Ask Tim Grey

Today’s Question: You said that you retain all layers for images you work in in Photoshop in part to maximize image quality. How does using layers improve image quality?

Tim’s Quick Answer: Using layers in Photoshop primarily provides for a more flexible workflow. However, using adjustment layers rather than adjusting pixel layers directly can also benefit image quality by virtue of not compounding the effect of multiple adjustments.

More Detail: When you apply an adjustment to an image you are obviously altering the pixel values. Each time you apply an additional adjustment you’re compounding the effect, and this can have a negative impact on image detail and overall quality.

For example, increasing contrast can cause a loss of some detail in the dark shadows and bright highlights. That loss is generally very modest, and the improved appearance from the enhanced contrast is likely worth the very slight loss of detail.

However, if you enhance contrast in multiple steps rather than a single step, there can be a more significant loss of detail than if you achieved the same result with a single adjustment. For example, if you gradually increase contrast in ten steps that can cause more degradation than if you achieved the same degree of increased contrast in one step.

When using an adjustment layer in Photoshop, that adjustment layer provides the same effect as if you had only applied one adjustment, no matter how many times you fine-tune the settings for the adjustment layer. Therefore, when using an adjustment layer there is no compounding effect cause by altering the same adjustment multiple times.

To be sure, with normal adjustments for a photo the benefit in terms of image quality when it comes to using adjustment layers is very modest. However, I strongly recommend using layers when optimizing photos to take advantage of the greater flexibility and the potential benefit in terms of image quality, minor though it may be.

Today’s Question: As a follow-up question, do you flatten the image before saving it as a TIFF, or maintain the layers in case you need to make any adjustments as you refine the printing processing?

Tim’s Quick Answer: While I would save a flattened copy of the source image when sending a file for printing, I always retain the “original” version of the image with all layers intact so I can refine the image later if needed.

More Detail: When I process an image in Photoshop, I often make extensive use of layers, including adjustment layers, additional image layers for cleaning blemishes, layer masks with adjustment layers for targeted adjustments, and more. My preference is to preserve all layers for that image so I can always go back and make changes later if needed.

My original capture, of course, is generally a raw capture. I typically adjust that image in Lightroom Classic, which in many cases might be all the editing I need for a photo. However, some images require a little more work, which means sending the image to Photoshop.

I prefer a layer-based workflow in Photoshop to help maximize image quality and give me more flexibility in my workflow. I save this image as either a Photoshop PSD file or a TIFF image, with all layers intact. I can then open the image in Photoshop at a later time so I can refine those layers as needed.

When sending an image to someone else, such as for printing, they generally don’t need to have access to the layers, and I generally don’t want them to have access to the layers I used to create the final version of the image. Therefore, when preparing an image such as for printing, I will save the result as a new file with all layers flattened.

Today’s Question: In the [August 3rd] newsletter you discuss converting a TIFF file to a JPEG file and the resulting file size. At the end you say, “JPEG images should only be used for derivative images that are shared digitally, not as archival images or as the basis of photo prints.” What file format do you recommend for photo printing, especially if sending the file to a third-party printer?

Tim’s Quick Answer: I recommend using the TIFF file format for saving photos that have been prepared for printing, as long as that is supported by the print service you’ll use.

More Detail: Most photographers are (I hope!) capturing their photos in a raw capture format. That is great for maximizing image quality and for optimizing the photo, but you wouldn’t want to send a raw capture format to a print service. Whenever you will share a photo with someone else for any purpose, you would want to create a derivative copy of the original capture.

As mentioned in my answer from August 3rd, for sharing online or other digital sharing the JPEG format is a good option because it provides a smaller file size. There is some degradation in image quality caused by the JPEG compression however, so I don’t recommend using the JPEG file format for images that will be printed.

In my view the best file format to use for printing is the TIFF format, either without compression or with lossless compression such as LZW or ZIP. This will ensure optimal image quality for the derivative created from your source capture, which in turn will help ensure the best print quality.

Of course, when sending a file to a third-party printer it is also important to meet the requirements of that provider. With some online printing services, for example, you may be limited to only being able to upload JPEG and possibly PNG images. This isn’t ideal, but in general you can still achieve good print quality even with a JPEG file as long as a high setting for quality was used.

If possible, however, I recommend saving photos intended for printing with a file format that will not involve any lossy compression. The TIFF file format meets this requirement, while also being widely supported by most printing services.

