"The Nikon D3x offers the finest image quality in a DSLR the world has yet seen"

[Ray]

Bill, using ETTR is fine by me, and I use it frequently. However, Ettr at an ISO above your sensor sweetspot does not gain anything. Even if you meter for ETTR, you're still better off holding those shutter and fstop settings and then cranking down the ISO that happens to be your camera's sweetspot, and that would be ISO 100 on the A900. After that, boost exposure in raw with highlight compression.    What raw converter are you using?


Ps. This all about eeking out best results. Shooting regularly, the A900 is competitive through ISO 800 with the other makes.

I'm not sure that 'sweet spot' is the correct term here, Doug. Most cameras have a sweet spot at base ISO. That is, the camera can perform at its full potential at base ISO with regard to DR, S/N and resolution, given proper exposure. The A900 seems to be unusual in the respect that best performance is at one of the expanded ISO settings, ie. ISO 100. Base ISO is 200, is it not?

Basically, I would characterise the A900 as a 'one ISO' camera. It seems that the only purpose served by using a higher ISO than 100 (which is actually ISO 119 according to DXOmark), is to brighten the image review on the camera's LCD screen, and/or allow the camera to produce decent jpegs with its internal processing. For those who shoot RAW, it would seem best practice to always have the camera set on ISO 100 for all shooting conditions.

There is the concept of the 'false' ISO setting. For Canon cameras such as the 5D, there is little point in using ISO 3200, if you shoot RAW. DR and shadow noise are no better than they would be using the same exposure at ISO 1600, which is effectively a one stop underexposure. In fact it can be disadvantageous to use the false ISOs and attempt to get a full ETTR at such ISOs, because there's always a danger of miscalculating exposure and blowing highlights. Using the same exposure at a lower ISO greatly reduces the risk of blown highlights whilst presenting no disadvantages.

I guess one could claim that all ISOs on the A900 above ISO 100 are false ISOs. They serve no image quality purpose. This is not the case with Canon cameras and, it seems clear, is not the case with the D3X.

[douglasf13]

I would have agreed with mostly everything you just said a while back, but after jumpin down the rabbit hole with Iliah Borg and others involved in raw conversion, I'm beginning to understand choosing the right raw converter, exposing for that converter, the nuances in the ways these cameras are built...and the nuance and dependability of reviews/tests, and that includes DXO mark.

  When I say sweet spot, I'm referring to the point where it becomes more advantageous to boost in raw devoloper rather than boost ISO in camera. This rarely correlates to extended camera ISOs. With the A900, that number is between ISO 100-400, depending on the converter used (maybe higher with Adobe, as it's about the worst out there.) That number goes a stop higher to ISO 800 with the D3x. I don't know what that number is on your Canons, but I can almost guarantee it's not what you think it is, given what you've been saying.

Saying that the A900 is a single ISO camera is absurd, unless you shoot in the way I mentioned above, in which all cameras are essentially single iso cameras. If using in camera ISOs, the A900 does fine in the noise department compared to the competition until iso1600, where it starts losing noticeable ground. Of course, it also has distinct advantages over other cameras at ISOs lower than 1600.   The issue that some are having with the 5dii is that's a jack of all trades, master of none. If shooting in the lower ISOs, the D3d and A900 reign supreme, with each having advantages over the other depending on the scene. At the higher ISOs, the D3/700 is still king.

[Ray]

Saying that the A900 is a single ISO camera is absurd, unless you shoot in the way I mentioned above, in which all cameras are essentially single iso cameras.

No, I don't think you've grasped the point. It is never better on a Canon to underexpose at a lower ISO instead of using the same exposure as an ETTR at a higher ISO, unless one uses the expanded ISO range, which is ISO 3200 on the 5D, 40D, 1Ds2, 1Ds3; ISO 6400 on the 50D and ISO 12,800 on the 5D2.

Maybe I've misunderstood what you are saying which is why I asked for clarification in a previous post, so I'll ask again. If you meter at ISO 800 and get a reading of 1/100th sec at F8, which you estimate would give you a full ETTR exposure maximising DR and S/N at that specific ISO of 800, you are saying that you will get better results by using that same exposure of 1/100th sec and F8 at ISO 100, and compensating for the underexposure of 3 stops in the RAW converter. Right?

Now I know that different RAW converters can produce slightly different results, depending on the camera. I used to prefer RSP before Adobe bought them out. DPP seems to produce a marginally finer noise pattern than ACR with Canon RAW files, but I consider such differences to be of the nature of extreme pixel-peeping. But sometimes they may not be, so let me ask you a few more questions.

