Sony A7RII Review - Dynamic Range...

[dreed]

I know I should probably watch the video first (and please spank me if that addresses this but...)


now that Sony has announced a lossless raw format for this camera is to be forthcoming, will this need to be redone? Or is the feeling that while it will help deliver better quality images (in various cases), it won't impact DR?

[Kevin Raber]

Yes, maybe it will have to be updated.  Most likely though if Sony does what they say they will with the update the results would only be better.  We'll wait and see.  When this project was started by Michael the release of uncompressed RAWS was not known.

[Josh-H]

Is DR all people care about these days? What happened to caring more about 'Good photography'?

I must throw out all my images.. all shot with sensors with poor DR.. oh well.. better go buy a Sony so I can make a good photograph *shakes head in utter amazement*

[Telecaster]

Is DR all people care about these days? What happened to caring more about 'Good photography'?

It's among the few remaining aspects of photo tech where you can geek out on specs while claiming this means something when applied to actually taking photos. Thus the intensified attention paid to it. No doubt when all cameras have 16+ usable stops of DR we'll come up with something else to avoid confronting our own limitations as photographers. 

-Dave-

[Tony Jay]

Is DR all people care about these days? What happened to caring more about 'Good photography'?

I must throw out all my images.. all shot with sensors with poor DR.. oh well.. better go buy a Sony so I can make a good photograph *shakes head in utter amazement*

As an owner of both the A7R and the mark II version DR is not the only thing I care about.
My equipment is currently both Canon and Sony.
Although I own some Sony FE lenses (and they are very good) I still mostly use my Canon glass (also top-notch).
The bodies I use are dependent on what I am shooting - 5D mark III for wildlife/birds/action and the Sony bodies in preference for most of the rest.

I have a saying that almost everyone who has tried to use a camera to make meaningful images can probably relate to: is that it is the easiest thing in the world to shoot rubbish - shooting well is a little harder.
In this regard the equipment is almost irrelevant.
However, on some occasions where the images produced might justify the equipment used a greater DR could have been useful.
DR, and any other sensor characteristic, can never save a rubbish image. A sow's ear ever remains a sow's ear, DR never magically converts it into a silk purse.

Like you Josh I shoot almost exclusively outdoors and the gain in DR from the 5D mark III to the Sony A7R series cameras has made a difference in the images obtainable and the quality obtained.
Hopefully I know a little about exposure and getting the best out a sensor.
Not knowing how to expose well and lever the advantages of whatever sensor I am using in whatever camera just makes the images I want unobtainable. The equipment used then becomes completely irrelevant.

For what it is worth I am comfortable still shooting with both Canon and Sony bodies.
The 5D mark III still provides me with capabilities that the Sony bodies cannot match especially when wielding super telephoto lenses.

Tony Jay

[amolitor]

I have to say, I use AIS lenses on a D3100, which doesn't couple at all with them. So all exposures are by guess and glancing at the back of the display.

It is a very rare circumstance when even my initial shot is 5 stops off, and generally that only happens when I last used the camera under radically different light, and have time to burn, so I just blast off a test shot as the fastest way to tell where I am at. If I am more pressed for time, I dial in a guess, which is rarely more than 2 stops off.

So this notion that one might need to push as much as 4-5 stops is.. odd to me.

[Alan Klein]

I agree.  The last section that compares which of the 4 look best at 6400 is really silly.  Even the best looks awful. What does this really prove? 

[michael]

Humm....

What's with the negativism folks? Is this test really so hard to understand?

It's like testing which car goes round the track faster than another. This can tell you something about how the car handles at the extremes, including acceleration, top speed, braking, handling, etc, etc.

This isn't saying that you should drive on the roads at reckless speeds, and certainly not that you have to. Just, that when a device is pushed to its limits, which one fails first.  As a consequence, we learn something about how well the car (camera) does overall when pushed.

Testing something under ordinary circumstances....driving to the grocery store in traffic tells us nothing, just as pretty pictures at normal ISO isn't terribly instructive.

So, the comments that this isn't about real world photography, are simply gratuitous. Of course this isn't what we do in the real world, any more than is pushing a car around the track at top speed. But we learn something because of the stress circumstances, don't we?

Also, I find comments such as "the easiest thing in the world to shoot rubbish - shooting well is a little harder.." to be an unhelpful non-sequitar under the circumstances.

Michael

[haplo602]

Humm....

What's with the negativism folks? Is this test really so hard to understand?

It's like testing which car goes round the track faster than another. This can tell you something about how the car handles at the extremes, including acceleration, top speed, braking, handling, etc, etc.

This isn't saying that you should drive on the roads at reckless speeds, and certainly not that you have to. Just, that when a device is pushed to its limits, which one fails first.  As a consequence, we learn something about how well the car (camera) does overall when pushed.

