Category: Ask Tim Grey

Today’s Question: As I understand it, the color temperature for light is rated by the Kelvin scale. The higher the number, the cooler or more blue. The lower the temperature, the warmer or more red. Yet we photographers with Lightroom and Photoshop seem to be using the Kelvin scale in reverse. Lowering the Kelvin temperature makes the image cooler and more blue, whereas increasing the temperature warms the image. What am I not understanding about how the Kelvin scale is applied to color and light temperature?

Tim’s Quick Answer: When it comes to the Temperature slider in most imaging software, the reason the adjustment seems backwards is that you are choosing a color temperature to compensate for, not choosing what color temperature the overall color in the image should look like.

More Detail: First off, your overall understanding of the Kelvin scale is correct. A low Kelvin value (temperature) reflects a warmer color tone (more yellow, for example). A higher value reflects a cooler color tone (more blue, for example). This scale is based on the behavior of a theoretical “black body” that radiates heat in the form of visible light as it is heated up.

Of course, it is worth noting that what is really backwards here is our definition of color values in what I refer to as a “psychological” way. We refer to colors in the red and yellow range as being “warm” colors, and we refer to colors in the blue range as being “cool” colors. But in reality, hotter objects emit what we would refer to as cooler colors. So our psychology of colors could be described as “backwards” to some extent.

When referring to the color temperature of visible light, the Kelvin scale is often used to describe the color of that light. So a warmer (or more yellow) color would have a relatively low Kelvin value, while a cooler (or more blue) color would have a relatively high Kelvin value.

In most imaging software, reducing the color temperature value cools off the image, while increasing that value warms up the image. This certainly seems backwards based on the actual Kelvin scale for color temperature.

However, the color temperature adjustment is actually allowing us to define what color temperature we want to compensate for. So, if the light illuminating a scene was very blue, we would shift the color temperature adjustment toward blue. In essence, we are saying, “The light was very blue, so please shift the image toward yellow to compensate”.

In other words, when you adjust the color temperature Kelvin value up or down, you’re really attempting to define what color the light actually was, so that an appropriate compensation can be applied.

So, if you think of the white balance adjustment as enabling you to define what color temperature the light was in order to produce appropriate neutral values in the image, I think the slider will make more sense.

Today’s Question: I just watched one of your videos from a B&H Photo presentation and thoroughly enjoyed it very much. I focused in on your discussion about the neutral density (ND) filter and I’m convinced that I want to get one. So, what do you recommend? Variable or solid, and what brand?

Tim’s Quick Answer: My personal preference is to use solid neutral density filters, rather than variable. In terms of brands, I’ve been very happy with filters from Singh-Ray and I consider B+W to offer excellent quality at a good price point. However, I’ve also recently discovered Breakthrough Photography filters, and am very impressed with what I’ve seen so far.

More Detail: There is certainly a convenience factor when using a variable ND filter, since you can achieve a range of densities with a single filter. However, it can be challenging to achieve just the right density or exposure. For me those challenges slow down my workflow (and sometimes frustrate me). Therefore, I prefer to use solid ND filters, so that I always know exactly how many stops of light I’m blocking for the exposure.

For a long time I’ve considered Singh-Ray filters (http://www.singh-ray.com) to be among the very best in terms of quality. These filters do tend to be a bit more expensive than those from other companies, but the quality in my experience has been excellent.

I’ve also been a fan of the filters from B+W, such as the 10-stop filter I’ve been using for some time now (http://timgrey.me/bw10stop). I have found the filters from B+W to be of excellent quality, while being offered at a competitive price.

More recently, I’ve been introduced to the filters from Breakthrough Photography, and I am very impressed. A quick test has demonstrated that the ND filters from Breakthrough Photography are very neutral, and the quality is excellent. You can find an example of one of the ND filters from Breakthrough Photography here:

http://timgrey.me/tgtv-bp-nd6

Today’s Question: In all my years of reading your newsletters I’ve never heard you address the question of file degradation due to resizing for printing. If a person has a finished PSD image file and uses that file to print various size prints and at the end chooses to “not save” at any particular image size, will that process after a short number of times begin to degrade or muddy the pixels of that original PSD file?

