Category: Ask Tim Grey

Today’s Question: How can Lightroom export at higher than the native resolution of the file? I thought Lightroom never created pixels but apparently it can. I tested it and sure enough, a capture from my camera at 5472 pixels on the long side was successfully exported to JPEG with 6000 pixels on the long side. Who made those extra pixels and is this as good as upsizing in Photoshop?

Tim’s Quick Answer: While Lightroom does not enable you to, for example, create composite images, that isn’t the same as saying that Lightroom can’t create pixels. Lightroom can indeed create pixels, including the ability to resample an image upon export so that the exported image contains a smaller or larger number of pixels than the original capture. The quality of this resampling in Lightroom is comparable to Photoshop.

More Detail: Lightroom is non-destructive, so when you are working in the Develop module you aren’t actually replacing the pixels in your original capture. Even things that might appear destructive at first glance, such as replacing pixels to clean up blemishes, are actually non-destructive. However, I think it is fair to say that Lightroom is creating pixels (perhaps in an indirect way) when you perform this type of work.

But when Lightroom is really creating pixels is when you export a photo. That is when your original capture is processed based on the adjustments you’ve applied in the Develop module to create a new image file. So, for example, you might use Lightroom to create pixels in the form of a JPEG image based on your original RAW capture.

When exporting a photo, you also have the option to change the pixel dimensions of the photo. Based on the typical display resolution for computer monitors, when presenting an image online (such as on a website) you don’t need much more than perhaps around 500 pixels or so of width for the image. Viewed at the actual pixel dimension size, an original capture that was over 5,000 pixels wide would be too big for a web page.

Similarly, if you are exporting an image so it can be printed, you might want to produce a larger file to create a larger print. For example, if you want to prepare an image to be printed 20 inches wide at 300 pixels per inch, you need 6,000 pixels across for the photo. If the original capture is “only” 5,000 pixels wide, the image would need to be resized to produce the desired print size.

Whenever an image file is created with pixel dimensions that differ from the dimensions of the original capture, pixels must be added or removed to produce the final pixel dimensions for the new image. This is a somewhat complicated process that involves calculations to ensure that color and tonal fidelity, fine detail, and overall sharpness, for example, are retained. Lightroom (and other applications) do a generally good job with this resampling work.

So yes, Lightroom is most certainly able to create (or destroy) pixels in the context of a derivative image being created through the Export facility, by virtue of being able to resize the image being exported to specific pixel dimensions based on your specific needs for the photo.

Today’s Question: Is viewing a file at 1:1 in Lightroom the same as viewing the same file at 100% in Photoshop? I have read, and heard, others say that most images don’t look sharp at 1:1 and should be viewed at 1:2 in Lightroom.

Tim’s Quick Answer: The “1:1” zoom setting in Lightroom is a 100% view, with each pixel in the image being represented by a single pixel on your monitor display. Just as in Photoshop, the 100% zoom setting is best for evaluating sharpness, though you may want to zoom in further to evaluate fine details (such as when checking for noise). It is important to keep in mind, however, that the most accurate preview for an image will be seen in the Develop module, not the other modules.

More Detail: The previews in Lightroom’s Develop module are created differently compared to the previews shown in the other modules, such as the Library module. In the Develop module the preview is based on a full rendering of the original image, while in the Library module the previews are based on JPEG renderings of your images.

At a 100% zoom setting the differences will be very subtle, but you can see some variations with close examination of the image when switching between the Develop and Library modules.

So, the Develop module should be used for all critical image evaluation. When evaluating sharpness, you should view the image with the 1:1 setting (100% zoom) so you are getting an “actual pixels” view, where one pixel in the image is represented by one pixel on the monitor display.

For evaluating noise and other fine details where you want to be able to see those small details better and aren’t as concerned about visible sharpness, I recommend using the 4:1 (400%) or higher zoom setting. And for evaluating the overall composition and framing, of course, I recommend zooming out to the Fit option, so you can see the entire image.

