LWN.net Logo

Advertisement

E-Commerce & credit card processing - the Open Source way!

Advertise here

Dynamic range and file formats

Dynamic range and file formats

Posted Mar 30, 2007 12:29 UTC (Fri) by StuHerbert (subscriber, #15663)
Parent article: Working with raw images on Linux

Just a quick correction ... the dynamic range of an image is limited by the capabilities of the sensor, first and foremost. It isn't that RAW images have a *higher* dynamic range - the top and bottom of the range isn't different between RAW and JPEG. It's just that, having more bits per channel, RAW images have more data between the top and bottom of the dynamic range than an 8-bit JPEG image can hold.

You need a better sensor to achieve a higher (or, strictly speaking, a *wider*) dynamic range. One such sensor is the specialist sensor used in the Fuji S5 Pro.

Best regards,
Stu


(Log in to post comments)

Dynamic range and file formats

Posted Mar 30, 2007 13:06 UTC (Fri) by corbet (editor, #1) [Link]

I think we disagree slightly about what "dynamic range" means. It's not the difference between the upper and lower ends, instead, it's the ratio between the highest value and the smallest possible change you can represent. If you take 12-bit output from your sensor and cram it into an 8-bit file format (compressed or not) you've lost dynamic range.

In other words, the "more data between the top and bottom of the dynamic range" you mention is the dynamic range. It's what lets you record a bit more information about what's in the shadows without overexposing the image as a whole.

Copyright © 2008, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds