ACR Histogram different?

[ShaneS]

I was working on this image and wonder why the histogram looks different in ACR than in Photoshop with no further adjustments made? Is this due to different color spaces Adobe RGB vs. RGB?

Also, is the fact the the colors are off the chart vertically a bad thing in terms of image quality?

ACR Screenshot:



Photoshop CS6 screenshot:



I'm very confused...

[pnodrog]

They look reasonably similar to me. Just different graphical representations and they are only indicative at the scale shown.

In CS6 you can expand the view of the histogram by selecting the options given in the lines/box in the top right hand corner in the histogram window.

[IzzieK]

I agree with Richard, Shane. I can't see the difference either. It is just another program...Even CS5 can show the expanded view.

[Manfred M]

Both pieces of software show the same thing in two different ways. Having individual channels "go off the chart" at the top are really related to the colour distribution in the shot, versus the way the vertical scale is laid out; so no issues or impact on image quality. The vertical scale simply shows how many times that specific colour occurs in an image. If you took a picture of a green piece of paper that has the same shade throughout, and lit it evenly, so there are no different shades of that green, you would have a single vertical line, as every pixel in the image would be identical. The image (while perhaps a bit boring) would still be displayed properly and could be printed properly as well.

The only time that image quality will suffer is if you see data at the extreme left or extreme right of the chart (i.e. pure black and pure white), regardless of how high the bar is.

[davidedric]

I have long been puzzled how the pictorial representation of the histogram is generated. The total area under the histogram (AUC) must somehow represent all the pixels in the image. But it must also be "normalised" otherwise a low resolution image would have a much flatter histogram than a high resolution.

My tentative conclusion is to look at the general shape of the histogram, and not to pay too much attention to the height of individual peaks.

As to why the two versions of ACR produce different representations - no idea

[ShaneS]

Thank you! I see what you mean about the shape being essentially the same and the vertical scale of the graph changing.

Much appreciated