Today’s Question: When I resize and resample to enlarge a print in Photoshop and want to check the pixels to see how the image might look printed I get confused. When I look at it “print size” it doesn’t appear nearly as large as the intended print. However if it looks good using this selection, does that mean it will look this good when printed? And what is 100 percent zoom for?  I feel unsure how to use these two options.

Tim’s Quick Answer: The Print Size view option is intended to provide a sense of how large the current image will appear when printed, not for evaluating image quality. To evaluate sharpness and other details related to overall image quality, the 100% (“actual pixels”) view option is best, as it ensures that each pixel in the image is represented by a single pixel on your monitor display.

More Detail: It is important to keep in mind that the Print Size option will not produce an accurate indication of the actual print size for most users, based on the default settings in Photoshop. Fortunately, you can resolve this issue so the Print Size preview actually reflects the print size for the current image.

First, you need to determine the pixel per inch resolution of your display. First, determine the actual pixel dimensions of the display within the settings for your operating system. Then measure the actual width of the display. Divide the number of pixels across by the width of the display. For example, my MacBook Air is set to a resolution of 1440×800 pixels, and the width of the display is 11.25 inches. Therefore, the pixel per inch resolution for this display is 128 pixels per inch (ppi). Note that the resolution of most displays, contrary to popular belief, is not 72 (or 96) ppi.

To adjust the display resolution setting in Photoshop go to the Edit menu on Windows or the Photoshop menu on Macintosh, and then choose Preferences > Units & Rulers. In the New Document Preset Resolutions section at the top-right of the Preferences dialog enter the resolution value you calculated into the Screen Resolution field, making sure that the popup is set to the correct unit of measure. Click OK to close the Preferences dialog.

With the actual pixel per inch resolution for your display established in the Preferences dialog in Photoshop, when you set the view option to Print Size the image will appear at the actual size that it will be printed. Keep in mind, of course, that the print size is based on the current output resolution and dimensions for the photo, which can be found in the Image Size dialog (Image > Image Size from the menu).

As stated above, the Print Size view option should be used for getting a sense of the output size, not for evaluating image quality. Instead, evaluating for sharpness and other factors related to image quality should be done at a zoom setting of 100%. At a 100% zoom setting one pixel in the image is represented by a single pixel on the display, so you are getting an accurate view of the information contained within the photo.

You can switch to the 100% zoom setting by choosing View > 100% from the menu (this option was called “Actual Pixels” in earlier versions of Photoshop). You can also establish a 100% zoom setting by double-clicking on the button for the Zoom tool on the toolbox, or by pressing Ctrl+Alt+0 on Windows or Command+Option+0 on Macintosh. Note that the last character of this keyboard shortcut is the number zero, not the letter “O”.

Today’s Question: I just learned that my camera is actually capturing in 14-bits per channel when I shoot RAW, while I’ve been working in 16-bits in Photoshop. How much detail am I missing because my camera doesn’t support 16-bits?

Tim’s Quick Answer: Most digital cameras do perform the conversion of analog information (light) to digital information (pixel values) at a bit depth of 14-bits per channel. While most software provides a 16-bit per channel mode as the “high bit” option, the difference isn’t something you really need to be too concerned with.

More Detail: With most imaging software you have the option to work in either the 8-bit per channel mode or the 16-bit per channel mode. The primary advantage of a higher bit depth is that you have a wider range of tonal and color values available for the images. This is especially important with images that require relatively strong adjustments. The greater bit depth provides more overhead so that even with strong adjustments to the image there will be enough information “left over” to ensure smooth gradations of tone and color in the image. In other words, you’ll avoid posterization.

While the high bit depth option is generally a 16-bit per channel option, you can think of this as simply being a container that allows for more than 8-bit per channel information. Some cameras convert data at 12-bits per channel, most convert the data at 14-bits per channel, and a few actually process the data at a full 16-bits per channel. However, the differences aren’t as critical as they might seem.

