DPI A simple approach

DPI is an acronym for "Dots per inch" (or pixels per inch)

One inch is equal to 25.1 mm.

Some of the used expressions are often mixed up: Pixels, DPI, resolution, imagesize, filesize and so on ...

Let's try to make things simple:

The imagesize is a fixed size ... let's say an image has 1000x1000 pixels. This means the image is 1000 pixels in width and 1000 pixels in height.

This would be 1,000,000 pixels or 1 megapixel ... which would actually be the image's resolution.

One pixel matches one dot in the image (except it's size). The difference is that we talk of pixels if the image is displayed on a computer's monitor and we name them dots if we print.

So DPI means nothing else than the number of dots we print on 25.4 mm or one inch.

300 DPI would tell us now: We're talking about prints (because of the dots) and there are 300 of these little dots arranged on 25.4 mm / one inch. Since the dots have always the same size, they are pretty close to each other and there are rarely gaps between them. This 300 DPI is a pretty good size for prints and it will deliver very good results on up to date printing units and home printers.

72 DPI is the standard resolution for monitor displays. It means that there are only 72 dots on one inch / 25.4 mm. Well, you see: If now 72 dots have to cover the same distance (one inch) as before 300 of those dots did, the "pixels" need to be a bit bigger than the "dots" because we don't want to see gaps between them, right? But - the 72 pixels are the best resolution for your monitor, even if the image is a little more "grainy" ...

Now, has your image 72 or 300 DPI?

Both is true ... ;-) ... Let's come back to the initial size for your photo: 1000x1000 pixels

If we now try to display it with 300 DPI we have to ensure that there are 300 pixels on every inch, which would mean that your photo has the following width and height:

1000 pixels / 300 DPI = 3.33 so your photo is 3.33 x 25.4 mm in width and length ... :-) (equals about 84.7 mm)

Now let's see which size you get in a resolution of 72 DPI:

1000 pixels / 72 DPI = 13.89 ... this means 13.89 x 25.4 mm = 352.81 mm ...

Let me show you some of these things in the referring Photoshop displays:

Resolution and imagesize - get to the display by clicking Image - Image size... in Photoshop.

Filesize - get to the display by clicking file - save as then select a name and the file format JPG. Click ok and you'll see the screen.

I want to print my image ... how can i esthimate the width and height the file will have?

Well, the easiest way to do that is by using the displays shown above.

Select the imgage size display, make sure that the box close to "Resample Image" is UNCHECKED and select the resolution for you photo ... For prints select 300, for on-screen display 72 ... you will see how the width and height change when entering the values.

And finally a nice trick ... resample your image!

Now you found out that your photo (released from the examples above) has a size of 10x15 cm at 300 DPI ...

"To bad" you think, because it is one of your favourite shots and you want to have it printed on 20x30 cm ...

If we resize the photo as shown above, we will receive a resolution of 150 DPI (because we stretch the dots to the double distance to each other) ... but we heard that this resolution might cause poor print results.

Here comes the trick: Resample your image

Go to the image size display in Photoshop. It looks like that:

Image - Image size ...

Now do the following steps:

Enter the new width or height (i used the height in my screenshot), but make sure that the selections for "resample image" and "Constrain propotions" are checked before you do so!

IT'S A MIRACLE ... you might think now ... Photoshop has done it - your photo has now exactly the sizes, you wanted it to have.

What happened?

Simple ... Photoshop is an editing software and one of it's capacities is to recalculate the resolution. A real professional would probably protest, but I think to make things easy:

The software calculates that the original dots have to be spread to double size ... That means it needs to double the dots by inserting new dots in the gaps which were opened by the enlargement. Now the software has a close look to the surrounding dots and esthimates the color the new dot has to get ...

Actually the software works pretty good in doing so, nevertheless it's not perfect and some resharpening or other slight works have to happen on the photo ...

This was a simple example and it works in certain borders ... you can imagine that things get more tricky, if the gaps are huge and the software has to do a lot of this esthimating ... the quality of the photo will suffer ...

Well, finally i hope you got an basic idea out of this ...