Month: January 2024

Today’s Question: With a subject that includes horizontal and vertical lines, such as a building, how important is it to get the final image to have lines that are perfectly horizontal and vertical, versus being a little askew due to perspective issues?

Tim’s Quick Answer: While accurate alignment in a photo can be important, I don’t think it is always necessary to completely correct perspective issues for all images. In some cases, having lines that are not aligned perfectly can actually add to the impact of a photo.

More Detail: One of my biggest pet peeves when it comes to evaluating a photo is seeing blemishes or other distractions in the photo. In some cases that distraction takes the form of improper alignment, such as when a horizon is crooked. A more extreme example of a lack of proper alignment is when straight lines within the image don’t appear straight, and perhaps aren’t perfectly horizontal or vertical.

However, in many cases a lack of proper alignment can actually be a good thing. For example, if you use a very wide fisheye lens that has perhaps an angle of view of around 180 degrees, straight lines within the scene are most certainly going to appear curved and distorted. But that distortion is a big part of what makes a photo captured with a fisheye lens so interesting and eye-catching.

I think the most important consideration when it comes to correcting for perspective and alignment issues in a photo is what looks right or most pleasing for an individual photo. I rarely correct an image so that all lines are perfectly horizontal or vertical, though in some cases I most certainly do.

To provide a very general guideline, if you have a photo where the key (or solitary) subject clearly stands out in the frame and looks like it should have perfect alignment, then it probably makes sense to ensure that all lines are perfectly horizontal or vertical.

For photos that have a relatively wide angle of view, and especially when the perspective effect makes the image more interesting, I might apply some slight corrections such as to ensure that the lines of a key subject in the center of the frame are perfectly vertical. But I most certainly won’t correct all images to remove all distortion or to ensure that all lines are perfectly horizontal or vertical.

Again, in my view the emphasis should be on what looks appropriate and pleasing for the individual image. If lines look like they should be perfectly horizontal or vertical, it is generally worth making sure they are aligned properly. But keep in mind that in many cases having lines that are somewhat askew or perhaps are slightly curved can very much enhance the look of an image.

Today’s Question: I’m a little confused. Many, if not most, professional photo labs request that you send them images in a JPEG format. If artifacts are an issue, why don’t they request images in a TIFF format?

Tim’s Quick Answer: In my view photo labs should not suggest using the JPEG file format for printing because of the risk of visible artifacts. They do so, I assume, because the JPEG format is widely supported and results in a small file size that is easy to transmit via the internet.

More Detail: In some ways you could say that photo labs request JPEG images because photo labs have always requested JPEG images, at least in the context of online submission of images. The JPEG format is very widely supported and yields smaller file sizes than other file formats. However, that smaller file size comes at a potential cost in terms of print quality. Therefore, for printing I suggest submitting images as a TIFF or PNG file.

To provide some context, I saved a 20-megapixel image as a TIFF file with ZIP (lossless) compression and the file size was 146MB. The same image saved as a PNG file with maximum (lossless) compression was 95MB in size. The same image saved as a JPEG image with maximum quality (but lossy compression) was only 12MB.

In my view the risk of visible artifacts is a very real concern whenever printing a photo, and especially when that print will be relatively large. I therefore strongly recommend not using JPEG files for photos that will be printed. With high-speed internet being relatively ubiquitous today, sending a file that is around 100MB rather than around 10MB is not a major inconvenience, and in my view well worth it when it comes to helping ensure optimal print quality.

If a printer only supports uploading images in the JPEG format, I suggest finding a different printer. For example, I have had very good results getting prints from Bay Photo (https://bayphoto.com), and they allow you to upload images in a wide variety of image formats, including TIFF.

Today’s Question: If pixel per inch (ppi) resolution doesn’t matter for the digital display of photos, wouldn’t saving at 300 ppi create a larger file than with 72 ppi?

Tim’s Quick Answer: No. The pixel per inch resolution setting for an image has absolutely no impact on the file size.

More Detail: The total number of pixels in the image combined with the file type and related settings determine the file size for an image. The ppi resolution is simply a metadata setting used for determining the output size of an image when it is printed.

As may have been made abundantly clear by the number of questions I’ve addressed recently related to resolution in general and ppi resolution in particular, clearly this is a topic that leads to confusion.

The pixel per inch resolution is simply a way to explain the overall pixel dimensions in an image in a way that is intended to be more clear. For example, I think most people would agree it is more meaningful to say that an image is 10-inches wide by 10-inches tall rather than that the image consists of 3,000 by 3,000 pixels. But in order to say that an image is 10-inches across, we would need to specify the pixel density, which is what the ppi resolution provides. So instead of saying the image is 3,000 by 3,000 pixels, we can say the image is 10-inches by 10-inches at 300 pixels per inch. Both are saying the exact same thing in terms of pixel dimensions, just using different terminology.

Changing the pixel per inch resolution for an image will not affect the file size, all other things being equal. For example, let’s assume an image that is 1,000 by 1,000 pixels, in the 8-bit per channel bit depth, saved as a TIFF file without compression. The file size will be about 3MB. If you change the ppi resolution to 300 ppi, the file size is still 3MB because it still contains the same number of pixels. If you change the ppi resolution to 1,000,000 ppi the fie size is still 3MB because it still contains the same number of pixels.

Just remember that the ppi resolution simply provides a way for you to translate a set number of pixels to a set number of inches on the printed page. The ppi number is just a metadata value, intended to provide helpful translation information but unfortunately leading to confusion for many photographers.