How to Tell Whether Raw Photos are Keepers?

[21limited]

I shoot Jpeg plus Raw. When I shoot on vacation, and the shots will be used just a travel slide show for myself and family, 98% of the jpegs are acceptable. I might apply an increase in saturation to all the shots and that;s it. I still can use the RAws for the 2% that need more help. It;s just a waste of time to edit from Raw for a vacation. For "artsy" shots, then I can spend my time on RAW. So the point is, what you intend to do with the photos defines your process.

Which brings us back to the original question… how do you decide which need to be raw.

When I load the raw into the computer loads the jpeg treatment of the image, and that’s the first thing I see. If I press the edit button, then I see an interpretation of the raw files. Because I see the SOOC jpeg image before I change anything anything, I know what it looks like. WHen I push the magic wand button, I see the AI generated raw version, that is always, 95% of the time, better than the jpeg version, but I can keep it if I wish. I don’t have to load the jpeg treatment to see the jpeg treatment, but I can use it if I wish. I only load one version. the raw, but I can still use the jpeg rendition. Even using the single AI "magic wand” almost always improves the images over teh jpeg. Once in a while the jpeg images matches the output, and makes no changes.

The questionable thing here is that you say 2% need more help, but unlike my process, I wonder how you determine that. Do you actually look at what the raw does, and decide the hpeg is good enough or is it only in cases where the jpeg clearly didn’t work that you try the raw, not the ones where it might be better but you don’t care because what you see is good enough.

 

[21limited]

This post highlights not only what you can recover from a RAW file that would be unusable in a JPEG file, but it also highlights the fact that the camera cannot record what you're looking at with your eye, and it's vital that photographers understand the difference.

As you look at a scene with your Mark 1 Human Eyeball, the only part of the scene being processed by your brain is what you're directly looking at. If you're looking at the far shoreline, the foreground and campfire are irrelevant, but you see the far shoreline very clearly. As your eye moves to the campfire, everything else become irrelevant. You can see the dancing flames of the campfire pretty well, unless the flame is bright enough to actually be painful to look at. Then you move your eye to the near shoreline, or to the rocks immediately in front of you. Again, everything else in the scene becomes irrelevant, and your brain process and displays that near shoreline or those rocks perfectly. Your brain closes or opens the iris in your eye according to what your eye is pointed to, looking at directly. You have auto-exposure for the small piece of the scene you are actually looking at. As you look around the scene, if you don't realize how much exposure difference there is in the pieces you're looking at, you'll be thinking to yourself, "What a lovely shot this is going to make!"

The camera is completely incapable of making those exposure decisions around the entire scene. It records the entire scene with a single exposure value setting combination of aperture and shutter speed, either making a best guess from its metering, or using your best guess that you've applied to the settings manually. The range of exposure values in the scene exceeds the camera's ability to record them all, and you get excessive brights and then dark areas with no detail. The image was salvaged (if that's really the right word) on the computer by bringing up the exposures in the dark areas, and accepting the highlights in the sunset glare, which were pretty much blown out anyway. His description of the process, using color correction as well as exposure correction, points out how saving such an image might be difficult, but that workflow gets easier with experience, as you try things and find what works. For example, with the second image, he's thinking, "Well dang, that's better in the foreground but it really screwed up the sky! That looks like a cartoon or something!" and that's when he started in with the color corrections he described.

My point overall is that the photographer must realize that the camera is incapable of accurately recording what he's looking at when there is a wide range of light values (dynamic range, as 21limited pointed out using the letters DR) in the scene. HDR might have done justice to this scene, or might have introduced its own set of difficulties. HDR would require several shots from the exact position, rather difficult if handheld, even with continuous high-speed shutter release.

You nailed it, thanks…. it takes so long to explain everything. All help appreciated.

 

[smoke665]

OP is not participating in the discussion.

Considering most of the comments have been off thread he likely got tired of reading. LOL

 

[smoke665]

OP is not participating in the discussion.

Considering all the off thread comments he probably got tired of waiting fir an answer to his post. LOL

 

[Ysarex]

I believe full size would refer to resolution. But what about compression? How is that determined?

Right -- resolution is the same but the compression rate could be different. The most investigating I've done is to just look at the two side by side and they appeared the same.

 

[AlanKlein]

Which brings us back to the original question… how do you decide which need to be raw.