Today’s Question: Should you do noise reduction before you start editing an image or after?

Tim’s Quick Answer: I recommend applying noise reduction early in your workflow for optimizing a photo, though depending on which software you’re using that timing may not be as important.

More Detail: Various adjustments can amplify the appearance of noise in an image or make the noise more problematic. For example, sharpening involves an enhancement of edge contrast, and this can cause the noise to be exaggerated to some extent. Because of these issues I recommend applying noise reduction as early in your workflow as possible.

However, the software you’re using may make this timing less critical. For example, if you’re using Lightroom Classic to optimize your photos, the normal noise reduction that can be applied manually can be used in any order. It doesn’t matter whether you apply this noise reduction before sharpening or after, for example, because the adjustments aren’t applied directly to the image. Rather, the adjustments effective represent metadata values, and they aren’t applied until you create output from the source image, such as when exporting a derivative copy.

In situations where the noise reduction is being applied to the image before other adjustments, I do recommend applying noise reduction as one of the first adjustments to an image.

For example, if you’re going to use the AI Denoise feature in Lightroom Classic (or Camera Raw) that should be the first step you apply. After that noise reduction is applied to the image you can continue with any other adjustments in any order you prefer.

Whenever noise reduction is actually altering pixel values in an image, I recommend applying that noise reduction first (or as early as possible). In situations where the adjustments aren’t altering the pixels directly, such as with most adjustments in the Develop module in Lightroom Classic, the various adjustments can be applied in any order you’d like.

Today’s Question: I have negative scan which is a fairly reasonable size TIF file (6 MB). However, when I export as a JPEG, the file size is tiny (300 KB). What am I doing wrong? I tried 300 pixels per inch and 2400 pixels per inch, but these resolutions had no effect on the JPEG file size.

Tim’s Quick Answer: You’re not doing anything wrong. The smaller file size is simply a matter of the compression being applied to the JPEG copy of the image.

More Detail: In most applications, saving an image as a JPEG file automatically means that lossy compression is applied. This means that the file size will be reduced by simplifying the description of the pixel contents of the image, resulting in some degree of loss of fidelity and image quality. In other words, if you save the exact same image with the same pixel dimensions as both a TIFF and JPEG image, the JPEG will always have a smaller file size. In most cases the difference in file size will be quite significant.

As a general rule, a TIFF file with a bit depth of 8-bits per channel will have a file size that is about three times the number of megapixels in the image. So, for example, a 20-megapixel image saved as an 8-bit per channel TIFF image would result in a file size of about 60 megabytes. If that same image is saved as a 16-bit per channel image the file size would double to about 120MB.

A JPEG file saved from the same 20-megapixel image would have a file size in the general vicinity of 10MB, though the actual results can vary significantly.

There are three key factors that affect the file size for a JPEG image. The overall pixel dimensions are obviously a significant factor. The JPEG quality setting is also significant, as it determines the strength of the compression being applied. A higher quality setting obviously helps maintain better image quality, but results in a larger file size. A lower quality setting will result in a smaller file size, but image quality can be degraded somewhat significantly in the process.

The third factor is the relative complexity of the image. The simpler the image is, the more it can be compressed without degrading image quality. For example, a blank white image saved as a JPEG will have a significantly smaller file size than that of a normal photographic image. If the 10MB JPEG image referenced above were filled with white and saved as a new JPEG, that file size would be about 1MB.

So, JPEG files will always be smaller than most other file formats because of the compression applied to those JPEG images. Of course, quality can be degraded as part of the process of compressing the data, which is why JPEG images should only be used for derivative images that are shared digitally, not as archival images or as the basis of photo prints.

Note, by the way, that the pixel per inch (PPI) resolution won’t have any impact on the file size. That is just a metadata value that affects the output size when printing but doesn’t affect the file size because it doesn’t affect the overall pixel dimensions of the image.

Today’s Question: As a follow-up to the “Defining Your Workflow” presentations, I wonder about the culling software that are out there. I would be interested in your opinion as to whether or not you think they are worthwhile and worth the cost. If so, is there one you might prefer?

Tim’s Quick Answer: While there are some good and interesting tools for culling software automatically, I find that these are most useful for portrait photographers and can create a more cumbersome workflow for other types of photographers, especially in the context of a workflow that revolves around Lightroom Classic. One good option you might look at is Aftershoot (https://aftershoot.com).

