falls in the infrared, which is indicative of the excellent efficiency for this type of lighting. As light bulbs go, these operate with very little heat.You will note one small anomaly with this light source. There are two large peaks in the output, one at about 430 microns and another at about 540 microns. Fortunately, our eyes tend to average out the light's emissions and so we don't notice these peaks. Because one does not encounter pure colors in real life, the peaks do not significantly impact the light balance for a digital camera.
Overall, my view is that a daylight balanced fluorescent light source is well suited to photography. The negatives for this light source are that the intensity degrades and the spectral output tends to shift over the life of the bulb. Also, the intensity cannot easily be varied, so most light fixtures that are based on this technology, use arrays of bulbs that get switched on or off to obtain a desired output level. This of course limits the number of settings and increases the expense.
A studio strobe outputs a very bright pulsed light with a continuous spectrum that closely matches daylight. On account of this, it is one of the best artificial light sources for photography. Because it is pulsed, the energy required is very low compared to a continuous light source. The power output for a studio strobe can be accurately controlled over a broad range of power settings. The significant disadvantage is that the photographer cannot see the lighting until after making the exposure and processing the resultant image.
To allow the photographer to visualize the lighting, most studio strobes come equipped with incandescent modeling lights. These are much lower power than would be needed for most purposes, but allow one to at lest see where the shadows fall. Typical wattages are around 150 – 250 W. They tend to generate a fair amount of heat, as is common for all incandescent lights. In the future, we will likely see then incandescent modeling lamp replaced with LED based modeling lamps.
For more information on artificial lighting sources, visit the National Lighting Product Information Program website. http://www.lrc.rpi.edu/programs/NLPIP/lightingAnswers/fullSpectrum/abstract.asp
In the next postings I will begin to demonstrate the effect of using the light sources that have been discussed and how one can effectively achieve a proper white balance whether shooting raw or jpeg.
Continue on to "Temperature of Light - part 3 (An Introduction to Histograms)
Monday, May 11, 2009
This next set of postings will deal with the temperature of light. This is a rather broad subject and really warrants some additional reading, but I will attempt to offer up a very basic background of the physics and touch on the more common issues with how the temperature of artificial light affects photography.
In the first post of this series on light I touched briefly on what light is. In this last section we are going to be discussing an aspect of light that is very dear to photographers, namely the temperature of light. This is sometimes referred to as the color of light, but it more properly refers to the spectrum of light.
Any light source, with the exception of a laser, will emit a variety of colors, some more than others. When the source approximates the spectrum emitted by the sun, it is said to be a full spectrum light source. The sun emits light in all of the visible colors, but not all of the colors are emitted at the same intensity. Sunlight is also affected by the earth's atmosphere, so the light that we receive on earth is certainly not the same as the sunlight in space, nor is it consistent from day to day. In fact, here in the Seattle area it is not consistent from minute to minute due to our ever changing weather. However, on clear days, the emission of light from the sun is relatively consistent, and so this is what gets used as our standard for a model of sunlight emissions.
The Wikipedia tells us that “Color temperature is based upon the principle that a black body radiator emits light whose color depends on the temperature of the radiator. Black bodies with temperatures below about 4000 K appear reddish whereas those above about 7500 K appear bluish.” This reference is based on the kelvin temperature scale and is commonly used to describe light for photographic applications. Essentially, you heat up a black body and at some point it starts radiating in the visible light spectrum, starting with red. The more it is heated, the more the emissions spread toward blue, violet, ultraviolet, and beyond. If we heat it to 6000 K, then we get light emissions that approximate noon day sunlight. Go a littler hotter, say 7500 K and the light gets a somewhat blue. Go a little lower, say 4500 K and the light gets somewhat reddish. Do note that in spite of the temperature, blue is considered a cold color and red is considered a warm color. This comes from the mood of the color and has nothing to do with its kelvin temperature. For more information on color temperature, visit the Wikipedia or do a Google search.
Here is a chart showing the emissions of sunlight (approximate) for a cloudless day. Note that its emissions extend into the ultraviolet on the left, and far into the infrared at the right, but the majority of its spectral intensity is concentrated within the visible spectrum.
Light sources are rated as either continuous or non-continuous. A continuous light source will reproduce light across the entire range of visible light, from red to violet. A non-continuous light will produce only some of the colors from the visible range. An example of continuous light is a standard incandescent light bulb. While it is very weak in blue light, it does contain enough light across the visible range that our eyes are able to properly identify colors when illuminated under this lighting. An example of non-continuous light is a sodium vapor lamp, which produces mostly an orange/yellow color and has virtually no blue color. If you view a color chart under a sodium vapor lamp, you will see a very poor rendering of color, such that everything will have a distinct orange tint and many colors will not be discernible.
When we consider light sources for photography, we obviously want to use a continuous light source so all of the visible colors can be seen. Our eyes are optimized for sunlight and consequently, our digital camera's sensor and image rendering has been optimized to mimic the way our eyes see. Because of this, we would like a light source that closely matches the emissions of the sun, within the visible light spectrum. Ideally our light source will contain little or no ultraviolet and little or no infrared. The camera's sensor is sensitive to both infrared and ultraviolet. Most digital camera sensors use filters and coatings to reduce the absorption of light outside the visible spectrum, but the effectiveness of these filters vary and there are trade offs with light loss and cost.
In the next post we will take a look at some common light sources and discuss their characteristics and suitability for photography.
