A story about JPG image compression

Following are 11 images that, on first glance, might seem to look all the same. In fact, even after second and third looks they are still likely to appear to be all exactly the same.

But they aren't!

Although they are all precisely the same screen size, which is exactly 1200 pixels high by 842 pixels wide, each has been saved with a different JPG compression setting. The tool I have used to save these pictures is Adobe Photoshop CC (v2015.1.2). You need to know this because different tools apply different degrees of compression aggression when saving JPG files.

Going from left to right, top to bottom, each picture of the teddy bear and the YSL Tribute high heels has been saved using increasing degrees of compression in line with each of the 11 compression settings provided in Photoshop.

The first image has been saved using the compression slider set fully to the right at C12. Now some folks might think that setting the slider hard to the right means that no JPG compression is going to be applied. Well; you would be wrong.

While this does cause Photoshop to apply the least compression—'least' is far from zero. A little searching around on the Web will reveal that setting the slider to the hard right is typically going to result in about 50 percent compression.

In the case of my teddy bear and heels picture the compression that Photoshop applied was 57 percent, which resulted in the image being 896KB when saved. The original non-compressed image size is 2,090KB.

Moving the slider one notch to the left, shown as the 11 position in Photoshop, has a dramatic impact on compression. In the case of this test picture it went from 57 percent compression to a whopping 70 percent compression. However, I defy anyone to spot any difference in the picture when viewed full size on a good screen—click any image to see it full size.

With each subsequent picture I have increased the Photoshop compression by one click.

The last picture, being the eleventh picture in the series, was saved at compression setting C2. This resulted in a file of only 94KB which represents a massive 96 percent compression.

The following picture shows a clip from the first save side-by-side with the same clip from the last save.

If you look at this full size on a large screen you will now see some of the 'damage' that has been introduced by the 96 percent compression applied in the last save. If you look at the heel on the left shoe you will notice it is nicely defined and glossy in the C12 picture but there are colour artefacts causing blotchiness in the C2 picture. Ditto for the ankle strap.

You can also see artefact blotchiness in the grey patch of the pillow that the right-side shoe is resting against. The definition of the cotton thread in this patch has basically been wiped out by compression.

The more you look the more you will notice the degredation introducted by the compression.

However, the JPG compression algorithm is doing a pretty awesome job when you consider the insane amount of compression that has been applied to this last picture to get it down from 2,090KB to just 94KB while maintaining the original size of 842x by 1200y.

In my next post I will cover some of the upsides and downsides of JPG compression and compression in general.

Previous
Previous

Introducing my new 'apex predator' wildlife lens

Next
Next

USB-C: Not as simple as it seems (when it come to video)