Losing image size / resolution when exporting!

First: DPI is only an indication how a picture should be printed at 100%. Infact you could add this parameter from 1 to very high, without changing one single pixel of your image. As a rule of thumb this parameter is set to something between 200 to 300, without having any influance on the quality.

This is different for an other parameter pair: resolution in pixels. You sensor has a physical resolution and that one will determine the size of your picture. There is generally no good reason to increase that resolution to get better image quality.

The sensor is normally colour blind (favon sensor from Sigma is an exception to this). Colour is captured by adding filters to the sensor pixels. Normally the are arranged in what is called a Bayer pattern. Developping RAW pictures means interpolating these Bayer pattern values into real colour information. Ech pixel is availablein 12 to 14 (most modern sensors have 14 bits) bits, compressed in a nondestructive way. So this means that in a worst case scenario and with 14 bits a Raw picture has a storage need of Xres*Yres*14bits/8bits+some bytes for storing parameters needed.

A Jpeg file has 3x8bitsxResolution, so this file should be bigger in size, if there would not be mathematical transformations, allowing to rearrange the pixels in such a manner that it is easy to compress. The aim here is to transform in a way that is effective but not or barly noticable for the human eye. This process is destructive, meaning that a compressed image cannot be reconstructed to be exactly equal to the original. The quality factor tells here, how much of the data gets lost. Highly compressed images show noticable artifacts. At a quality factor of 100 (12 in Photoshop), you do not notice any loss of quality, but even then, your picture may be highly compressed. The compression rate depends on the image. An image having only one colour would be best compressed. In the same sens, images with a lot of changes are dificult to compress.

A property of Jpeg is that when you open a Jpeg file and you resave as Jpeg without modification, the introduced artefacts will increase. So resaving will make your image worse.

I suppose that your customer needs 16x24 inch pictures, which is quite big. You need to know at what dpi this size is needed. Inches and dpi will translate into pixels. There is the use of this obscure dpi factor. Or for a given pixel size and a given print density you have a given size you can print.

A 24 x 16 inch picture can be printed in good quality at 200 dpi meaning that your picture should have a pixel size of 4800x3200. If you camara produces smaller images, you can enlarge a little bit. If it is bigger, export with more dpi, ie keeping the picture at its native pixel size.

Btw, print quality could be still good enough at 150dpi or even 100dpi. Generally I prefer resizing in Photoshop, because I have more influence on the quality. Photoshop is quite good in interpolating pixels in an intelligent way.

If we learn your camera, I could advise on how to proceed best.

Remark: Bayer pattern and Jpeg compression and also dpi can be looked up on Wikipedia in detail. There are also other formats of interest explained: Tiff, raw, PSD, ...

JPG is a compression technology. It compresses your image with extremely minimal loss of quality. So there is nothing wrong with a result of 1.2mb. It does not mean a loss of quality (other than the extremely minimal amount caused by compression); it does not mean a loss of resolution.

In fact, looking at megabytes is the wrong thing to do. It misleads you and causes you to take unnecessary and even harmful actions.

What you need to look at is the exported image with your own eyeballs; how does the quality look? I'm sure it looks fine (because it is fine). Also, look at the number of pixels in the image, are these the number of pixels you expected when you did the export? I'm sure it is (because Lightroom gets that right too).

Moved from Creative Cloud forum.