Testing something under ordinary circumstances....driving to the grocery store in traffic tells us nothing, just as pretty pictures at normal ISO isn't terribly instructive.

So, the comments that this isn't about real world photography, are simply gratuitous. Of course this isn't what we do in the real world, any more than is pushing a car around the track at top speed. But we learn something because of the stress circumstances, don't we?

Also, I find comments such as "the easiest thing in the world to shoot rubbish - shooting well is a little harder.." to be an unhelpful non-sequitar under the circumstances.

Michael

hmm ... the test is more like testing straight line acceleration. all the other parts (braking, handling etc.) are not tested :-) Most notably I am missing highlight clipping test for red and blue channels. green only helps recovering the luminance part of highlight information but you need all 3 channels to recover color. I hope this is tested in the next videos in the series.

anyway the test is extremely interesting. I was very suprised how the 1DX handles things compared to D810/A7 ...

[pegelli]

I think it's an interesting, useful but non-scientific test, just like in the old days you needed to know developer strength and times to get the best of your negatives I think it's useful in the digital days to know how far you can push your sensor without too much compromise on image quality. No need to be cynical or derogatory about it, the technical aspect has always been part of photography and we all know that mastering the technique alone doesn't produce worthwhile images, nothing new there either. 

[marcmccalmont]

I'd like to applaud Michael Tapes for taking the time and effort to do the testing and make a video for all of us to benefit from.
I would like to see the maximum capabilities of the cameras tested like iso 64 on the D810 (min and max iso for each camera)
These tests are extremely difficult at best because the scientific method is flawed to begin with(I forget the name of the chemist who proved this statistically I think he worked for Dow?) ....lets set up a test, hold all variables constant except one, change the one variable and draw a conclusion.
To test properly one needs to optimize the 2 interfaces to the changed variable, run the tests then draw a conclusion. An example, I'm testing the best tire of 2015 and pick a Corvette, stock wheels, mount the tires and run the tests. I then draw the conclusion XYZ is the best or ABC has the best dry braking etc. A proper test would optimize both interfaces first, change the width and diameter of the wheel to optimum(interface of the tire to the wheel) and optimize the tire pressure(the interface of the tire to the pavement). When this method is used the conclusions are accurate but it takes exponentially more time and tests.
Bottom line these tests are useful but not 100% accurate like optimizing the D810's dynamic range by using iso 64. The point is one variable is not changing rater 2 interfaces are, these need to be optimized. But one can conclude for overall performance the A7rII is the winner and for landscapes on a tripod the D810 would be my choice. Can't wait to see what Nikon does with the next Gen Sony sensor in a D8xx!
Marc

[Tony Jay]

Humm....
...
Also, I find comments such as "the easiest thing in the world to shoot rubbish - shooting well is a little harder.." to be an unhelpful non-sequitar under the circumstances.

Michael

Michael, I make the connection in the next sentence - shooting well or shooting badly is almost always completely independent of the equipment used.
Josh's comments are little over the top - a lot of hyperbole there.
He actually produces excellent images with whatever he uses (Canon equipment I think) and clearly not Sony mirrorless.
However, even Josh must (should) agree, that even if it is only occasionally, a bit more DR can help.

Tony Jay

[jeremyrh]

Is DR all people care about these days? What happened to caring more about 'Good photography'?

I must throw out all my images.. all shot with sensors with poor DR.. oh well.. better go buy a Sony so I can make a good photograph *shakes head in utter amazement*

No, it's not all people care about (and there's no implication that good photography is dependent on having the ultimate in DR), but it's a good thing to discuss in a test entitled "Sony A7RII Review - Dynamic Range"

[Josh-H]

However, even Josh must (should) agree, that even if it is only occasionally, a bit more DR can help.

Thats why we have Grad ND filters 

Actually in all seriousness.. more DR can really help in certain limited situations. However, the fascination with it (speaking generally for what is rampant on the web) as the 'be all and end' I find over the top.

Measuring DR seems to be the trendy thing to do these days.. I for one would rather see more general reviews about how these machines actually perform in the field in real world use. With thoughts on battery life, ergonomics, general usability, weather sealing etc. In other words, look at the full gamut of what makes a camera great; rather than distilling it down to the 'DR' argument.

[ErikKaffehr]

Hi Michael,

Absolutely agree. Specific tests are often the best way to find out special aspects.

Just to say, web size images without processing information can be nice, but they don't yield any information...

Best regards
Erik

Humm....

What's with the negativism folks? Is this test really so hard to understand?

It's like testing which car goes round the track faster than another. This can tell you something about how the car handles at the extremes, including acceleration, top speed, braking, handling, etc, etc.

