<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.0.12-alpha" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: Why I Shoot Raw: Recovering From Disasters</title>
	<link>http://regex.info/blog/2007-10-08/598</link>
	<description>Not a photo blog, but sometimes I play one on TV</description>
	<pubDate>Tue, 02 Dec 2008 20:48:20 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.0.12-alpha</generator>

	<item>
		<title>by: acl</title>
		<link>http://regex.info/blog/2007-10-08/598#comment-18650</link>
		<pubDate>Mon, 07 Jul 2008 00:05:41 +0000</pubDate>
		<guid>http://regex.info/blog/2007-10-08/598#comment-18650</guid>
					<description>Jeffrey,
I have no idea if you've resolved this, but it certainly looks like the difference you get when you apply auto tone on the camera's jpeg and the jpeg from lightroom is due to a different curve used. That is, if the curve used by lightroom pushes the "shadows" (which end up being midtones after autotone, assuming autotone is sort of like autolevels or something like that) lower than the camera does, this should happen. No?

Anyway, I enjoy your blog when I read it (much more than I enjoy fighting with regexps, I could add).</description>
		<content:encoded><![CDATA[<p>Jeffrey,<br />
I have no idea if you&#8217;ve resolved this, but it certainly looks like the difference you get when you apply auto tone on the camera&#8217;s jpeg and the jpeg from lightroom is due to a different curve used. That is, if the curve used by lightroom pushes the &#8220;shadows&#8221; (which end up being midtones after autotone, assuming autotone is sort of like autolevels or something like that) lower than the camera does, this should happen. No?</p>
<p>Anyway, I enjoy your blog when I read it (much more than I enjoy fighting with regexps, I could add).
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: Bob Johnston</title>
		<link>http://regex.info/blog/2007-10-08/598#comment-15904</link>
		<pubDate>Wed, 02 Apr 2008 17:25:41 +0000</pubDate>
		<guid>http://regex.info/blog/2007-10-08/598#comment-15904</guid>
					<description>Jeffrey, believe the difference between the JPG made from the RAW and the in camera JPG, is normal. 

 When you make a JPG from a RAW, just exporting it, much of the detail contained is lost in the processing and compression.  In the in camera JPG, no processing is done, it is exactly as recorded....

&lt;span class='jfriedl'&gt;I'm not exactly sure what you mean here, but it sounds wrong. The in-camera JPG is not "exactly as recorded" (that's the raw data), it's the processed and compressed result of the camera's sensor-to-jpg engine, and it shares the same data limitations as any other 8-bit jpg (limits that are inherent in the file format).&lt;/span&gt;

NOW, if you take that in camera JPG, and convert it to a TIFF file, you can modify it even more, and many  times get a better image. Modifying contrast, brightness, color, etc.  Then, convert it back to a JPG for display.  

 Many "bad" JPGs made in camera by relatives, which had no modification, have salvaged pictures many would have discarded.  Even large group flash exposures where the foreground were over exposed a bit, and background faces almost unrecognizable.  In Photoshop circled the face with a "magnetic lasoo tool" then adjusted them individually, and brought out all the detail.

Sometimes using Dodging and Burn in Brushes to modify small areas, etc..

Bob

&lt;span class='jfriedl'&gt;Indeed, many seemingly-poor jpgs can be salvaged by good editing; I intended to imply only that the range is wider when working from raw data. &#8212;Jeffrey&lt;/span&gt;</description>
		<content:encoded><![CDATA[<p>Jeffrey, believe the difference between the JPG made from the RAW and the in camera JPG, is normal. </p>
<p> When you make a JPG from a RAW, just exporting it, much of the detail contained is lost in the processing and compression.  In the in camera JPG, no processing is done, it is exactly as recorded&#8230;.</p>
<p><span class='jfriedl'>I&#8217;m not exactly sure what you mean here, but it sounds wrong. The in-camera JPG is not &#8220;exactly as recorded&#8221; (that&#8217;s the raw data), it&#8217;s the processed and compressed result of the camera&#8217;s sensor-to-jpg engine, and it shares the same data limitations as any other 8-bit jpg (limits that are inherent in the file format).</span></p>
<p>NOW, if you take that in camera JPG, and convert it to a TIFF file, you can modify it even more, and many  times get a better image. Modifying contrast, brightness, color, etc.  Then, convert it back to a JPG for display.  </p>
<p> Many &#8220;bad&#8221; JPGs made in camera by relatives, which had no modification, have salvaged pictures many would have discarded.  Even large group flash exposures where the foreground were over exposed a bit, and background faces almost unrecognizable.  In Photoshop circled the face with a &#8220;magnetic lasoo tool&#8221; then adjusted them individually, and brought out all the detail.</p>
<p>Sometimes using Dodging and Burn in Brushes to modify small areas, etc..</p>
<p>Bob</p>
<p><span class='jfriedl'>Indeed, many seemingly-poor jpgs can be salvaged by good editing; I intended to imply only that the range is wider when working from raw data. &mdash;Jeffrey</span>
</p>
]]></content:encoded>
				</item>
</channel>
</rss>