(1) Are you claiming that your RAW converter of choice for A900 images, does a better job than ACR in compensating for underexposure across the whole range of ISO's, or are you claiming that it mainly does a better job with the A900 only at ISO 100?

(2) Are you claiming that ACR conversions of A900 images underexposed by 3 stops at ISO 100 are the same, better or worse than ACR conversions at ISO 800, using the same exposure for a full ETR at ISO 800?

(3) Are you claiming that full exposures at ISO 1600 on the A900 are the same, better or worse than 4-stop underexposures at ISO 100, using your favourite converter?

[jjj]

As the pixel peepers seems to be clustered in this thread like a decay of zombies around a still breathing human.
Try this quiz to see how good you really are.  
Spot the gifference

[douglasf13]

I grasped the point. I just believe you're wrong.  

 Really, outside of highlight recovery, ACR is poor for all cameras, comparitively.  Making IQ judgements and using ACR don't mix well, and it happens to be particularly bad for the A900.  I agree that this is all pixel peeping to a point, but so is comparing the high ISO noise of all of these cameras, so we should be consistant.  

 Yes, bumping the ISO down to around 100-400 and boosting in raw (not ACR) is better than shooting ISO 1600 with the A900, but that is not unique to Sony.

[douglasf13]

jjj, I agree that this is all pixel peeper mumbo jumbo, but the title of this thread evokes exactly that. If you're not printing large, there is little difference between any of these cameras.

[dwdallam]

I'm always happy to be proved wrong. That's how I learn. I do a few tests with the material available. I use my common sense, experience and what little knowledge I possess to make certain deductions and draw certain conclusions.

If I've overlooked something, or my testing material was flawed in some way, or perhaps I didn't use the best converter available for a particular camera, then I'm happy to have such flaws in my methodology mentioned. I've got no barrow to push here except getting at the facts.

What I don't go along with are claims that because so and so are very experienced photographers and actually own the equipment under discussion, they must always be right on all issues.

Show me the images. Describe the methodology.

I think Lepig's point is that all things being equal, the D3X out resolves the D3 and I can understand that because the D3X is a superior resolution camera. That's pretty common sense. And when I say all things being equal, I mean that the Nikon sensor is at least as good at what it does as the Canon sensor. There is yet another way to put this:  What "sensor" is more efficient at what it does? The Nikon sensor would have to be "better" if it were to have, say, lower noise, since the same size sensor has more MPs, and thus more possible noise. If the Nikon has the same noise as the DS3, then the Nikon sensor (or other electronics) are doing a better job at canceling noise and keeping detail. Lepig is a scientist in this respect. What he is talking about are those variables that can be measured by testing equipment where the eyes fail.

The next question is something like: All things being equal, such as in a lab environment, which camera can produce better prints when viewed by the naked eye? And that question would be of no concern for some application that is screen based and seen at 100%. I don't know of any of those, except maybe some sort of scientific application, or maybe forensics. Again, Lepig's point, I think, is that using specific test equipment and parameters that are measurable, even if you can't see them with the naked eye, Nikon wins. Likewise, if you scan in a perfect negative from a medium format camera and scan it with a film grain resolution, the film MF camera will out resolve a 50MP camera because at that scan resolution, I think you get around a 128MB file.

My point here is that to argue scientific tests with your eyeball is missing the point of the scientific test. I guess a really simple analogy would be something like this: A 1MT Nuclear bomb releases less power than a 1.0000000000000000000000000000000009MT bomb, and that is surely true as true can be. Does it make a difference when bombing? Nope. And this these are the two arguments we are confusing here, I think.

Why do I get sucked into these types as analysis? It must be a guy thing.

[Ray]

As the pixel peepers seems to be clustered in this thread like a decay of zombies around a still breathing human.
Try this quiz to see how good you really are.  
Spot the gifference

jjj,
Why are you seeking the company of a cluster of decaying zombies? Is it because of your fascination with cats? These creatures cause great havoc in our pristine Australian environment (the cats, not the zombies).

[dwdallam]

"If someone tells you D3x does not record any more details than the D3 at 1600 ISO, what it tells is only that you were listening to an incompetent tester, even he or she actually has the means to own the equipment."

Calling others incompetent is rude in my book.



Actually it is only one variable, and the shot I showed was done with a tripod.  Another variable that you completely ignore is spectral properties differences between the two cameras and what happens when those are compensated for by conversion; then you impugn the competence of those that are aware of these issues because you saw a comparison made under studio lighting.