More Detail: The idea of automatic culling software for photos is pretty straightforward. The software analyzes your images, and helps to identify those you might be most interested in reviewing and those that may be outtakes. This can be helpful when it works well, but in my view the software is really only helpful if you can trust the results without having to review all the images to confirm those results are accurate.

My experience has been that while software such as Aftershoot can certainly mark photos based on attributes such as being out of focus or featuring a person with their eyes closed, the result does not necessarily replace a manual visual evaluation of your photos.

For portrait photographers who capture a large number of photos of people, I would say this type of software is definitely worth evaluating, as it can most certainly help you more quickly focus on the photos that are most likely to represent your favorites. For other types of photographers, such as landscape and travel photographers, I would say the software provides less meaningful utility.

My main issue with using culling software beyond Lightroom Classic is that it can create a more cumbersome workflow that doesn’t necessarily save considerable time when compared to a manual review within Lightroom Classic. That’s especially true considering that even after using culling software you’re almost certainly going to want to review the photos yourself.

In other words, the culling software can help you focus on the photos that are most likely to become favorites, but a manual review is still necessary in my view. I would use the analogy of spam filtering for email. In many cases the spam filter does a good job of identifying the most obvious spam, but many non-spam messages still end up in the spam folder, and some spam messages still slip through.

Having said all that, if you think culling software might aid your workflow, I recommend taking a look at Aftershoot. You can find more information and download a free trial here:

https://aftershoot.com

Today’s Question: Is there a way to copy white balance from one image to another in Lightroom Classic? I am shooting a lot of infrared, and the custom white balance is usually under 2000 Kelvin, which is the low limit for Lightroom Classic. Occasionally I forget to white balance in camera. Is there a way to copy the custom white balance from an earlier grey card for that frequency of infrared to another photo in Lightroom Classic? Could I create a custom profile to do this?

Tim’s Quick Answer: You can copy the white balance settings from one image to another in Lightroom Classic. In situations where you need a stronger adjustment for color, you can use a custom Tone Curve adjustment for individual color channels. These adjustments could be saved as a preset if they represent common settings needed for multiple images.

More Detail: In Lightroom Classic you can copy settings from one image to another in a variety of ways. One quick option is to copy the settings from one image and paste them to another. For example, you can select the image that has a good white balance adjustment applied and click the Copy button at the bottom of the left panel in the Develop module.

In the dialog that appears you can enable only the adjustments you want to apply to the other image. For example, in this case you may want to click the “Check None” button so that none of the adjustments are enabled. Then turn on the checkbox for White Balance, along with any other adjustments you want to copy from the source image. Click the Copy button at the bottom-right of the dialog to copy the selected settings.

Next, select the image you want to apply the adjustments to and click the Paste button at the bottom of the left panel. This will apply the adjustments you copied from the other image to the current image, which in this context will apply the same color temperature compensation.

If you’re finding that the adjustments aren’t strong enough for extreme situations, such as with infrared capture or scenes illuminated by very strongly colored light, you can supplement the White Balance adjustment with a Tone Curve adjustment focused on specific channels.

For example, in the Tone Curve adjustment section on the right panel in the Develop module you can select the blue channel from the Adjust control above the curve. Then click at the center of the curve for the blue channel and drag downward if you want to shift toward a more yellow (warm) appearance or drag upward if you want to shift toward a more blue (cool) appearance.

If you’ve applied these adjustments to an image and want to create a preset to use for other images captured under similar conditions, you can click the plus button to the right of the Presets heading on the left panel and choose “Create Preset” from the popup menu. In the dialog that appears enter a meaningful name for the preset in the Preset Name field and turn on the checkboxes for the adjustments you want to copy. Click the Create button at the bottom-right of the dialog and that preset can then be applied to other images that require the same adjustment settings.

Today’s Question: I’ve used Sky Replacement in Photoshop with some success, but in some images the clouds in the original sky “bleed through” the new sky and faintly appear. Is there an adjustment I am missing?

Tim’s Quick Answer: This type of issue is not uncommon with the Sky Replacement feature in Photoshop. When this happens, I recommend refining the layer masks associated with the new layers created by the Sky Replacement command.

More Detail: The Sky Replacement command attempts to ensure that the new sky blends into the surrounding areas of the image to create a more natural effect. However, this often leads to too much blending, allowing some of the original sky area to remain partially visible. In some cases portions of the sky might not be detected at all, leading to an area where only the original sky is visible.