Tim’s Quick Answer: Not at all. One of the great advantages of digital photography is that an image file will not degrade over time. As long as the actual image file remains intact, you can produce output from the same image file an unlimited number of times without degrading that image.

More Detail: In the days before digital photography, we didn’t have the ability to protect our images over the long term the way we can with digital photos. While many film images (or prints) were rather stable, there was still the risk of slow degradation or damage over time. With a digital file, as long as that file is preserved the contents of the file will not change.

There are other aspects of a photography workflow that can impact an image. For example, if you open a JPEG image, apply changes, and re-save the image file, you are degrading that image to some extent due to the updated processing of the image file for purposes of JPEG compression.

But in the context of a workflow that involves simply opening an existing image, processing that image to produce a print, but then not saving the changes to the original file, the original image file remains in its original state.

Naturally you need to be careful about backing up images, ensuring that the storage media on which your images is stored does not become obsolete, and other considerations. But in terms of the file itself, as long as it is preserved, so to are the contents of that file.

Today’s Question: My understanding has been that prints made on matte surface, inkjet printing papers are generally more archival than glossy surface papers. Do you agree? And how, if at all, does the coating on coated matte papers influence the print’s life expectancy?

Tim’s Quick Answer: There are a variety of factors that impact overall print longevity. However, in general matte papers will provide the greatest longevity. Glossy papers will generally provide the least longevity, and coated matte papers can be expected to provide longevity that falls somewhere in between.

More Detail: Let’s first assume that all other factors are equal in this case. That’s not a small assumption, of course. There are factors related to the composition of the paper substrate, the degree to which a given paper is acid-free, and many others.

However, among the various factors that impact print longevity is how close the inks ultimately sit to the surface of the paper. The closer the inks are to the surface, the more exposed those inks will be to environmental factors.

Glossy papers generally cause inks to stay at the surface rather than being absorbed. As a result, the inks are more exposed to environmental factors that can cause the inks to fade over time. In other words, with a glossy surface there is a greater risk that the print will fade relatively quickly.

Matte papers, on the other hand, absorb the inks so that the inks can be thought of as somewhat “sitting below the surface” of the paper. That helps to protect the inks, so they are not as prone to fading. Of course, that also reduces the overall color saturation and tonal contrast of the print.

A coated matte paper will generally fall somewhere in between. The matte paper itself is more absorbent than a glossy paper would otherwise be. However, the coating causes the paper to behave a little more like a glossy paper than a matte paper. Thus, in general I would expect coated matte papers to have a print longevity that is longer than that of a glossy paper, but shorter than that of an uncoated matte paper.

Today’s Question: I wondered if you could settle a curiosity for me. Do you know what the rubber rectangle on my camera strap is for? It almost looks like it could fit on the flash hot shoe, but it doesn’t. I can’t figure out what it is for. Help!

Tim’s Quick Answer: The rubber device you’re referring to is a cover for the viewfinder, used to prevent light from entering the viewfinder. You generally need to remove the eyepiece on the viewfinder in order to attach this cover.

More Detail: It seems to be conventional wisdom that during long exposures at night, you should cover up the viewfinder to prevent light from getting into the camera and causing aberrations. In my experience this is not an issue at all. There have certainly been cameras in the past (think large view cameras) where light was an issue in this way. But as far as I’m concerned this is not a concern for modern digital SLR cameras.

However, there is another reason you may still want to cover the viewfinder for night photography (among other situations). While light coming in through the viewfinder won’t impact your photos, it can have a significant impact on your camera’s meter. If you’re using an automatic or semi-automatic exposure mode, or, depending on the camera’s meter to measure the exposure for a night photo, it is a good idea to cover the viewfinder. Of course, it would also be fair to point out that the camera’s meter isn’t incredibly helpful with a lot of night photography anyway.

