Canon 5D Mark IV brings dramatic dynamic range improvements to the 5D line

[Don Haines]

Putting the RAW file out there is a big thing!

To do so, you have to be ready to say "this is the best edit that I could do" and be willing to see if others can do better. People are going to play with that RAW file and someone out there is going to do a better job editing it than you did. (not MAY do a better job editing it, WILL do a better job). This person(s) who do a better editing job will then respond and will (hopefully) tell us how they did it and we will learn from them.

by not releasing the RAW, the suppositions and suspicions continue, and the arguments continue with no hope of resolvement.

 

[davidmurray]

If I owned a 5DS line camera,I would want to see it updated real soon now.

Why would you own a 5DS camera if you aren't happy with it?

Premium camera in terms of MP, delivering less than premium IQ when compared to a "lesser" camera.

I suppose that's a pretty subjective point of view to take considering I don't own a 5Ds.

People bought the 5Ds/at when they wanted megapixels above all else.

That rule still applies because the 5Ds/at bodies are still 20mp higher than any other Canon bodies.

 

[mmenno]

Reading these forums it completely amazes me how so many people can blatantly deny more dynamic range is better..
There's a reason Rishi has the job of testing and reviewing cameras for a living and you guys don't. Dynamic range is an important quality in a camera, whether you know or use it or not. But rest assured, even as cameras get better, you'll still be able to add all the noise and banding you want in post

 

[Sporgon]

Reading these forums it completely amazes me how so many people can blatantly deny more dynamic range is better..
There's a reason Rishi has the job of testing and reviewing cameras for a living and you guys don't. Dynamic range is an important quality in a camera, whether you know or use it or not. But rest assured, even as cameras get better, you'll still be able to add all the noise and banding you want in post

Then you should read more carefully. No one here has ever denied more is a bad thing. The arguments start when someone under exposes an image shot on Canon, then claims you cannot do a satisfactory shadow lift without degrading the image quality, and defines this as the only quality the sensor can achieve, when in actual fact if the shot had been exposed more to the right without losing any highlights the lift would have been fine.

Of course the reason they have to do this is because under more optimum exposure the shadow lift required to make the image IQ look bad would be cartoon-like and rediculous, and this is because 12 stops of real EV range are more than enough in most uncontrived circumstances, and when they are not the required range tends to be many stops above even 13-14. Witness the recent post of a guy who's just bought a 5DIV and hasn't got enough DR to cope with a cruise ship lit up with bright lights in harbour at night.

I know, I know, the D810 would have done it :

 

[mmenno]

Those deliberately severely underexposed shots are meant to more obviously show the difference in dynamic range between sensors, and they serve that purpose very well. When camera A can be pushed 5 stops and camera B only 3, this simply shows Camera A has better DR, it doesn't mean every shot you take with it has to be underexposed by 5 stops, it just means you can, and with camera B you can't.

I still own a 7D1, and the lack of dynamic range in the 5D2 and 5D3 has kept me from going full frame for the past few years , as my motivation to do so is largely image quality. So personally I'm very happy that Canon has finally been pushed to upgrade their sensor technology, even though apparently not everyone sees it's benefits.

 

[Jack Douglas]

Those deliberately severely underexposed shots are meant to more obviously show the difference in dynamic range between sensors, and they serve that purpose very well. When camera A can be pushed 5 stops and camera B only 3, this simply shows Camera A has better DR, it doesn't mean every shot you take with it has to be underexposed by 5 stops, it just means you can, and with camera B you can't.

I still own a 7D1, and the lack of dynamic range in the 5D2 and 5D3 has kept me from going full frame for the past few years , as my motivation to do so is largely image quality. So personally I'm very happy that Canon has finally been pushed to upgrade their sensor technology, even though apparently not everyone sees it's benefits.

We are all happy with whatever is improved but don't fret about what we have, which for the most part is fabulous. Nothing like competition.

Jack

 

[mmenno]

I mostly agree with you, but what I don't get is why a serious professional reviewer gets whined at for even daring to touch on the fact that Canon's dynamic range has simply been demonstrably inferior for the last couple of years.

 

[scyrene]

I mostly agree with you, but what I don't get is why a serious professional reviewer gets whined at for even daring to touch on the fact that Canon's dynamic range has simply been demonstrably inferior for the last couple of years.

This discussion has gone over your head then.

 

[Sporgon]

I mostly agree with you, but what I don't get is why a serious professional reviewer gets whined at for even daring to touch on the fact that Canon's dynamic range has simply been demonstrably inferior for the last couple of years.