Galen Rowell was a hand hold person, he couldn't hold on to 55 lbs while hanging from a rock face and taking the shot you missed because you stayed on the ground.  Thom Hogan is his most notable protege; and he points out that the farther you get from the main road, the fewer big rigs you see.  You think pretty highly of yourself when you refer to such people as not being "capable landscape photographers".

It's not rude to call someone incompetent if it is true. The word just means not up to the task.
INCOMPETENT:
1: not legally qualified 2: inadequate to or unsuitable for a particular purpose 3 a: lacking the qualities needed for effective action
What is so rude about that?

[Ray]

I grasped the point. I just believe you're wrong.

You reckon my eyes are deceiving me and I should see a doctor?  

I've addressed this issue a number of times over the years and have made several test comparisons to confirm what others have also claimed. The early Canon cameras such as the D60 offered only a very marginal improvement in DR and S/N when using a specific exposure at a higher ISO instead of a lower ISO. This situation changed dramatically with the 20D.

Nevertheless, there's still not much difference between ISOs of one stop difference. An ETTR at ISO 1600 is still only marginally cleaner than the same exposure at ISO 800, which in turn is only marginally cleaner than ISO 400. But compare a shot at ISO 1600 with a shot using the same settings at ISO 100, and the difference is like chalk and cheese.

Yes, bumping the ISO down to around 100-400 and boosting in raw (not ACR) is better than shooting ISO 1600 with the A900, but that is not unique to Sony.

You haven't answered all my questions regarding this point. Are you saying that ACR produces a better image with the A900 at ISO 1600, than it does at ISO 100 when underexposed 4 stops, but that your favourite converter produces a better image at ISO 100 than it does at ISO 1600? Or, are you saying that your favourite converter produces better images than ACR at both ISO 100 and ISO 1600, in these circumstance where actual exposure is the same at both ISOs?

[jjj]

jjj,
Why are you seeking the company of a cluster of decaying zombies?

No, it's a 'decay of zombies' not a cluster. Girlfriend asked me the other day what the collective noun for a gathering of zombies was, not sure why, maybe trying to describe her workmates?    And a 'decay' seemed apposite.

Is it because of your fascination with cats?

I didn't design the kitten quiz, I just saw it after reading this thread.

These creatures cause great havoc in our pristine Australian environment (the cats, not the zombies).

So you've tamed the zombies but cannot control kittens!?! Weird country.

[douglasf13]

You reckon my eyes are deceiving me and I should see a doctor?  

I've addressed this issue a number of times over the years and have made several test comparisons to confirm what others have also claimed. The early Canon cameras such as the D60 offered only a very marginal improvement in DR and S/N when using a specific exposure at a higher ISO instead of a lower ISO. This situation changed dramatically with the 20D.

Nevertheless, there's still not much difference between ISOs of one stop difference. An ETTR at ISO 1600 is still only marginally cleaner than the same exposure at ISO 800, which in turn is only marginally cleaner than ISO 400. But compare a shot at ISO 1600 with a shot using the same settings at ISO 100, and the difference is like chalk and cheese.



You haven't answered all my questions regarding this point. Are you saying that ACR produces a better image with the A900 at ISO 1600, than it does at ISO 100 when underexposed 4 stops, but that your favourite converter produces a better image at ISO 100 than it does at ISO 1600? Or, are you saying that your favourite converter produces better images than ACR at both ISO 100 and ISO 1600, in these circumstance where actual exposure is the same at both ISOs?

Sorry, Ray. I'm trying to answer your questions, but Im posting on a cell phone and I can't review your questions as I'm typing. These long posts are tedious  BTW, if there is a shortness to my posts, please don't take it as being rude. It's just because of my input device. Freakin technology LOL.

Anyways, with ACR or Sony's IDC software, I shoot normally, because there does not seem to be any real advantage to underexposure with boosting in RAW.  A RAW converter with more advanced demoisacing, like RPP, does show me that underexposing and then boosting in RAW does show advantages. So I believe that the answer to both your questions is yes. I believe that if you dump ACR and DPP, you may form new opinions about this with your Canons.  

Basically, all of the fullframe cameras are very good now, and each has it's strong suits. The problem arrises with arm chair pixel peeping, as it's what's become popular for easy to understand, short reviews. However, if making judgements between cameras involves using ACR or similar, those judgements should be taken with a grain of salt.    
 