The 8-bit per channel mode provides 256 tonal values per channel, which translates to almost 16.8 million possible color values. It so happens that this is the number of color values that the human visual system is estimated to be able to perceive, for someone with normal vision. So it would be fair to say that for the final image, 8-bits per channel is all you really need.

However, again, a higher bit depth can be very helpful, especially when strong adjustments will be applied to the image. By starting with more color and tonal values than you actually need, you’ll still have enough color and tonal values in the image even after adjustments have been applied, so that smooth gradations of tone and color will be preserved.

For a typical image with typical adjustments, however, that doesn’t mean you actually need the full range of color values provided by a bit depth of 16 bits per channel.

At a bit depth of 16-bits per channel you have 65,536 shades of gray available per channel, for a total of over 281 trillion possible color values. That is far more color and tonal values than you ever truly need in a photographic image, considering the capabilities of human vision.

At 14-bits per channel there are still 16,384 shades of gray per channel, for a total of over 4 trillion possible color values. And even at “only” 12-bits per channel there are 4,096 shades of gray per channel, for a total of over 68 billion possible color values.

So, the bottom line is that while an analog-to-digital conversion performed at 16 bits per channel will provide more possible color and tonal values in the image, there is a diminishing return relative to the final presentation of the image. In other words, there’s no reason to have any concerns about the fact that your digital camera performs the analog-to-digital conversion at “only” 14-bits per channel.

Today’s Question: Is there a “best” way to apply a color tint to a black and white photo, or does it not really matter which method you use?

Tim’s Quick Answer: For a basic tint consisting of a single color, I would say any of the various options available would produce about the same result, all other things being equal. However, there are several options that allow you to exercise greater control and produce more interesting results, such as split toning or a gradient map.

More Detail: The way I generally describe a color tint for a black and white image is that you are replacing the shades of gray in the image with shades of a specific color. For example, a sepia effect involves replacing the shades of gray in a black and white photo with shades of a brownish-yellow color.

For this relatively simple effect, there isn’t a significant difference in terms of the final result (all other things being equal) among the various ways you could add such a color tint. Lightroom and Photoshop, for example, provide several options for adding a color tint to a black and white image, including the Tint control available with the Black & White adjustment and the Colorize option available with the Hue/Saturation adjustment (both of these specific examples being found in Photoshop).

For a slightly more sophisticated effect, you can use a Split Toning adjustment, which can be found in Adobe Camera Raw and Lightroom, among other software tools. With split toning you are adding a color tint similar to what you might use to create a sepia effect, with the difference being that you are able to apply two different colors. So, for example, you could create an effect where shadow areas are a bit more blue and highlight areas are a bit more yellow, rather than having the entire image appear with the same color tint.

If you really want to exercise significant control over this type of color tint that involves more than a single color, the Gradient Map in Photoshop is the option I recommend. You can start, for example, with a simple black-to-white gradient for the Gradient Map adjustment layer. You can then add additional color stops to define the gradient.

A very simple gradient for the Gradient Map adjustment might include black and white at the extremes, with a single additional color stop for sepia, resulting in a basic sepia-tone effect. You could include two color stops between black and white to produce the effect of a Split Toning adjustment. But you can add even more color stops to produce a very sophisticated tint effect for what would otherwise be a black and white image, with the Gradient Map assigning overall color and tonal values based on the luminance values of each pixel within the photo.

So, again, in terms of the basic result, you can feel perfectly comfortable choosing among any of the available adjustment options in Photoshop, Lightroom, or other software tools that allow you to apply a color tint to a black and white (or even color) photo. However, the various tools available for applying the color tint in the first place provide the potential to exercise greater control over the image. It is also worth noting that while it is perhaps most common to add a color tint to an image that is being interpreted as a black and white image, you can also add a color tint  or split toning effect to a full-color photo, without removing the original color.

Today’s Question: I am upgrading from Lightroom 2 to version 4 on a new computer (I am a good bit behind, I know). I am trying to do some basic work on images that were uploaded in version 2 and put into version 4. I’m not clear on the ‘upgrade image’ button in the bottom right area. Should I automatically click on this?

