I have downloaded and perused the User Manual - and it is very good, and clear.  However, there is little or no discussion of "how" to use the information provided and why you want to examine it.
I purchased the software not to "cull" RAW images, but to use the RAW display and historgrams to understand exposure and lightling and how to improve my shooting.  I am specifically interesting in uderstanding where HDR processing is needed or will benefit the quality of my photos.  I don't have a good sense of how to judge this just looking at a scene - I was hoping FRV could help with this.
It would be really nice to have some tutorials that discuss how to use the information provided and how to apply it to shooting "difficult" scenes.
For example, should the RAW histrogram be interpreted differently from the historgrams I see in Lightroom?  How much should I worry about the percentages shown for underexposed and overexposed pixels?
This is a very neat program and I am looking forward to better understanding how it can be used.

I, too, would like to see an advanced tutorial to help me develop additional understanding of some of what the program shows me.

Please ask, that would help us to understand what questions need to be addressed.

Dear David,

Histogram in the Lightroom is based on the current image rendition, that is it is based on the adjustments you've made and conversion Lightroom came up with based on those adjustments as well as its internal design decisions.

Raw histogram in FastRawViewer (and RawDigger, too) is "ground truth" - it is the histogram of the raw data in the shot, before any conversions at all, "as is".

Such a raw histogram is useful when evaluating what can be extracted out of the raw data during the raw conversion; that is, for example, how much of real highlight clipping is in the file. It is also important that the raw histogram in FastRawViewer and RawDigger shows true amount of underexposure on the shot (headroom in highlights), and in FastRawViewer it is easy to see how much exposure can be added.

Raw histograms allow to diagnose false channel clipping, like when shooting saturated objects (flowers, for example). It is quite an often complaint that when shooting a red rose under daylight the red channel clipping indication based on the "conventional" histogram (in-camera or in a raw converter; Lightroom being one of raw converters) is premature, thus leaving the detailed blue channel underexposed and noisy. For the similar reason, photographers complain on the noise in the red channel in the sky. This happens mainly because before displaying the warning in the conventional case white balance is applied, multiplying the raw values in red and blue channels by 1.6 to 3. Having raw histogram allows to avoid such false warnings and to select (for raw conversion) among the series of exposure-bracketed shots the one that is correctly exposed.

The amount of clipping to be allowed is somewhat of a personal choice, generally I consider 0.3% in the strongest channel OK, if not in a single important area (to see what is clipped press "o" to display clipping overlay). Normally 0.3% poses no problem at all for Lightroom highlight recovery feature (even if in a contigious area of the shot).

Underexposure limit is user-selectable. Depending on the camera and ISO setting as well as on my intended image use I may set it between 7 EV and 10EV. The default is reasonable 8 EV. The reason to keep the underexposure limit below 8 EV is not just noise, but also flare and glare considerations. Flare and glare result in the loss of linearity in shadows, which makes white balance and colour control problematic. At -8 EV typical flare in a well-controlled studio shot causes non-linearity of more that 2/3 EV.

I hope I addressed some of your questions. Please feel free to ask futher. Your questions help us understand what needs to be covered in workflow mini-manuals and tips.

That's helpful.  I shoot RAW+JPG most of the time and my main camera is a Sony A6000.  If I use FRV to look at both the RAW and JPG versions of the same shot, the histogram looks different, but the exposure stats show the same values (that is, the proportion of each channel that is under- or over-exposed).  I don't see the extra information from the RAW data in this case - if I am trying to understand what is not exposed correctly, the RAW and JPG historgrams and statistics are giving me the same information.
The A6000 does a very good job with respect to dynamic range - and the JPGs are generally very good.  I think I frequently misinterpret the need to "expand" dynamic range via HDR techniques.  Images where I would have guessed HDR was required really don't - the camera captures the entire range with only minor under- or over-exposure.  At least, that is what FRV is showing me for the shots I thought would be "difficult".  Beyond the obvious situatilons where you are shooting a dark room with a very bright window (rare for me), I am starting to think HDR is not needed.  And, the A6000 has an "Auto HDR" function (it takes multiple shots and combines them in camera) that works very well if I want "insurance" that I capture the shot.
I realize this is a very complex issue.  Perhaps some sample photos that illustrate the issues you describe would be helpful.  If I can "see" it - then I understand it better.  If these are theoretical issues that I can't see without doing a lot of "pixel peeping" (I have an iMac 27 Retina - so that is possible!), then I am not sure how useful this is.
Again, thanks for the response.  This is the kind of discussion that is very helpful to users, I think.

Dear David,

Histogram should change from raw histogram to JPEG histogram when you switch between raw and JPEG. On the title bar of the histogram window the change is indicatted by switching between title saying "Raw Histogram" and "JPEG histogram", like it is on these screenshots:



Statistics however is always derived from raw, as this kind of statistics can't be derived from JPEG.

Yes, as I noted, I could see differences in the "shape" of the histogram when viewing them first from the RAW image and then from the JPG image.   The exposure stats were the same.
If the exposure statistics are always derrived from the RAW data, then that is why they are the same for both the RAW and JPG histograms.
But, if I get what you said earlier, if I imported the raw image to Lightroom and looked at the histogram, and turned on the "show shadow clipping" and "show highlight clippling" that would not necessarily agree with what FRV was telling me.  (LR doesn't give me a count, but I could get a general indication of the extent of clipping by viewing the image itself with these settings on.) If FRV says, that, say, 10% of the pixels were underexposed, I should see that in LR too - but perhaps, FRV would show a smaller proportion than LR - more that could be recovered in Develop.
Yep, complicated!

Dear David,

if I imported the raw image to Lightroom and looked at the histogram, and turned on the "show shadow clipping" and "show highlight clippling" that would not necessarily agree with what FRV was telling me

That is one of the major FRV design goals - to provide statistics and histogram directly from raw data, independent of any conversion settings and algorithms. Using a good raw converter, you will always be able to bring the data existing in raw to the converted image.

If, say, FRV indicates clipping of highlights, it means the highlights are truely clipped in raw data. If some raw converter shows those as not clipped, it means highlight recovery is applied.

If FRV indicates raw data in highlights is not clipped, but a raw converter shows it as clipped, it means the "exposure" slider in raw converter can be pulled back without any fear of "inventing" data in the image.

Underexposure setting and warning in FRV seems to be not very well-understood. It does not mean the shadows are clipped, one still can push those shadows; it only means noise and non-linearity are higher than a user wants.; while what exactly he wants depends on the desired quality and experience. For high quality on the resulting images I would not go above 9 stops below highlight clipping point, keeping it mostly at 8. With accurate ETTR 8 stops is what the "beach scene" on the home page here has; so it is quite a lot for many shooting situations.

Thanks.  I am getting a better idea of what FRV can do for me.  This is exactly the kind of material that could go into tutorials to supplement the manual.
I used to use DxO, and liked their manual because it had a lot of illustations showing what their (sometimes obscure) adjustments really did in a visual sense.  A set of sample images that illustrate the features of FRV - along with a brief description - would be really helpful I think.  Beyond being helpful to your users, I think this would also be a very helpful marketing tool.  You are competing against (very powserful) complete tools like Lightroom - and many photographers (like me) would not understand why another RAW tool can be benficial.  And, you never know why people can be interested in your product.  It seems to me that the primary marketing feature for FRV is its speed - and potential use to review large volumes of RAW images.  That wasn't my motivation - I was looking for better understanding of what my camera was really capturing and how it was dealing with wide-dynamic images.

Add new comment