Wednesday, May 6, 2009
I spent six hours (plus travel time) at the Bellevue Hilton Inn last night to see David Ziser's “Digital WakeUp Call” seminar and I must say that David offers an impressive amount of information and goodies for the money. For me it was worth every penny and much more. David has many years of wedding photography experience and shares his approach to shooting, work flow, and business. There is a lot of information that transfers to studio photography, if not directly, then in concept. I highly recommend that you attend this seminar. At a minimum I guarantee that you will glean something of value. Often it is the little things that make the big difference. Go if for no other reason than to win a door prize. They gave away literally thousands of dollars worth of prizes and will be awarding a grand prize some time during the tour. Now, your's truly did not win any prizes, but happily, everyone came away with the following:
- One year membership to WPPI (Wedding & Portrait Photographer's International), which is a $100 value.
- DVD with 4-hours of extended tour content,covering lighting, lightroom, photography, Photoshop, marketing, & software, with tutorials from Scott Kelby & Matt Lkoskowski, David Ziser, and some vendors (show sponsors).
- Free offers and discounts from PPA (Professional Photographers of America), Zookbinders, Animoto, NAPP (National Association of Photoshop Professionals), SendOutCard, American Color Imaging, Professional Photographer Magazine, & BellaGrafica.
Let me give you a quick rundown on the sessions.
The first session was the longest and focused on lighting and how he got the lighting seen in the many resplendent photos that he shared in the form of 24” x 36” prints and in his slide presentation. Many of these appear as if elaborate lighting setups were used that must have taken considerable time. In reality, as David shared how he got the light, we could begin to understand how very little equipment can be used to make awe inspiring photographs. All were done with just one or two portable flashes and a reflector or bounce off of some available surface (even a person's white shirt). What impressed me about this segment was his ability to see light everywhere and the potential in virtually everything. If the lighting didn't exist, he made it happen using whatever tools he had at the moment. Weddings can be fast moving, high pressure affairs, which often leave little more than fractional minutes for setup and execution of a shot. Of course, there is always the photographer's lie, just one more shot, but even that buys only a little more time. With just a handful of gear and an understanding of light, David has made wonderful portraits in moments. It isn't rocket science, just a good understanding of light and an out of the box view for the potential of anything to interact with light. This is a subject that is near and dear to me, as I'm sure it is for every studio photographer.
The next section covered silver bullet techniques that he uses to speed up his work flow. This section covers mostly the software that he uses and endorses. It gets a bit commercial, but is well worth wading through because of the demonstrations. Of particular interest to me is a program called LumaPix Extreme 4. I thought that Corel Draw was the fastest way to do free form layouts, and of course, there are things that only it can do, but LumaPix Extreme 4 does most of what Corel can do and it does it many times faster. It comes equipped with awesome templates for a starting point, and has an automated setup wizard that can create an entire album (100+ photos) in a few seconds. It appears to be the perfect tool for setting up a photo montage, poster, senior book, family book, wedding book, and about anything that needs photos and text, or I should say graphic elements, arranged in a format suited to the genre and with a very polished professional look.
The last part of the seminar focused on the business aspects of running the business. This segment covered ways to make your business more profitable. Virtually everything David covered in this section is directly applicable for the studio photographer. David discusses how to leverage our relationships with vendors and other associates to increase our visibility in the local community. As he says, “get to know the people doing business with the people you want to do business with.” I don't want to attempt to reproduce his information here, but just let me tell you that it is reliable and relevant for today with an eye toward the future. Even for someone with honed business and sales skills there is fruit to be gleaned. For the rest of us it is life giving, crystal clear water fresh from the mountain stream.
I will get back on track with the series on lighting starting with the next post, but felt that this seminar is so good that it merits interrupting the series. If you are at all on the fence about going, just sign up and be there!
Saturday, April 25, 2009
In the diagram below, both lights are the same size, but the bottom light is positioned at about three times the distance. Again I have placed vector lines at three points along the light source to represent rays of light at the points that just touches the model's face. What becomes obvious when looking at these lines is that the angle of the lines is greatly affected by distance. When the light is close, the angle is greater than when the light is farther away. In this case the close light has about 55 degrees of spread from the light rays emanating from the extreme edges of the light source. The light that is farther away has about 12 degrees of spread from those same two edges of the light source.
This concludes the discussion for the quality of light. Next post we will delve into discourse on light temperature, which will be the last section for this series.
Monday, April 20, 2009
In this next diagram I have increased the size of the light source. It is still small relative to our model, so the light is relatively hard, but it is large enough that we can start to see a slight softening of the shadow.
I have placed lines emanating from three points along the face of our light source, one at each edge and one in the center.Looking at the magnified area in the circle we can begin to see how light from the three points interact to create a feathered edge for the shadow. The background area below the red line is illuminated by light rays from all points along our light source. Light from the red point cannot reach the area of the background above the red line, so that area only contains light from the gray and blue points. Light from the gray (center) point cannot reach the background above the gray line, so that area is lit only by the blue point. Now, if you imagine countless lines representing all of the points along the face of the light source, then you can see how the shadow's feathering is the interaction of all of those lines.
In this last diagram I have increased the size of the light source to be larger than the subject. What we see is that there is now much more interaction because light from either extreme of the light source reaches all the way to the center behind our model. This results in a gradual feathering of the shadow that is considered very soft.
An extreme in terms of size, would be the sky on an overcast day, which can leave virtually no shadow.
In the next posting we will take a look at how distance between light source and subject changes the effective size of a light source.
Continue on to part 4"Quality of Light - part 4 (effective size)"