Tim’s Quick Answer: Lightroom makes use of “Process Versions” to define the specific set of adjustments and algorithms available for an image, and you can change the Process Version for specific photos. In general it is advantageous to upgrade images processed with an older Process Version to the latest Process Version, though it is important to keep in mind that there may be some minor changes to the appearance of a photo when you make this change.

More Detail: When new adjustment features are added to a new version of Lightroom, a new Process Version is created to encapsulate those adjustments. In general it is advantageous to work in the most recent Process Version, so you have access to the latest features and algorithms for your photos.

The specific method of changing the Process Version varies in different versions of Lightroom, but you can find the Process popup in the Camera Calibration section at the bottom of the right panel in the Develop module. From this popup you can see the various Process Versions that are available, which currently include options for 2003, 2010, and 2012.

If you want to update the process version for multiple images, you can navigate to a specific location or apply a filter so that the images you want to update are shown on the Filmstrip. Then go to the Develop module, and click the alert symbol for the current image. This symbol will appear at the bottom-right of the photo in Lightroom 4, and below the Histogram display in Lightroom 5. When you click that symbol, you will be given the option of updating the current image to the latest Process Version, or updating all Filmstrip photos.

I recommend updating a single image to the latest Process Version as a test so you’ll see the degree to which the appearance of the photo changes based on the update. In most cases the changes will be minimal, and often represent a slight improvement in the image. But it is important to have a sense of this before committing to an update for multiple images.

But again, I generally feel it is advantageous to have the latest adjustment controls available, and also to have the latest algorithms for existing adjustments. Therefore, I usually prefer to update photos I’m working on to the latest Process Version.

Today’s Question: Should I make adjustments to an image before cropping, or crop first and then make adjustments?

Tim’s Quick Answer: In many respects it doesn’t really matter whether you crop before or after applying adjustments to an image. However, it is important to keep in mind that in either case you might cause yourself to make different decisions than if you had performed the steps in a different order. In other words, being thoughtful about the impact of your workflow order can be important in some cases.

More Detail: If you apply a crop early in your workflow, you might remove areas of the image that you might have otherwise kept if you applied your adjustments first. For example, there might be an area near the edge of the photo that you feel is distracting. In this case you might choose to crop the area out. If, however, you had applied adjustments first you may have found that the area actually adds to the image. Perhaps highlight details just needed to be toned down, for example. In this type of situation it would have been better to apply adjustments first.

On the other hand, applying adjustments before cropping the image might cause you to apply those adjustments in a way that is different than you would have after the crop. A good example would be adjustments for the black and white values in a photo. If the area of a photo you cropped out happens to include the brightest and/or darkest areas of the photo, that could have a significant impact on the settings you use for the black and white point adjustments. In this case it would have been better to save the cropping for after the adjustments.

Of course, because the challenge here can be a factor regardless of which order you apply adjustments versus a crop, it can be challenging to anticipate the best approach for a given image. This is one of the many great benefits of working with a non-destructive workflow. By applying adjustments and cropping with a non-destructive approach (in either Lightroom or Photoshop, for example) you preserve the ability to go back and forth among the various adjustments and the crop, so you can fine-tune each based on the impact of the other.

Today’s Question: With the recent talk about Adobe not updating older versions of Camera Raw, among other issues such as slow updates for new RAW formats, how bad an idea would it be to shoot in JPEG instead of RAW so you don’t have to worry about software updates?

Tim’s Quick Answer: Shooting in JPEG (rather than RAW) most certainly streamlines certain elements of your workflow, but it also increases the risk of quality problems in your photos. However, if you ensure excellent exposure and accurate white balance in the camera, the only major issue to be concerned with is the potential for visible artifacts in the images caused by JPEG compression.

More Detail: With JPEG capture, there are certainly risks related to reduced flexibility for optimizing your photos. For example, a JPEG capture will always be an 8-bit per channel image, whereas a RAW capture can be processed to a 16-bit per channel image. This translates into a greater risk that gradations of tone and color will not be as smooth in JPEG images if you need to apply strong adjustments to those photos.

