Micro 4/3 A question and a small rant on DR

wt21

Hall of Famer
This will be the thing that pushes me from m43, if anything does. Lack of ability to recover even minor blown highlights.

Using the 45/1.8 yesterday during some touring in the lovely weather, I took two shots of my wife. The EXIF is intact, but I'll give it here anyway: ISO 200, 1/640, F2.0 both shots. Both shot in RAW and run through Lightroom.

Conditions outside: perfect, blue skies (no clouds), pics done within 5 seconds of each other. Yet one is holding the highlights, and the other is blown. How can that be? As you can tell from the statue, all I did was shift over a couple of feet to try getting the statue in the background (turns out it didn't add much to the picture).

My question: how can the highlights be so blown in one, and OK in the other?

Rant: If I had shot this with my Canon DSLR, I could get those highlights back with a quick slider adjustments. The ORFs here are giving me nothing at all. I have one picture (shown here) with good highlights, and 5 or 6 (one example here) all with blown highlights. If only I had an assistant with a diffuser, lol.

All I can say is Grrrr...

(pics removed for privacy reasons -- wifey doesn't like pics of her posted on the internets)
 
The right shot still has some blow outs, but not as bad. On shots like these I underexpose. Ok, yes, there will be noisy shadows, but I'm not worried about noisy shadows on a portrait. Do the RAW files look drastically different with the same tone curve applied? What is the histogram showing for both images? Assuming all things were equal(shutter, aperature, lighting conditions) I don't think 1 stop difference in DR would have helped too much. If the blown out highlights even in RAW are blown out(meaning the pixel values for that area is set to white) there isn't much you can do. In bright conditions like these, I find myself always preserving highlights if highlights is part of the subject. Nice lens though, ain't it?
 
Whatever you do, don't post this on the Oly SLR forum on that major site.

I'd definitely want to see the histogram data on this. Even on 4/3, I'm always underexposing, then shadow recovering, to protect highlights. The thing is I really don't feel I should keep having to do that.
 
"Whatever you do, don't post this on the Oly SLR forum on that major site." -- lol, too true.

If I have to underexpose, and then deal with noise in the shadows, then that's still a DR limitations, though perhaps that's just one of those things I have to deal with. I find highlights recovery in my Canon images, though, were more like 2 stops+, especially in the case of the 5D. But, I don't want to lug around the 5D anymore, so here I am :)

The only thing I can think of that could make this difference, given all else is equal, is that by my moving one foot to the left, it caused her to turn her head slightly, and somehow the incident reflection of the light off her hair changed, being stronger in one shot than the other. Not sure how else this could be explained with no lighting changes and all else being the same.
 
"Whatever you do, don't post this on the Oly SLR forum on that major site." -- lol, too true.

If I have to underexpose, and then deal with noise in the shadows, then that's still a DR limitations, though perhaps that's just one of those things I have to deal with. I find highlights recovery in my Canon images, though, were more like 2 stops+, especially in the case of the 5D. But, I don't want to lug around the 5D anymore, so here I am :)

The only thing I can think of that could make this difference, given all else is equal, is that by my moving one foot to the left, it caused her to turn her head slightly, and somehow the incident reflection of the light off her hair changed, being stronger in one shot than the other. Not sure how else this could be explained with no lighting changes and all else being the same.

Cameras are really funny when it comes to relfected light. Looking at the shirt on the left side, there really isn't that much difference between the two. Both are blown out. Most of the difference I can see is in the hair, which could be because of the angle that the light is reflecting. All in all, both are good images.
 
I agree which is one reason I posted it. Just odd to get pretty different results, and like you said, the shoulders look exactly the same. I'm going to mark it down as a freak event, though I wish my camera always gave me the output on the right :)

The shoulder blow out I can live with. I just find the face/hair distracting :(
 
Yep, shooting with Micro 4/3 is all about preserving highlights. Unfortunately, the live histogram on both Olympus and Panasonic bodies is not really accurate. I have found I have to push it rather heavily to the left to avoid blown highlights.
 