So, while light can most certainly come in through the viewfinder and affect your camera’s meter, you don’t need to worry about that light in terms of the actual image being captured. More importantly, now you know what that little gadget is on your camera strap!

Today’s Question: Our photography club said not to use IS [image stabilization] on a tripod. With absolutely no movement, the lens will start vibrating to create motion that it can stabilize. I read this somewhere else too, but most people don’t seem to know.

Tim’s Quick Answer: This is one of those bits of advice that has become slightly out of date. With some lenses you can indeed leave image stabilization turned on even when using a tripod. Other lenses include stabilization modes specifically designed for situations where you are panning with a subject.

More Detail: With most older lenses that include image stabilization, it is absolutely true that you should leave the stabilization feature turned off when you are using the lens with a tripod. Otherwise the stabilization technology can actually cause movement that translates into a blurred shot, defeating the whole purpose.

Many newer lenses, however, have stabilization technology that works well even when using a tripod. Some lenses even include special features that enable image stabilization to effective when panning. For example, Canon has incorporated a “Mode 2” option for some of their lenses that is designed for use when panning.

The key is to make sure that the lens you’re using has image stabilization technology that works well with a tripod. But in general you’ll find that with modern lenses the advice to turn off image stabilization is no longer critical advice.

As a general rule, with a static subject when photographing with the use of a tripod, I myself am still in the habit of turning off image stabilization. But in truth, most newer lenses with image stabilization technology (going back to 1999 in the case of Canon lenses, for example) can be used with a tripod even with image stabilization turned on.

Today’s Question: As a part of my workflow in Lightroom, I hold down the Option key [Alt key on Windows] and move the masking slider to adjust the area in the image that I want to sharpen. Sometimes when I am doing this, I think that the result that I see would make an interesting abstract. I think this must be a little like the posterization effect in Photoshop, but when I try to use that approach, it doesn’t seem to produce the same outcome I get with the masking slider. What is it that is actually happening in Lightroom with the masking slider and how could I best reproduce that effect in either Lightroom or Photoshop?

Tim’s Quick Answer: The effect you’re seeing is somewhat similar to a result you can achieve with the Find Edges filter in Photoshop. The approach involves using the Find Edges filter and then creating additional contrast and possibly inverting the result.

More Detail: To get started, open the image in Photoshop. Then create a copy of the Background image layer. From the menu bar choose Filter > Stylize > Find Edges. There are no parameters for this filter, so you’ll simply see the result in the image.

Next, add a Hue/Saturation adjustment layer, and reduce the Saturation completely. This will remove the color created as part of the Find Edges filter, so that you have a black and white version of the image.

You can then add a Levels (or Curves) adjustment layer and drag the black and white point sliders inward to create extreme contrast for the photo. You will generally want to bring these end points very close to each other, possibly all the way until those end points are touching each other. Pay attention to the detail within the photo to help choose where those two end points should meet.

You can then add an Invert adjustment layer if you’d like to reverse the areas that appear black versus white, which will provide the same basic result you were seeing in Lightroom. You may also want to apply a slight blur to the Background Copy layer, using the Gaussian Blur filter.

Note that in some cases you may prefer the effect produced by first working with a lower-resolution image, and then increasing the size of that image to exaggerate the size of the lines created by the effect.

It is worth noting that you could also simply produce a screen capture of the effect directly from the Develop module in Lightroom when using the preview feature (by holding down the Alt/Option key on the keyboard) while adjusting the Masking setting for the sharpening effect. Because the image is abstract, having a relatively low-resolution screen capture won’t have a significant effect on the appearance of the final image.

Today’s Question: I rented a fisheye zoom lens (Canon 8-15mm) to try out, because I’ve never used one. I am not getting the circular images that I was expecting. At the widest zoom setting there is a little vignetting, but the image isn’t a circle. How do you achieve the circular image that to me represents a fisheye photo?

Tim’s Quick Answer: It sounds like you are using a camera with a “cropped” sensor, meaning a sensor smaller than full-frame. To achieve a circular fisheye image, you need to use the lens on a full-frame camera.