 

[Bill Caulfeild-Browne]

Bill, using ETTR is fine by me, and I use it frequently. However, Ettr at an ISO above your sensor sweetspot does not gain anything. Even if you meter for ETTR, you're still better off holding those shutter and fstop settings and then cranking down the ISO that happens to be your camera's sweetspot, and that would be ISO 100 on the A900. After that, boost exposure in raw with highlight compression.    What raw converter are you using?


Ps. This all about eeking out best results. Shooting regularly, the A900 is competitive through ISO 800 with the other makes.

I use Capture One for RAW conversion then LR2 for final adjustments, unless I need PS, which is very rare now. Aperture actually does a bit better than C1 at high ISOs but I rarely use it because my final product is a print - and even with 16 by 20s, I just cannot see the difference. By 24 by 36, it certainly becomes visible but I don't routinely print to that size from my a900 - I'd normally be using MFDB for those pix.

Pixel peeping is an interesting "academic" exercise but really it's the end product that should be critiqued, not 100% images on a screen.

But hey, it's all fun!

Bill

[douglasf13]

Good points, Bill. I've currently settled on C1 as a compromise between workflow an IQ, but I've been reading a lot about flat RAW conversion with PS corrections, and that is apparently the top of the IQ mountain so far.  However, like you said, at 16x20 the noise differences between any of these cameras and converters is minimal. Hilariously, I tend to add noise to a lot of my low ISO stuff. Go figure

[douglasf13]

Whoops, double post.

[Ray]

So you've tamed the zombies but cannot control kittens!?! Weird country.

Cannot is not the operative word. Don't want to, or can't be bothered is more like it. It seems unnatural to me to keep a cat locked up in the house 24 hours a day, or keep a close watch on it when it's let out into the garden, in case it starts playing with a near-extinct species of lizard. Animals should be allowed a certain degree of freedom. Do you want to spend your life as an animal baby-sitter?

However, cats in Australia present a different problem because they are not indigenous, just as rabbits and Cane Toads are also not indigenous and also create problems as a consequence.

[Ray]

Basically, all of the fullframe cameras are very good now, and each has it's strong suits. The problem arrises with arm chair pixel peeping, as it's what's become popular for easy to understand, short reviews. However, if making judgements between cameras involves using ACR or similar, those judgements should be taken with a grain of salt.

Doug,
I know that they are all very good, on balance, and that each camera has its strong points and weak points. The purpose of pixel-peeping is not necessarily an end in itself, as an armchair exercise. It might be for some, but it isn't for me. For me it's a way of finding out the true advantages of a piece of equipment so I can justify buying it. Most people simply haven't got the time and money to buy or hire each piece of equipment that they might be interested in, and test it for themselves in the field. And, of course, new equipment is often not available for hire. I once tried to hire a D3 in Australia with Nikkor 14-24/2.8 lens. Neither were available at the time and the lens was never going to be available for hire because its bulbous front element prevented the attachment of a protective UV filter.

The problem is, essentially, that all sorts of claims are made about equipment that may be broadly true but need to be quantified, otherwise one could waste a lot of money buying certain equipment with the understanding that there is an advantage in some important respect, DR, noise, resolution, whatever, and then find out later that in practice the advantage is insignificant. In order to quantify such differences, someone has to engage in pixel-peeping and the results ideally should be accompanied with images demonstrating such difference.

The possibility that RPP might do a better conversion job than ACR is interesting, but again, some of us want to know how much better. It would have to be very much better for me to use that program because it seems to be restricted to the Mac operating system and I'm a PC guy.

[Tony Beach]

It's not rude to call someone incompetent if it is true. The word just means not up to the task.
INCOMPETENT:
1: not legally qualified 2: inadequate to or unsuitable for a particular purpose 3 a: lacking the qualities needed for effective action
What is so rude about that?

Okay, I think you're incompetent too -- no offense intended.

[Ray]

No, it's a 'decay of zombies' not a cluster. Girlfriend asked me the other day what the collective noun for a gathering of zombies was, not sure why, maybe trying to describe her workmates?    And a 'decay' seemed apposite.

Can't agree. Decay is more appropriate for corpses. A decay of corpses. Sound right, doesn't it? Zombies are still living, or at least animated, aren't they? Incidentally, Wikipedia uses the collective noun of 'horde' and 'plague' for zombies. For cats it's a clutter.  

[douglasf13]

Ray, nearly any raw converter...rpp, raw therepee, rawmagick, bible, capture one, etc is better than ACR, outside of highlight recovery. I'm using C1 currently.  ACR is still fine for most, but if you care about the slight nuances between these cameras, using acr is contradictory.