I'm wondering about this and how Color Efex Pro could help render the image and redraw lost areas, as the Silver EFEX does.

I've been quite surprised with how well these two programs recover blown items better than LR.
 
"

The only thing I can think of that could make this difference, given all else is equal, is that by my moving one foot to the left, it caused her to turn her head slightly, and somehow the incident reflection of the light off her hair changed, being stronger in one shot than the other. Not sure how else this could be explained with no lighting changes and all else being the same.

This can and will make a difference. You moved your position, and she turned her head. She also seems to be leaning a bit forward in the second shot, and her smile is more contained (so her cheek won't be as raised to the light as in the first shot). If you had shot two in succession without any of these changes, I'd say you have a problem... but not in this circumstance.
 
I don't tend to find the percieved lack of dynamic range in the m4/3 sensors to be a problem for how I want to use the cameras, since I'm just as likely to blow a highlight or shadow in PP as I am to attempt to retrieve it. Having said that, I'm not often doing portraits so the issue described here isn't really a common scenario for me. I guess the thing about about a portrait is that is most frequently a "forced" shot, where the photographer has the oppurtunity to place the subject where they want them to and avoid exposure imbalances like this. The current 45/1.8 image thread at mu-43 shows that m4/3 cameras are more than capable of taking great portraits, although ironically I find the ones that have been professionally lit and exposed perfectly to be unnatural looking and lifeless. Kind of defeats the purpose of taking an image of a living human being for me, but I digress...

So...I guess it could be an issue and is something to keep in mind, but isn't something that effects me too often.
 
I like the posted image as it is. It has a maximum black and a maximum white. Good images should have that. Perhaps the issue is not so much the camera format as maybe the screen. In my calibrated monitor, things look ok. I like the feeling of the harsh sun glistening on her hair and blouse.
Before you ditch your camera, get some prints made. That will tell the tale.
Remember that old guy Ansel? Well, he fought with process and invented the Zone System.
So, you have maybe a Zone 10 or so and that's considered natural. You will never get every tone in a print unless you shorten the scale and then you could loose the dynamic range all together.
Get a few prints made..... Then decide.
 
You are right about printing -- I bet the white wouldn't be so bold (prints not having back lighting and all). My Dell IPS panel, though, is calibrated.

Not at all ditching the camera. That's what I meant by "small rant." If you want a compact system camera, where else are you going to go? Oly is smooth as butter to use, fantastic lens assortment, classic styling, just the right size (not too big, not too small), but I could pull those highlights back with CR2 files. Gone in ORF. That's all. Just a small rant. If I can't find some sympathy here, then I'm not sure where else I could find it ;)
 
Pout, boo hoo...I'm shedding a tear fer ya my friend....
Geeze ya got it tuff ya do. I mean after all is said and done... Ya got a compact camera system that delivers big time...
Yeah, me heart is aching fer ya and ya gotz my sympathy...
 
Awright. I get it. No honesty here. We'll just all pretend there can't be anything even slightly off kilter with any of our equipment. I know it's me. It's not the camera. I'll go in the corner and flog myself with the cat o' nine tails for 20 minutes.
 
Nah.... Just get some prints made. You'll be surprised at the difference from the screen image.
Seriously this is a reason I moved to APS sensors. The tonal range is just more defined.
I did love my Pen1 and GF1 but the APS just gives me what I want.
Anyway, I like the image posted the way it is.
Don
 
I agree that trying to retain the original mood or luminosity is sometimes impossible without blowing out the highlights. It is frustrating & in scenes lit by harsh directional light, the mood can often be destroyed by reduced exposure. On the other hand, I have been surprised by the extent to which one can recover shadow detail by using the 'Fill Light' feature in LR.

I haven't used APSC and am gratified to learn that the same problem doesn't exist. It may indeed be a reason to switch to larger sensors.
 
Don't blame me for going to APS sensors....ok blame me...

The fill slider is a great way to fill in shadows and midtones.
The most important slider is really the Clarity slider. It will give details where they are low and in SEP2, the structure is a lifesaver...
 
Back
Top