Of course, if the JPEG capture looks perfect (or close to perfect) right out of the camera, then you don’t have to worry about strong adjustments causing problems with image quality. So, if you’re confident in your ability to achieve accurate exposure and white balance settings, you don’t have to worry too much about image quality problems being introduced by strong adjustments.

That said, even with absolutely perfect photographic technique, you can’t avoid the issue of compression artifacts with JPEG captures. Even at the highest quality setting, JPEG captures will have compression applied to them in order to produce a smaller file size. That compression always has at least a slight negative impact on image quality for JPEG captures.

In some cases the compression artifacts may be very difficult to see. But they will be there. Frankly, the risk of having JPEG artifacts in photos is a critical factor from my perspective. In other words, while I am perfectly comfortable in my ability to capture images that require little or no adjustment (most of the time), I’m not willing to risk having compression artifacts visible in my photos.

The “insurance” provided by RAW captures when it comes to applying strong adjustments to tone and color is certainly appealing. And the potential for greater dynamic range and other benefits to overall image quality is appealing. But the real reason I avoid JPEG capture is the presence of JPEG compression artifacts in those captures.

Today’s Question: Last week you talked about printer profiles for various papers. When a profile is not available for the paper and printer you’re using, is it possible to build one yourself?

Tim’s Quick Answer: It is possible to build your own profiles if you have a tool that allows for this, such as the ColorMunki Photo from X-Rite (http://timgrey.me/photocolormunki). There are also a variety of companies and individuals offering custom printer profiling services if you prefer not to build your own profiles.

More Detail: Building a printer profile is relatively straightforward. You start by printing a specific image that includes a series of color swatches. You then “scan” the color swatches on that print using a spectrophotometer, which is a special device for measuring color precisely. The resulting information is used to build a custom profile for the specific printer, ink, and paper combination you used to print the image with the color swatches. That profile can then be used to produce highly accurate prints with that specific print setup.

You can also find various companies and individuals who will build profiles for you at a fixed fee (generally around $30 to $100 per profile). Many years ago I actually used to offer just such a service. Later I recommended a couple of service providers. However, the providers I recommended in the past no longer offer this service, and I prefer not to make recommendations for specific providers unless I have personally tested their services. However, an online search for “Custom Printer Profiles” will provide many options you can choose from.

And again, if you want to build your own printer profiles you can use a tool such as the ColorMunki Photo, which you can learn more about here:

http://timgrey.me/photocolormunki

Today’s Question: Is there a general source for ICC printer/paper profiles where the manufacturers of the printer and paper are different? I have an Epson 3880 printer and bought some HP premium plus glossy photo paper.

Tim’s Quick Answer: As a general rule you can find profiles for most third-party photo papers, supporting a variety of different printer models, directly from the paper manufacturer. However, that generally excludes support for a combination of printer and paper from two different printer manufacturers.

More Detail: If, for example, you had purchased paper from a company that does not manufacture printers (such as Red River Paper, for example, available at http://www.redrivercatalog.com), you would generally find that profiles are available for most popular printer models.

However, in your case you have purchased paper from one printer manufacturer (HP) with the intent to print on that paper using a printer from another manufacturer (Epson). As you can imagine, HP would very much prefer that you only print on an HP printer, and so they have an incentive to provide ICC profiles for their papers only for use with their own printers.

In some cases you may get good results by using an ICC profile for a different paper that has similar properties to the paper you are actually printing with. However, more often than not you’ll find that this approach doesn’t provide the most accurate prints.

Therefore, if you want to use paper from one printer manufacturer to print with a printer from another manufacturer, you will generally want to skip the ICC profile altogether. Instead, you will need to choose the option to let the printer (rather than your software, such as Lightroom or Photoshop) manage the colors for the print.

You can then use the various settings in the printer properties dialog to make adjustments through a trial-and-error process to find settings that will produce an accurate print. Once you have those settings established, you can save them for use anytime you are using that specific printer and paper combination.