colour saturation and Raw v JPEG issues

the entire raw vs jpg concept is being missed here.

If I had the time I would write about it... someone can provide a synoptis or a link ?

Jpeg is an 8 bit, 256 colour, lossy, compressed file format. In a compressed file format data is lost from the original and when it comes to photos data means picture quality. Since reality has a lot more than 256 colours, it means for example that 20 shades between red and black with gradual transitions in a sunset are compressed to perhaps 8 with harsher transitions and banding. Editing 8 bit images results in losing more colours as shown in the histogram, so you could end up down to 200 colours or less. Artifacts such as halos and stray pixels also end up added by the editing process further degrading the image.

RAW is straight uncompressed data from the chip. Nothing is lost. It can also be edited in a raw program without any degradation as well. It can be saved as a 16 bit TIFF file with thousands of colours which can be edited withough noticeable colour loss. Some plug-ins such as software filters for Photoshop or PaintShop Pro open up more sophisticated opportunities for editing in the higher quality 16 bit format.

skieur
 
Clearly the OP was not a JPEG v RAW question - shoot RAW by default and I'm not keen on JPEGs

Because of the course I was on and to speed up viewing between shoots I was requested to shoot JPEG. In case I wanted to do some processing I shot the two together and because of this had a chance to view the images together.

Now I have to start to question my work flow and processing style etc. This has moved on as my experience has, and is now spilling into further questions regarding WB and colour space that both effect the image and setting it up in RAW etc
 
Jpeg is an 8 bit, 256 colour, lossy, compressed file format.

Not true. Each of three channels (RGB) have 256 colours (or shades) each so you can achieve over 16 million colours (256x256x256).

Since reality has a lot more than 256 colours, it means for example that 20 shades between red and black with gradual transitions in a sunset are compressed to perhaps 8 with harsher transitions and banding. Editing 8 bit images results in losing more colours as shown in the histogram, so you could end up down to 200 colours or less. Artifacts such as halos and stray pixels also end up added by the editing process further degrading the image.

Not sure where you got this info but it's completely wrong. Sure compression deletes info from the image but your figures are way off the mark.

RAW is straight uncompressed data from the chip. Nothing is lost.

RAW is still a COMPRESSED file but is a lossless compression (similar to tif in that respect).

It can be saved as a 16 bit TIFF file with thousands of colours......

It can indeed be saved as 8bit or 16bit tiff. An 8 bit tiff has the same number of colours as an 8 bit jpg (16 million) however a 16 bit tiff has many many more. 65536 per channel as compared to 256 in an 8 bit file. So a theoretical 65536^3 = 281,474,976,710,656 (281 trillion)!!

Although that doesn't matter too much as all printers print in 8 bit I believe so that 16 bit file is converted to 8 bit either by you or by the printers driver.

.....which can be edited withough noticeable colour loss.

There is no "noticeable colour loss" because there is none when edited, because the tif files are also a lossless compression. If you edit the image you control the appearance and loss/addition of colour info.

Can you advise where you got your info as it seems flawed in every way.

JD
 

Most reactions

New Topics

Back
Top