quattro, what about dynamic range and highlight recovery?

Started Jul 13, 2014 | Discussions
(unknown member) Senior Member • Posts: 1,857
quattro, what about dynamic range and highlight recovery?
4

I've seen some very nice images from quattro lately.
But i'm afraid there is a real issue with highlight recovery at base ISO 100.
In fact it is much worse than the DP2m at ISO 100.
Especially skies are often blown out without the ability to recover it. Very similar to the SD15.
SPP has a new "overexposure correction". Is it because they know about the issue?
Unfortunately this is only a "fake" correction, filling the missing information with color.
Until now i've seen only very few good DP2m vs. DP2q comparisons with available X3F to play with.
So maybe the DR of quattro is hidden in the shadows, and we have to learn a totally different way to photograph with the new Sigmas.

Some crops to make my point visilble:
There are more, but i don't want to copy too much pictures from other people in this thread.

Maceo

Johan Borg Senior Member • Posts: 2,508
Re: quattro, what about dynamic range and highlight recovery?

maceoQ wrote:

I've seen some very nice images from quattro lately.
But i'm afraid there is a real issue with highlight recovery at base ISO 100.
In fact it is much worse than the DP2m at ISO 100.
Especially skies are often blown out without the ability to recover it. Very similar to the SD15.
SPP has a new "overexposure correction". Is it because they know about the issue?

It may be because the Quattro sensor has a different way of overexposing which can be corrected for with this new feature? After all, the lower layers receive light over 4 times the area from before. This doesn't apply to the top layer, so the recovery is likely quite different from Merrill.

In practice I'm not sure there's much difference as I would deliberately overexpose the Merrill to get most out of the files in SPP, while that is not really useful with the better shadow color in Quattro.

jeanpierre19
jeanpierre19 Forum Member • Posts: 59
Re: quattro, what about dynamic range and highlight recovery?

The highlight recovery setting in DP2Q is for the JPEG-OOC (out of cam). The JPEG out of cam are very good. So you will less overexpose clouds. The X3F-File you can switch off the overexposure correction in SPP6!

 jeanpierre19's gear list:jeanpierre19's gear list
Sony RX100 III
Roland Karlsson Forum Pro • Posts: 26,336
Re: quattro, what about dynamic range and highlight recovery?

maceoQ wrote:

But i'm afraid there is a real issue with highlight recovery at base ISO 100.
In fact it is much worse than the DP2m at ISO 100.

OK. Lets assume it is true.

Better shadows and worse highlight recovery.

Might be so that the DR is unchanged from Merrill. The better shadows might only be because Quattro expose further to the right.

NOTE - I said might. More tests are needed to really know this.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma dp2 Quattro Sony RX100 III Pentax K-3 Pentax K-1 +14 more
Johan Borg Senior Member • Posts: 2,508
Re: quattro, what about dynamic range and highlight recovery?

jeanpierre19 wrote:

The highlight recovery setting in DP2Q is for the JPEG-OOC (out of cam). The JPEG out of cam are very good. So you will less overexpose clouds. The X3F-File you can switch off the overexposure correction in SPP6!

Agree about the excellent JPEGs, but the feature is also useful for RAW files in SPP and brings results closer to a Merrill. I definitely prefer turning in ON for challenging shots.

OP (unknown member) Senior Member • Posts: 1,857
Re: quattro, what about dynamic range and highlight recovery?
1

Roland Karlsson wrote:

Might be so that the DR is unchanged from Merrill. The better shadows might only be because Quattro expose further to the right.

Well, that's exactly what i mentioned already in my first post 

Roland Karlsson Forum Pro • Posts: 26,336
Re: quattro, what about dynamic range and highlight recovery?
1

maceoQ wrote:

Roland Karlsson wrote:

Might be so that the DR is unchanged from Merrill. The better shadows might only be because Quattro expose further to the right.

Well, that's exactly what i mentioned already in my first post

To test this you can set both M and Q at ISO 100 and then take a series of exposures manually, starting from under exposed and stopping at over exposed.

And then evaluate the result.

From that evaluation you can find the relative sensitivity of the two sensors and also the relative dynamic range.

To get absolute values is harder as you then have to make more thorough measurements using some standardised method.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma dp2 Quattro Sony RX100 III Pentax K-3 Pentax K-1 +14 more
xpatUSA
xpatUSA Forum Pro • Posts: 10,913
Re: quattro, what about dynamic range and highlight recovery?

Roland Karlsson wrote:

maceoQ wrote:

Roland Karlsson wrote:

Might be so that the DR is unchanged from Merrill. The better shadows might only be because Quattro expose further to the right.