This isn't saying that you should drive on the roads at reckless speeds, and certainly not that you have to. Just, that when a device is pushed to its limits, which one fails first.  As a consequence, we learn something about how well the car (camera) does overall when pushed.

Testing something under ordinary circumstances....driving to the grocery store in traffic tells us nothing, just as pretty pictures at normal ISO isn't terribly instructive.

So, the comments that this isn't about real world photography, are simply gratuitous. Of course this isn't what we do in the real world, any more than is pushing a car around the track at top speed. But we learn something because of the stress circumstances, don't we?

Also, I find comments such as "the easiest thing in the world to shoot rubbish - shooting well is a little harder.." to be an unhelpful non-sequitar under the circumstances.

Michael

[marcmccalmont]

Measuring DR seems to be the trendy thing to do these days.. I for one would rather see more general reviews about how these machines actually perform in the field in real world use. With thoughts on battery life, ergonomics, general usability, weather sealing etc. In other words, look at the full gamut of what makes a camera great; rather than distilling it down to the 'DR' argument.

I think DR was very important when digital cameras or slide film did not have the dynamic range (static) of human sight. Now that current sensors accomplish this you can capture and print a scene that represents what you saw. Now its how much highlight recovery or shadow recovery or sloppy exposure you can tolerate. But you are buying "a refrigerator full of the same film for 3 years" so why not buy the best film for your purpose? I think we have some great resources to evaluate a new purchase, LULA, DPReview, DxOmark, Photozone reading all will give you a balanced review.
Marc

[pegelli]

Measuring DR seems to be the trendy thing to do these days.. I for one would rather see more general reviews about how these machines actually perform in the field in real world use. With thoughts on battery life, ergonomics, general usability, weather sealing etc. In other words, look at the full gamut of what makes a camera great; rather than distilling it down to the 'DR' argument.

I'm sorry Josh, I don't agree with you. I think the trendy thing on the internet (and the reactions to this test are no exception) is to complain about what it is not rather then appreciate the information it provides and the effort that went into it.

[Tony Jay]

Thats why we have Grad ND filters 

Actually in all seriousness.. more DR can really help in certain limited situations. However, the fascination with it (speaking generally for what is rampant on the web) as the 'be all and end' I find over the top.

Measuring DR seems to be the trendy thing to do these days.. I for one would rather see more general reviews about how these machines actually perform in the field in real world use. With thoughts on battery life, ergonomics, general usability, weather sealing etc. In other words, look at the full gamut of what makes a camera great; rather than distilling it down to the 'DR' argument.

I agree that no camera can really afford to be a one-trick-pony - I don't think that the new Sony mirrorless options are that.
Nonetheless they do have limitations.
Individual battery life is not great.
I also don't think that the weather sealing is quite up to the standards of the high-end Canons but I intend to use my Sony's in all sorts of weather conditions - thats where the fun images are - but I am not sure how they would stand up to one of your expeditions Josh.
I mentioned in another thread that sensor dust is a major pain with mirrorless cameras however one gets used to regularly cleaning sensors using both blowers and the built-in cleaning mechanisms. Sometimes wet cleaning is also required.
I don't think that, even with the new Metabones firmware updates, that I could use the A7R mark II mated with Canon super telephoto lenses for bird and wildlife.
Also I am not sure that Sony's repair services are necessarily up to that of Canon Professional Services. However time will tell.

Tony Jay

[Bart_van_der_Wolf]

No, it's not all people care about (and there's no implication that good photography is dependent on having the ultimate in DR), but it's a good thing to discuss in a test entitled "Sony A7RII Review - Dynamic Range"

+1

The Part 1 video focused on Dynamic Range, because it is useful to have.

Cheers,
Bart

[Hans Kruse]

Thats why we have Grad ND filters 

Actually in all seriousness.. more DR can really help in certain limited situations. However, the fascination with it (speaking generally for what is rampant on the web) as the 'be all and end' I find over the top.

Measuring DR seems to be the trendy thing to do these days.. I for one would rather see more general reviews about how these machines actually perform in the field in real world use. With thoughts on battery life, ergonomics, general usability, weather sealing etc. In other words, look at the full gamut of what makes a camera great; rather than distilling it down to the 'DR' argument.

+1.

DR is good to have and never be a bad thing to have, but I agree there are many other things that are important for a good tool for the picture taking. I have been shooting now for a couple of weeks on Isle of Skye in Scotland with the Canon 5Ds R and the camera really delivers. DR (as we know) could be better but more extreme DR situations would need blending anyway on my D810 or ND grad filters (which I don't use). Blending in Lightroom CC is amazing and so far I have not found any faults for the type of pictures I take. With that the lack of DR for landscapes has become a much lesser issue in my opinion.