Because the information is presented as the Canon sensors not being able to cope with the desired EV range and tonal adjustment to taste, when in fact they can when not crippled by under exposing and putting the dark tones onto the sensor floor. This is exactly what was done with the girl half and hour after sunset shot on the 5Dsr.

Everyone knows by now that the older generation Canon sensors were behind on "DR", but those who come here to demonstrate it produce, without fail, a contrived under-exposed shot and then state this is the best the sensor can do. That's the issue. In fact some of the under-exposed shots shown here as examples of poor "DR" have even the highlights under exposed by about four stops !!

If I have to choose between the 5Ds with its older, off chip ADC, and the newer 5DIV with its on chip ADC the difference in DR potential of the two cameras wouldn't even be considered in my decision. Now of course if we were talking about the difference between 6 and 13 stops it would.

 

[mmenno]

I've seen the picture you mention on dpr, but as I'm on my phone at this time I can't do any serious assessment of it's precise exposure. As I recall though, there was mention of the highlights being pushed all the way to the right, and it would surprise me if this was an outright lie.
But aside from all the 'test pics', whether done properly or not, the fact remains that the sensor performance, DR wise, on canon cameras has been lacking for quite a while. When I picked up my 7D I was surprised to see so much noise and banding, not only in deep shadows but sometimes in clear skies as well. I didn't even know what caused it at first, but after some reading I found out it was just a sponsor design issue, not a fault in my specific body. Ever since then I've been waiting for Canon to get with the times, and now finally they seem to have..
Don't get me wrong, I love a lot about my camera, and Canon camera's in general, but DR has been a serious gripe for me, and others as well, so I am glad that Sony made better sensors and reviewers compared them to Canon's, so we all benefit in the long run.

 

[Sporgon]

I've seen the picture you mention on dpr, but as I'm on my phone at this time I can't do any serious assessment of it's precise exposure. As I recall though, there was mention of the highlights being pushed all the way to the right, and it would surprise me if this was an outright lie.

You are right, in the 5DIV DPR article he says "highlights just short of clipping". Earlier, in the arguments here, he stated "about one stop" or "just under one stop" off clipping. So regarding lying; well you judge.

I calculated from the time after sunset and the exposure triangle info given in the original 5Ds article, that the highlights were about 1.5 stops under clipping. That puts the deep shadows 1.5 stops deeper in very thin (dark) light, which would make a huge difference in the ability to lift in post. On any current camera actually; even on a D810 you would have been losing tonality.

 

[KeithBreazeal]

Personally, the 5D mark IV meets my DR needs. Balancing the Lights & darks so that finished edits represent what my eye saw is what I want. The improvements in shadow information is noticeable- especially color noise.
The other thing I have noticed is that I'm not having to mess with color temps as much. I think the RGB metering is a huge improvement.

As soon as LR updated to support the Mark IV, I ran this shot as a test for bringing up the shadows.

Canon 5D Mark IV test LR6 3255 © Keith Breazeal by Keith Breazeal, on Flickr

One thing I noticed in this photo is that the metering system nailed the skin tones despite the yellow light cast from the canopy over the pit area.

Reno 2016 Steve working on Voodoo 4255 © Keith Breazeal by Keith Breazeal, on Flickr

And just because I was curious, the 5-stop push...

Canon 5D Mark IV 5 stop push LR6 3379 © Keith Breazeal by Keith Breazeal, on Flickr

I was actually a bit surprised that the deep shadows pulled up that well. In the past, I would have had to use DXO to pull it up that much- with a noise penalty.

 

[mmenno]

I've seen the picture you mention on dpr, but as I'm on my phone at this time I can't do any serious assessment of it's precise exposure. As I recall though, there was mention of the highlights being pushed all the way to the right, and it would surprise me if this was an outright lie.

You are right, in the 5DIV DPR article he says "highlights just short of clipping". Earlier, in the arguments here, he stated "about one stop" or "just under one stop" off clipping. So regarding lying; well you judge.

I calculated from the time after sunset and the exposure triangle info given in the original 5Ds article, that the highlights were about 1.5 stops under clipping. That puts the deep shadows 1.5 stops deeper in very thin (dark) light, which would make a huge difference in the ability to lift in post. On any current camera actually; even on a D810 you would have been losing tonality.

If that is indeed the case I'd call that a lie, as the position of the highlights makes all the difference when trying to prove a point about dynamic range.. But either way, even if things are exaggerated to prove a point, which shouldn't be necessary or practiced by serious reviewers I must add, ,,in general, the fact that canon had a 'DR issue' and it was mentioned in various reviews in itself is a good thing in my opinion.