Well, that's exactly what i mentioned already in my first post

To test this you can set both M and Q at ISO 100 and then take a series of exposures manually, starting from under exposed and stopping at over exposed.

And then evaluate the result.

Are we sure that Sigma uses the same 'ISO' method in both the M and the Q? The Standard, as Roland knows, allows three different methods to determine the number and allows quite a latitude (tolerance) on that!

-- hide signature --

Cheers,
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2s Sigma DP2 Merrill Panasonic Lumix DMC-G1 Sigma SD10 Sigma SD14 +15 more
Roland Karlsson Forum Pro • Posts: 26,336
Re: quattro, what about dynamic range and highlight recovery?

xpatUSA wrote:

Are we sure that Sigma uses the same 'ISO' method in both the M and the Q? The Standard, as Roland knows, allows three different methods to determine the number and allows quite a latitude (tolerance) on that!

Sometimes "the standard" is not even interesting. I have a K-5 and a K-3. The K-5 tends to clip, so I regularly put it at a compensation of -0.7. The K-3 do not have that problem. so I set it at 0. I do not think that the K-3 exposes less or have another opinion, it just have much more detectors in its evaluating exposure meter. So, it finds the clipping easier.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma dp2 Quattro Sony RX100 III Pentax K-3 Pentax K-1 +14 more
MarkWW
MarkWW Contributing Member • Posts: 619
Re: quattro, what about dynamic range and highlight recovery?

Roland Karlsson wrote:

xpatUSA wrote:

Are we sure that Sigma uses the same 'ISO' method in both the M and the Q? The Standard, as Roland knows, allows three different methods to determine the number and allows quite a latitude (tolerance) on that!

Sometimes "the standard" is not even interesting. I have a K-5 and a K-3. The K-5 tends to clip, so I regularly put it at a compensation of -0.7. The K-3 do not have that problem. so I set it at 0. I do not think that the K-3 exposes less or have another opinion, it just have much more detectors in its evaluating exposure meter. So, it finds the clipping easier.

If you read the post he's responding to, he says that the exposures should be taken at the same settings in manual mode and bracketed. So it's not a question of metering, it's a question of how each handles ISO.

If one is better at highlights and the other at shadows the question is - is it because they're metering differently or because the sensor is handling the the same metered scene differently.

MOD Kendall Helmstetter Gelner Forum Pro • Posts: 19,884
Mixed Results

In one picture strong light in clouds, I could not recover more than a half-stop of exposure!!!

However in others I've been able to recover more like a stop...  so it makes me wonder if there's a processing issue that can improve things.  I would especially love to see the raw data from a few shots I have that did not seem very over-exposed but had pretty large areas of cloud that could not be recovered.

The over-exposure compensation can sometimes get things about right but with a large enough blown area as you can see it simply turns it flat.  Adobe Camera Raw does the same thing to try and recover highlights too.

Basically I would not over-expose as works for the Merrill, I'd keep to default metering and use exposure compensation to guard highlights.

Luckily shadow recovery is better, the shadows are usually a lot cleaner.  I think overall there is more DR but not sure how much more (if any).

 Kendall Helmstetter Gelner's gear list:Kendall Helmstetter Gelner's gear list
Sigma 70-200mm F2.8 EX DG OS HSM Sigma 8-16mm F4.5-5.6 DC HSM Sigma 24-70mm F2.8 EX DG HSM Sigma 50-500mm F4.5-6.3 DG OS HSM Sigma 85mm F1.4 EX DG HSM +4 more
maple Veteran Member • Posts: 3,191
Re: Mixed Results
2

Thank you, Kendall, for your honest opinion. I'm glad that Maceo re-initiated this subject. I started a thread in the same vein, http://www.dpreview.com/forums/post/54010482, but got little attention.

There I linked a review: http://www.fotopolis.pl/n/19452/sigma-dp2-quattro-test/?page=8 which shows images that look encouraging regarding DR. They show more natural shadows and shaded areas, not as deep as Merrill typically renders them, while still retaining high lights reasonably well. I had another more careful look at the exposure parameters, to see how they did that. The finding is, well, interesting and still encouraging, though not conclusive. By referencing to the sunny 16 rule, all images are over exposed by between 1/4 of a stop to more than 2 stops. There are some high light clipping, but not overly so. Here’s the list of the “sunny” images, their respective exposure parameters and relative exposure level compared to sunny 16.

Unless Q over states its ISO, there seems to be about one stop latitude for over exposure with it.

-- hide signature --

Maple

OP (unknown member) Senior Member • Posts: 1,857
Re: Mixed Results

Thanks for sharing your experiences. This confirms the feeling i had.
Did you also shoot ISO 200 once? Does it behave the same way as Merrills ISO 200?
From what i see, i would shoot the quattro with an exposure compensation of about -0.7 in sunny weather.

