Is this true about the camera's Histogram

[m_tt_donuts]

<p>I purchased a book by David duChemin, and the author states that the data of images shown in the histogram of the camera is not spread equally. When looking at the histogram of your camera, the left most quadrant holds 3% of the data, the next 7%, followed by 10%, and lastly 80%. He says it's a mathematical thing. I made in image as an example in paint demonstrating what he says, but I haven't found another site/place that says the same. He's making the point that if the histogram is shot like the top, you could edit to your hearts content, shoot like the bottom and you'd quickly start showing noise, etc. compared to taken like the top. Is that true? I didn't know the right most quadrant holds 80% of the data available, previously had I seen the histogram on the bottom I'd been content but now I'd reshoot. DThanks<img src="http://s8.postimage.org/r0al99ev9/histogram.jpg" alt="" width="640" height="433" /></p>

<p> </p>

[john_deerfield]

<p>This is just another way of saying Expose to the Right (ETTR).</p>

<p>http://www.digitalphotopro.com/technique/camera-technique/exposing-for-raw.html</p>

<p>And that only works with Raw files. With JPEGs, I prefer a slightly underexposed file. Personally, yes, I shoot with ETTR in mind. I thing the above examples are quite extreme. To shift that curve as much as it's shifted is something pretty close to 2-f/stops and depending on your camera that might be outside what you can comfortably capture. The real problem in looking at these histograms is that your subject is very consistent in tone which is rare. There are no shadows? There are no highlights? And each of these are going to reveal more information in regards to <em>where</em> you want to put the exposure.</p>

[jcuknz]

<p>These people may be correct, they probably are, but as to if you can or cannot edit the 'poor' example I think is generally over stated and I simply get on with it. If adjusting improves the image I do it irrespective of what the histogram shows me. The eye is the final arbitrator not some graph.</p>

[aqualarue]

<p>Certain images of mine, mainly the flowers in my gallery, were all exposed about one stop over the edge of the right side of the histogram in camera, in other words 4 stops over in total in my camera, you can only do this with camera RAW, not any other setting. In some photography this information is absolutely vital in order to capture as much info as possible in camera. In the darkroom you expose for the darkest point and develop for the highlights, it is the other way around digitally. In the darkroom you can manipulate a film during development, in the light room you can manipulate the digital file to bring out exquisite highlight detail if you really Push the exposure slightly beyond the limits. In Documentary and documentary portraiture I would never do this, I behave as if I have a film camera and take exposures quickly by predicting what might be needed, in artistic photography one has absolute control over everything and can really push the histogram even beyond its range on the right and then through Camera raw bridge can devlope that image. Also the camera, as a rule, tells you when the highlights are over exposed, if you shoot raw Ignore this, you can often expose up to 1 and 1/3 stops over this without blowing out highlights, sometimes a little more, practise this if you can you will be surprised. </p>

[m_tt_donuts]

<p>Thanks, my drawing wasn't the best but at least it got the point across :) The Figure 1 of page <a href="http://www.digitalphotopro.com/technique/camera-technique/exposing-for-raw.html?start=1">http://www.digitalphotopro.com/technique/camera-technique/exposing-for-raw.html?start=1</a> really explained to me visually what's going on. Looking at it in that manner, it's easy to see the darks don't have a lot of tones, if I break up that chart into 3rds I see darks hold 8% of the tones, mids hold 25%, and lights hold 67% so it would agree with what the author was saying. Explains a lot to me about editing dark images vs. lighter too. Thanks so much!</p>

[bebu_lamar]

<p>The reason is primarily due to the fact that when the camera reads the analog values from the pixels and digitizes them it doesn't do it in a linear fashion. There are fewer steps for the dark part than the bright part because the camera use a gamma</p>

[jaydesi]

<p>It's true. The further to the right, the more light that is hitting the pixels on the sensor, so there's more data.</p>

<p>As an aside, this is how fill flash is able to brighten the dark areas without blowing the highlights. The flash needs to fire at much less power to bring those dark areas up substantially, whereas its impact on the already bright areas is negligible.</p>

 

[Laurie1]

<p>Mätt, though I have forgotten most of the detail I do remember having this link to a website which gives a very good run down on it. Here it is.....<br>

<a href="http://www.luminous-landscape.com/tutorials/expose-right.shtml">http://www.luminous-landscape.com/tutorials/expose-right.shtml</a></p>

<p>And here is another I found very useful.....<br>

<a href="http://www.luminous-landscape.com/tutorials/understanding-series/understanding-histograms.shtml">http://www.luminous-landscape.com/tutorials/understanding-series/understanding-histograms.shtml</a></p>

<p>best regards<br>

Laurie</p>

[Laurie1]

<p>Sorry, I just found there is an updated version of the article <a href="http://www.luminous-landscape.com/tutorials/optimizing_exposure.shtml">here</a>.<br>

Laurie</p>

[pete_s.]

<p>No, the author is wrong. Not completely wrong but still wrong.</p>

<p>The histogram shows what tones the camera generated jpeg contains, ranging from 0 on the left to 255 on the right.<br /> Even when setting the camera to raw only a jpeg is still generated and it's the jpeg that you see when your preview your images on the back LCD.</p>

<p>To generate the jpeg the camera takes the raw data and applies white balance then demosaics the image. Then lens correction and other functions are done. After that shadow, highlight recovery if those functions are active. Then saturation, contrast and a tone curve (often s-shaped) is applied. Sharpening and noise reduction might also be done at this stage. Finally the image is transformed into the selected color space (sRGB or aRGB) and gamma is applied.</p>

<p>Basically there is a lot of stuff going on so the transformation from what the camera sees to what the histogram looks like is not straight forward. And it doesn't look like the histograms above.</p>

<p>What the author is right about though is that the more photons that are captured by the sensor the better. And that means that the histogram will show more information to the right. And that you have a better signal to noise ratio so the noise doesn't show as much.</p>

<p>PS. Sorry if this was too much information. I didn't realize that I was in the beginners questions forum.</p>