Interesting article on DxO about 5D III and D800 resolving power

Started Apr 4, 2013 | Discussions thread
qianp2k Forum Pro • Posts: 10,350
Re: Interesting article on DxO about 5D III and D800 resolving power

schmegg wrote:

qianp2k wrote:

Hans Kruse wrote:

schmegg wrote:

Hans Kruse wrote:

The diffence is quite remarkable and with a full frame 645 sensor the sensor area is around 2.6 times larger so it is to be expected that the resolution is much higher.

The size of a sensor is not the governing attribute that determines its resolving power.

If you photograph the same scene (=same FOV) which is how everybody does this kind of comparisons, then yes, it is!

Just like APS-C sensors resolve less detail than a full frame sensor with the same number of pixels.

Given the exact same detail presented to the sensor plane (which is the only really valid way to determine which sensor can resolve more), an 18MP Canon crop will resolve more detail than a 22MP Canon FF sensor.

Not if you photograph the same scene which is what DxO does and everybody else.

I think you are considering situations where the detail presented to the sensor plane is different to begin with - something that isdue to the difference in sensor size in many situations.

Every review site I know of compares resolution via shooting the same scene (same FOV). You have a different way, but this is unusual and therefore creates more confusion


If you shoot from the same distance with the same focus-length lens, you ended with two different AOV photos, two different scenes, then you compare apple to orange.

What happens is that you end up projecting exactly the same image detail onto each sensor in the comparison (providing you are using the same lens).

they are projected to different AOV as shown in above article.  So what you compare?

Then is very simple to determine which sensor has the better resolving power - you simply look at the detail that has been recorded!

You invented sensor resolution There is no such sensor resolution but only system sensor+lens resolution.

In addition by shooting in the same AOV (or FOV), it just reflects how we use cameras in real world, such as you use 15mm on Canon APS-C while you use 24mm on Canon FF to get the same AOV.

However, when you do this you can make no claims about which sensor resolves more. The best you can claim is that the system (camera and lens), when used in a particular circumstance, can resolve more.

Exactly as all test sites do for a system resolution.  There is simply no sensor resolution.

But I can easily give you a different scenario where the result will be the opposite. Which makes this type of comparison, in terms of determining which sensor is capable of resolving more detail, simply a waste of time.

For example, when shooting a portrait at, say, six feet and using the same composition in terms of AOV, a 5D3 will almost certainly out-resolve a 60D. But, when shooting macro, or BIF, the 60D will almost certainly out-resolve a 5D3.

Certainly in former case, but not necessarily in latter case if you can move closer to birds for example then we can easily see 5D3 outresolves 60D in feather details. As for macro photo, I know what you mean in shooting in min focus distance so still on your point by shooting from the same distance then crop out FF photo to match crop's AOV. But in reality most macro photos are not taken from the min focus distance but from a bit of distance such as in flower case (I know you might describe as close-up photo). Or you can use a longer focus length for macro shots. Therefore I bought Sigma 150/2.8 OS (a bit more expensive) instead of 100L for macro purpose as I need 150mm on FF and I want to be able to shoot from a bit distance.

It's just silly - the resolving power of a sensor does not vary with the shooting situation! The idea is totally ridiculous, and any test for sensor resolving power that comes to this conclusion is flawed.

You're among minority trying to define in your own sensor resolution as simply no such thing.  On your logic (seems more pixels = more resolution), then 41mp Nokia 808 outresolves 36mp D800   I rather to trust DXOMark, Photozone etc creditable sites than you.  And I trust OP in this case.  At least Hans own and experience both and he tested and printed.

What you guys are talking about is the resolving power of the system (camera and lens), not the sensor, and the comparable resolving power of two given systems (as opposed to two given sensors) is a movable feast and may very well vary depending upon the application - such as the example above.

Exactly that's what we are talking about a system resolution.  Sensors or lenses alone don't take photos but only thru sensor+lens system, right?  There is simply no so-called sensor resolution as many factors are involved that depends on what crop format and what lenses used.  Even on the same crop format, a 16mp with a better lens can outresolve 18mp with an inferior lens.

BTW, another thing is that you always downplay DXOMark lens (with system) resolution as sharpness or acutance. However if you read DXOMark lens test carefully they are talking resolution from the lens on sensors. DXOMark P-MPix tests are based on MTF resolution but translate in an easier understanding expressing way. Nevertheless I have not noticed P-MPix has changed any comparison results when it used MTF resolution then. P-MPix is resolution that human eyes can see. 41mp Nokia 808 seems has lots of paper resolution but it simply cannot deliver the real world 41mp resolution human eyes can resolve.

-- hide signature --
Post (hide subjects) Posted by
MOD Mako2011
MOD Mako2011
MOD Mako2011
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow