Panasonic G7 camera - Controling the Dynamic range

DMillier wrote:

alcelc wrote:

Brisn5757 wrote:

alcelc wrote:

Highlight Shadow can't expand DR. It is to play inside the DR of a frame under the setting. It is the same (but in a much simplier way) as PP a shot in Photoshop. I might use the wrong word, but it will actually clip on certain tones. Yes, noise would be affected.

However, my custom curve -5,0 is used in another way round. i.e., at the expense of highlight overblown (on non essential areas of that frame), I can tune down the remaining highlight and so could have more room to push the exposure on the shadow, so will indeed reduce noise.

It is for SOOC jpg only.

Hi Albert.

Thanks for the info.

It's time I learn't more short hand term. What is PP?

Sorry, it is Post Processing.

Are what your say is that you slightly under expose the photo so you can push the exposure and not blow out the highlights?

I should better make it clearer. It is referring to the Custom HS of -5 (highlight) and 0 (shadow).

As explained earlier, it will be used to push on exposure, says +ev to increase the exposure to light up the shadow area. To have more light on the shadow can reduce the noise. (If I use a HS of XX, +5, it will bosst the noise in shadow.)

E.g., a scene has correct exposure value of 15ev (says. f/5.6, 1/2000"& ISO200). At that exposure value it will have no highlight overblown. But it is a high contrast scene and my aim is to better reveal the detail in the shadow. Under normal shooting condition, if I dial in +1ev (say, reduce shutter speed to 1/1000"), the highlight sections will of course be also brightened up and some places would be overblown.

By applying the custom HS, push 1ev, let highlight overblown (say those overblown areas are non essential area like gray sky originally without much detail). Due to the -5 in highlight the bright area (not yet overblown) will be tuned down. 1ev over exposure will light it up again to a result of looking like no HS be applied. I gain 1ev additional exposure to the shadow area.

The result is: I have more exposure on the shadow, a better noise control in shadow and highlight area been proper exposed at the cost of overblown certain non essential highlight areas.

By it I can take a SOOC image similar to edit it by Post Processing.

It is an interesting feature that we can tailor made it to fit our requirement like we do in PhotoShop.

From what I understand Highlight Shadow is something that can be done in Photoshop to alter the photo. Maybe there is some advantage in having the camera do the Highlight shadow instead of Photoshop?

Yes, PhotoShop can do it in and can be in a more detail manner other than just a S or inverse S curve. The in-camera HS is relatively too simple.

But if we can do it wisely, I found that I can reduce my Photoshop time. It is also a motivation to produce better (to me) SOOC jpg than shooting in raw. Back in the old days, after a 10 days trip I would normally have to deal with 7~8K+ raws when back home. Normally 1~2 months sitting in front of computer to do PS was never an easy task... If I can produce more usable SOOC jpg, I can plan my next trip <1 month after my last trip .

Brian

I'm not convinced this really works - at least not the way you think it does.

Try it you will know how it work.

It works well for me, but of course only on limited conditions (high contrast, I need to expose on the shadow, dispensable highlight area to trade off, not intend to do it by PP/ HDR...).

Before all, I suppose we must on the same ground that HS does not affect the physical DR of a sensor that can be recorded. So, manipulated on the curve is to tune down or boost certain light value, in case of HS, in a more smooth S or inverse S shape only. So, the brightest and the darkest won't be affected by HS. My method is to overexpose the shot, let certain brightest sector overblown but better exposure on shadow. The -5 in Highlight of HS is to protect the highlight area (not white saturated). Result: an image of similar exposure as no overexposure be made on the highlight, blown some non essential area and have higher exposure on shadow with noise in check.

"Correct" exposure with a digital camera is to expose as hot as you can without clipping highlights. Once highlights are really clipped, the data is lost forever and cannot be recovered.

Yes. But the problem is that for any high contrast scene, if HDR is not prepare to use, but I need to expose on the shadow, I will never insist to have every bit of data be retained (including something like gray sky without detail in it?)... I will overexpose the shot no matter HS be applied or not anyway. YMMV.

The in-camera contrast settings for jpegs do a very simple thing: they control the contrast curve that is applied during conversion to jpg.

Yes, I'm playing on it like I used to do in PS when my older models didn't have this feature. Of course, when more complicated manipulation be required (more than a simple S shape curve etc), I must go back to use PS and in such case, HS should be better staying away.

Now, even if you have exposed correctly, an excessive contrast boosting curve might brighten the highlights too much

Please note that my custom curve is -5 on highlight, when push exposure physically, those not yet highlight white saturated area will more or less be compensated by the HS effect.

- even though there is data there -

No, the blown area would be blown as per my decision.

and the final jpeg might end up looking clipped. That would be a bad thing.

No difference from what we do in PS(?).

But I doubt it works in reverse. If you genuinely overexpose and blow the highlights,

Not all highlight. It is just to blown those area of value 0 or will be boosted to 0 by the overexposure. Other highlight area of not white saturated will be retained.

applying a soft contrast curve cannot recover highlights that are gone.

Once gone, they remain gone. Pulling down the curve will just turn your blown out white areas to light grey which usually looks horrible.

Not intend to do so and so, in my custom curve there must be dispensable highlight area that overblown them won't affect the image.

If your method appears to work, I would suggest it is an illusion. Most likely, even with the extra exposure, your highlights were not actually blown even if you thought they were. So the highlight flattening still has something to work with.

The key point of my handling is to give up the non essential area (not much meaning to the entire image) in exchange for a better exposure of the shadow area exactly similar to what we do in PS. Applying HS is a simple way to handle it for SOOC result only.

Of course, if you shoot raw all this is moot as you have access to the full recorded data and can apply any amount of curves, exposure boots or reductions in post processing.

It is for SOOC result. In fact I can do it in PS on jpg (less head room) or raw (more headroom).

It is a challenge on getting SOOC through these new feature. YMMV.

-- hide signature --

DPReview gallery: https://www.dpreview.com/galleries/0286305481
Website: http://www.whisperingcat.co.uk/ (old)
Flickr: http://www.flickr.com/photos/davidmillier/ (very old!)