More Detail: The Canon 8-15mm fisheye lens (http://timgrey.me/atg815fish) is indeed a genuine fisheye lens, capable of capturing the classic circular images with an extremely wide perspective. I should clarify that the image itself will still be rectangular, but the circular fisheye image will appear within that rectangle, with the rest of the image area being black.

A fisheye lens like this projects a smaller image circle than other lenses, enabling you to capture the circular image effect that is commonly associated with photos captured with a fisheye lens. However, this lens is designed for a full-frame sensor. A smaller image sensor will crop the image circle, so that you aren’t able to see the full circular effect. Instead, at the shortest focal length you will simply see some vignetting in the corners of the image.

This lens includes a “Limit” switch, enabling you to lock the lens so that you can’t zoom out beyond a 10mm focal length. At focal lengths from 10mm onward, you won’t get the vignetting at the corners of the image, because the image circle is effectively being enlarged to more than fill the area of the smaller sensor.

In this case if you want to achieve circular fisheye images you have two options. You can either work with a full-frame camera using one of the various fisheye lens options available, such as the 8-15mm noted above. Or you can use a fisheye lens designed for cropped sensors, such as the Sigma 4.5mm (http://timgrey.me/atgcropfish).

Today’s Question: You had recommended setting your camera [the Canon EOS 7D Mark II] to record to two memory cards at the same time. My camera has this feature, but when I capture video it is still only saved to one card. Photos are going to both cards. How do I fix this for video?

Tim’s Quick Answer: You can’t fix this on the 7D Mark II (or as far as I know on any other digital SLR). While photos can be saved to two cards at once thanks to two cards slots on this camera (among others), you can’t save video to two cards at once.

More Detail: While you can’t record video to two cards, you can choose which card the videos get saved to. The result is that your photos will be saved on both cards, but your videos will only be saved on one of those cards.

In general I would tend to use the fastest media card of the two for saving video. However, I actually opt for the SD card rather than the CompactFlash card as the card that my videos get stored to. That is because my primary computer has a built-in SD card reader. In other words, I prioritize convenience in my overall workflow rather than the performance of the card. And, of course, I use relatively fast cards in any event.

In the case of the Canon 7D Mark II you can choose which card to record videos to from the menu on the LCD. Under the “Set Up 1” menu you can choose the “Record fun+card/folder sel.” option. Then, with the “Record func.” option set to “Rec. to Multiple”, choose the Playback setting and choose the card you want to use for playback. That card will then also be used as the card that video is recorded to.

Today’s Question: You made reference to setting a custom white balance in the camera based on a gray card. How can this be done?

Tim’s Quick Answer: The specific process will vary from one camera to the next, but in general that process is the same. You simply capture a photo of a gray card (or similar neutral object) and then use that photo as the basis of a custom white balance setting in the camera.

More Detail: The concept here is that if you capture a photo of a neutral object (such as a gray card) under the same lighting conditions as the subject you will be photographing, the color of that object as it is captured can be used as the basis of an automatic white balance adjustment for subsequent photos captured under the same conditions.

While a gray card is certainly a great tool to use in this process, it is also possible to employ any other neutral (non-colored) object, such as a bright white sheet of paper. Perhaps the most important part of this process is to be sure you are photographing the gray card (or alternative) under the same lighting conditions that will illuminate your subject.

In other words, this process works best when the lighting on your subject is relatively consistent and uniform. You can then place the gray card in that light, and fill the camera’s frame with the gray card. Once you have captured a photo of the gray card under the lighting conditions for your subject, you can set that photo as the basis of a custom white balance adjustment.

Again, the specific process will vary by camera model, but in general you can simply choose the custom white balance option within your camera’s menu system and then select the appropriate photo as the basis of that custom white balance.

I should hasten to add that this custom white balance option is different from the similarly named option that enables you to dial in a specific Kelvin value for the white balance compensation. The process described above offers a relatively automated approach to compensating for the color of light on your subject through the use of a “sample” photograph captured under the same conditions.