Kendall Helmstetter Gelner wrote:

In one picture strong light in clouds, I could not recover more than a half-stop of exposure!!!

However in others I've been able to recover more like a stop... so it makes me wonder if there's a processing issue that can improve things. I would especially love to see the raw data from a few shots I have that did not seem very over-exposed but had pretty large areas of cloud that could not be recovered.

The over-exposure compensation can sometimes get things about right but with a large enough blown area as you can see it simply turns it flat. Adobe Camera Raw does the same thing to try and recover highlights too.

Basically I would not over-expose as works for the Merrill, I'd keep to default metering and use exposure compensation to guard highlights.

Luckily shadow recovery is better, the shadows are usually a lot cleaner. I think overall there is more DR but not sure how much more (if any).

MOD Kendall Helmstetter Gelner Forum Pro • Posts: 19,884
Re: Mixed Results

maceoQ wrote:

Thanks for sharing your experiences. This confirms the feeling i had.
Did you also shoot ISO 200 once? Does it behave the same way as Merrills ISO 200?
From what i see, i would shoot the quattro with an exposure compensation of about -0.7 in sunny weather.

That's a really good question, I did do a bunch of ISO 200 shots I've not really processed much yet... I'll let you know if it's any better.  That may well still be the case as I was playing with a few and now that I think about it the highlight recovery may have worked better.

 Kendall Helmstetter Gelner's gear list:Kendall Helmstetter Gelner's gear list
Sigma 70-200mm F2.8 EX DG OS HSM Sigma 8-16mm F4.5-5.6 DC HSM Sigma 24-70mm F2.8 EX DG HSM Sigma 50-500mm F4.5-6.3 DG OS HSM Sigma 85mm F1.4 EX DG HSM +4 more
Roland Karlsson Forum Pro • Posts: 26,336
Re: Mixed Results

Kendall Helmstetter Gelner wrote:

That's a really good question, I did do a bunch of ISO 200 shots I've not really processed much yet... I'll let you know if it's any better. That may well still be the case as I was playing with a few and now that I think about it the highlight recovery may have worked better.

If Quattro have no variable ISO dependent amplification, then exposing after higher ISO (i.e. really under exposing) would surely result in a better high light recovery.

 Roland Karlsson's gear list:Roland Karlsson's gear list
Sigma DP3 Merrill Sigma dp2 Quattro Sony RX100 III Pentax K-3 Pentax K-1 +14 more
Scottelly
Scottelly Veteran Member • Posts: 7,827
Re: quattro, what about dynamic range and highlight recovery?

My guess is that Sigma realized that people have been over-exposing their shots, just so they can get better shadows, by recovering the highlights. Rather than having people do that, which makes their JPEGs overexposed and look like garbage, Sigma has decided to make the camera work in a way that the JPEG versions of the photos look just right, when coming out of the camera. A good decision, if you ask me. I believe these are all arbitrary decisions, made by the programmers, at the request of management. I believe it was silly of Sigma to make the Merrills perform the way they did. Then again, they may have underestimated the way photographers would use the cameras to try to eliminate noise in shadow areas of the Merrill photos, and now they have adjusted for what they see happening.

It's an evolutionary process, I guess.

 Scottelly's gear list:Scottelly's gear list
Olympus TG-860 Sony SLT-A65 Sigma SD1 Merrill Sony DT 18-55mm F3.5-5.6 SAM Tamron SP AF 10-24mm F/3.5-4.5 Di II LD Aspherical (IF) +14 more
OP (unknown member) Senior Member • Posts: 1,857
Re: quattro, what about dynamic range and highlight recovery?

Scottelly wrote:

Sigma has decided to make the camera work in a way that the JPEG versions of the photos look just right, when coming out of the camera. A good decision, if you ask me.

Yes, but i'm not sure if they went too far in this respect.
I see blown out skies in so many normal landscapes pictures with clouds.

jeanpierre19
jeanpierre19 Forum Member • Posts: 59
Re: quattro, what about dynamic range and highlight recovery?

I tested ISO 100 vs ISO 200 on my blog you can read it: http://sigmadp2quattro.blogspot.ch/

Well, I do have to say that ISO 200 is very good and it will be my default settings, when I shoot on location with much highlights. Maybe, I will shoot ISO 100 occasionally!

SPP6 is very good and I hope Sigma will improve it, till final version

 jeanpierre19's gear list:jeanpierre19's gear list
Sony RX100 III
Keyboard shortcuts:
FForum MMy threads