 

[romanr74]

Some examples of what I am referring for clarity :

http://petapixel.com/2014/11/24/creative-underexposure-nikon-dslrs/

Welcome to this world, Mk IVCanon shooters, I can't wait to see what you start producing.

To my taste, many of these look to "HDR-like"...

 

[romanr74]

I sense plenty of SPS in this forum...

 

[zim]

Some examples of what I am referring for clarity :

http://petapixel.com/2014/11/24/creative-underexposure-nikon-dslrs/

Welcome to this world, Mk IVCanon shooters, I can't wait to see what you start producing.

To my taste, many of these look to "HDR-like"...

Me too

Funny that the article is talking about 2 stop lifts! Welcome to the new world indeed all you 6d 5d3 5ds/r owners ;D

 

[GMCPhotographics]

I mostly agree with you, but what I don't get is why a serious professional reviewer gets whined at for even daring to touch on the fact that Canon's dynamic range has simply been demonstrably inferior for the last couple of years.

Because the information is presented as the Canon sensors not being able to cope with the desired EV range and tonal adjustment to taste, when in fact they can when not crippled by under exposing and putting the dark tones onto the sensor floor. This is exactly what was done with the girl half and hour after sunset shot on the 5Dsr.

Everyone knows by now that the older generation Canon sensors were behind on "DR", but those who come here to demonstrate it produce, without fail, a contrived under-exposed shot and then state this is the best the sensor can do. That's the issue. In fact some of the under-exposed shots shown here as examples of poor "DR" have even the highlights under exposed by about four stops !!

If I have to choose between the 5Ds with its older, off chip ADC, and the newer 5DIV with its on chip ADC the difference in DR potential of the two cameras wouldn't even be considered in my decision. Now of course if we were talking about the difference between 6 and 13 stops it would.

Any shot that uses controllable artificial light in addition to background natural light...which then needs the background to be pulled...shows poor balancing of exposure and not poor DR. It points to a lack of skill, maths and talent. Anyone who fails in this regard should be very careful in claiming to be an expert or authority in this arena. There are amateurs here who get this stuff right in their leisure time. Let alone claiming to be a subject matter expert. DPR's main audience is the general public and with that...a very low expectation of photographic experience. Here at CR, there are some very talented and very experienced photographers who can compare their work with some of the BS statements made by so called "subject matter experts" who happen to have a pre-production copy of a camera and a web site. Bring that kind of Horsehit here and expect to be roasted.

I have rarely needed two or three extra stops of DR, as Sporgon stated, if I need more DR, it's usually a LOT more. The need for expanded DR pales especially when I consider careful use of exposure. Cmos sensors seem to be more critical of exposure than even slide film of yesteryear. I have yet seen any photography genre that clearly shows the need for 4-5 stop pushed shadow areas which couldn't be achieved with either a superior technique (exposure blending / HDR) or better understanding of exposure metering. Sure we all pull a bit of exposure in Post Prod. But if we consider a 4-5 stop pull as necessary then that points to very poor metering or just being lazy and leaving it to be fixed later. If we then blame a camera for not being able to do this so well...that really points to the photographer's lazyness / lack of skill and then passing the blame onto the camera / tool.

Just remember folks that massive DR isn't the be all and end off of photography. Too much DR and silhouettes become impossible.

 

[Jack Douglas]

Recent comments have helped me understand why Richi got himself in trouble and I've learned from all this. So, thanks to those who have been thoughtful in posting comments. It's not easy for most of us to say I blew it - ego gets in the way.

Jack

 

[Deleted member 7828]

Recent comments have helped me understand why Richi got himself in trouble and I've learned from all this. So, thanks to those who have been thoughtful in posting comments. It's not easy for most of us to say I blew it - ego gets in the way.

Jack

+1

I've mostly stayed out of it...after having tested the 5D4, I find there is some extra latitude in post but I really do not see a huge difference in everyday shooting. The scenes that need more DR usually need a LOT more than the 2 stops so many have been clamoring for.

OTOH, there IS a huge difference in detail capture between the "normal" 5D and the 5DSR, and that is apparent in everyday shooting!

 

[IglooEater]

I sense plenty of SPS in this forum...

Standard Positioning System?
Special Services?
SocioPolitical Sciences?
Special Protection System?
Standby Power System?
Single Photon Source?
Solar Powered Sattelite?

I'm curious... No, honestly I don't know, yes I'm that dumb.