When I load the raw into the computer loads the jpeg treatment of the image, and that’s the first thing I see. If I press the edit button, then I see an interpretation of the raw files. Because I see the SOOC jpeg image before I change anything anything, I know what it looks like. WHen I push the magic wand button, I see the AI generated raw version, that is always, 95% of the time, better than the jpeg version, but I can keep it if I wish. I don’t have to load the jpeg treatment to see the jpeg treatment, but I can use it if I wish. I only load one version. the raw, but I can still use the jpeg rendition. Even using the single AI "magic wand” almost always improves the images over teh jpeg. Once in a while the jpeg images matches the output, and makes no changes.

The questionable thing here is that you say 2% need more help, but unlike my process, I wonder how you determine that. Do you actually look at what the raw does, and decide the hpeg is good enough or is it only in cases where the jpeg clearly didn’t work that you try the raw, not the ones where it might be better but you don’t care because what you see is good enough.

I'll use the raw file if I feel the exposure needs improvement more than the jpeg allows. The other reason is for cropping. The jpeg image is set to the selectable format is 3:2, 16:9, etc.. The raw file allows more room to maneuver what part of the image I want that the camera actually caught, which is a full uncropped sensor range.

Question for you. What editing program do you use that allows getting the jpeg image that would have been created in the original camera?

 

[AlanKlein]

Right -- resolution is the same but the compression rate could be different. The most investigating I've done is to just look at the two side by side and they appeared the same.

The compression in a separate jpeg file might be better than the one buried within a RAW file. I wouldn't;t rely on the buried jpeg without knowing what those values.

 

[Thiophilos]

Considering all the off thread comments he probably got tired of waiting fir an answer to his post. LOL

It was a hit-and-run post. OP hasn't been seen since post #3.

If OP's pics look bad on the LCD, most likely they are bad pics. A preferred solution is to find better subjects and/or better lighting.

But the situation could be different, and only the OP knows.

 

[21limited]

I'll use the raw file if I feel the exposure needs improvement more than the jpeg allows. The other reason is for cropping. The jpeg image is set to the selectable format is 3:2, 16:9, etc.. The raw file allows more room to maneuver what part of the image I want that the camera actually caught, which is a full uncropped sensor range.

Question for you. What editing program do you use that allows getting the jpeg image that would have been created in the original camera?

I shoot raw on one card, and jpeg on the other. I have to compare the jpeg engine with my raw output, and when conditions are right, sometimes the jpeg can do marginally better than I can, but not enough to tell which is which of you’re ot looking t both side by side. I once set out to prove that with correctly exposed image that no matter how good the jepg was, I could do better with the raw. Turned out I proved the opposite. The two images were virtually identical, except in a curious pat of the image the jpeg was better. It was only about 5% of the iamge, but still, the rest was equal. As an exercise it was quite frustrating because I was trying to match an image rather thn just doing the best I coud with the raw. It was 45 frutratig minutes with the last 30 minutes concetrated on minutae. I don’t think I still have the orgianls.

I still shoot all raw, on one card, and all jpeg on the other for every image.. If it’s quick and dirty job, I’ll use the jpegs. The last time was accompanying a friend cottage hunting and putting together a folder of the interiors and exteriors of the places we visited for future reference. That was maybe 4 yeras ago.

 

[William Baroo]

I haven't been getting notifications, so I thought this discussion was dead.

 

[Thiophilos]

I haven't been getting notifications, so I thought this discussion was dead.

It is dead.

 

[DanOstergren]

I haven't been getting notifications, so I thought this discussion was dead.

Folks in this forum love to whine about thread authors not engaging as much as they think they should. It's their problem, don't take it to heart.

 

[greybeard]

My RAW shots look the same as JPEGs on my camera monitor and with LR or NX Studio. Big difference is that I have a lot more processing options with RAW than I do with JPEG.

 

[AlanKlein]

My RAW shots look the same as JPEGs on my camera monitor and with LR or NX Studio. Big difference is that I have a lot more processing options with RAW than I do with JPEG.

I believe LR might be automatically applying the same jpeg camera edits as the camera does to create its jpeg file when you first open the RAW file in LR. That's why both the raw and jpeg files images look alike. You have to reverse the LR edit to get back to the camera's original RAW image. Or you can start on the RAW image where LR started with its initial jpeg edits and just make minor adjustments as required from that point. That way, you don't have to start from scratch with the RAW image. The edits could work quicker.

 

[21limited]

Aplple photos is the same. As soon as you press the edit button it reverts to a raw interprretation, But if you just save without editing, you get the suggested jpg output.