When using the Sky Replacement command I recommend making sure that the “New Layers” option is selected from the “Output To” popup. This will cause the replacement sky and additional effects layers to be added to a layer group named “Sky Replacement Group” to be added above the image layer you’re replacing the sky for.

Within that layer group you’ll find several layers, with layer masks associated with some of those. In particular you’ll want to evaluate the layer mask for the Sky image layer. You can then paint with white on that layer mask to reveal additional areas of that replacement sky layer, or paint with black to block the sky to reveal portions of the underlying image.

It can also be helpful to perform a little bit of dodging and burning on the mask to reduce the blending along edges between the replacement sky and the underlying image. To do so you can set the blend mode for the Brush tool on the Options bar to Overlay. Then paint with white over the areas of the layer mask where you want to reveal the new sky without blending, and paint with black over areas of the layer mask where you want to reveal the underlying image without blending.

With the Overlay blend mode you can paint with white without affecting any black areas of the layer mask, and you can paint with black without affecting white areas of the layer mask. This provides a helpful way to shift shades of gray on the layer mask to black or white to completely block or reveal pixels, respectively.

Today’s Question: I am interested in doing an incremental backup of my photo folders to two to three external drives in Windows 11. By incremental I mean only the changes that were made since the last backup. Am I correct in calling this type of backup incremental? Can this type of backup be accomplished in Windows 11 and if so, how?

Tim’s Quick Answer: Yes, what you’re referring to is an incremental backup, and what I recommend is to create a synchronized backup using software such as GoodSync (https://timgrey.me/goodsync), which is available for both Windows and Macintosh.

More Detail: In my early days with computers an incremental backup could be a rather involved operation, and recovering from such a backup was an even bigger undertaking. To restore from an incremental backup the software would often need to rebuild from the backup based on the original backup plus all incremental updates. Back when it was common to back up a hard drive to a series of floppy disks, this could be a time-consuming and frustrating experience.

Fortunately, there are software tools available today that make the process much easier, especially when it comes to restoring from a backup. I recommend a synchronized approach, and I use and recommend GoodSync software (https://timgrey.me/goodsync) for this purpose.

With GoodSync the first backup can require considerable time since the entire contents of the source drive need to be copied to the backup drive. From that point forward, however, GoodSync will analyze the source and backup drives and update the backup with only what has changed since the last backup.

The result of using GoodSync is a backup that is an exact match of the original. For example, if the external hard drive that contains all my photos were to fail, I could simply connect the backup drive that had been updated by GoodSync and use it in place of the failed original. The only additional step at that point would be to change the drive letter of the backup drive to match the original on Windows, or to change the volume label to match the original on Macintosh.

I demonstrated the use of GoodSync for backing up photos in my GreyLearning Ultimate Event on “Backing Up Your Photos”. This event is included at no additional cost in the GreyLearning Ultimate Bundle, and you can learn more about the course here:

https://www.greylearning.com/courses/greylearning-ultimate-event-backing-up-your-photos-may-2023

Today’s Question: In Lightroom Classic is there a way to remove a keyword that was inadvertently added to all photos in an entire folder? Or does each photo need to have that keyword removed individually?

Tim’s Quick Answer: Yes, you can easily remove a keyword from multiple images in Lightroom Classic by selecting those photos in the grid view and then turning of the checkbox for the applicable keyword on the Keyword List.

More Detail: If you’ve accidentally added a keyword to a large group of photos, such as by adding a wrong keyword in the Import dialog when importing new photos, it is easy to remove that keyword from the images.

Start by switching to the grid view (not the loupe view) and selecting the photos you want to remove the keyword from. In this example that would mean selecting all photos in the applicable folder, such as by choosing Edit > Select All from the menu.

With the photos selected in the grid view, go to the Keyword List on the right panel and locate the keyword you want to remove. Turn off the checkbox to the left of that keyword, and the keyword will be removed from all selected images.

Note that there are three possible status options for the checkbox to the left of each keyword on the Keyword List. If there is a checkmark that indicates that all selected images have that keyword assigned to them. If the checkbox is empty that means the keyword has not been assigned to any of the selected images. If there is a dash in the box that is an indication that the keyword is assigned to some but not all the selected images.

What that means is that if you see a dash, you’ll need to click more than once to remove the keyword. The first time you click on the checkbox the dash will change to a checkmark, meaning the keyword will be assigned to all selected images. You can then click on the checkbox one more time to remove the keyword from all selected images.