Comparing resolution

Lens and Camera reviews by our members...

Moderator: Stany Buyle

Post Reply
Iliah Borg
AFFILIATED MEMBER
Posts: 20
Joined: Tue 16 Feb 2010 17:53

Comparing resolution

Post by Iliah Borg »

With the introduction of high megapixel count sensors resolution and noise are two of most hot topics. There is no single number that can quantify noise or resolution in a way that is meaningful for a practical photographer, but there are couple of very useful approaches that result in meaningful comparisons between two different cameras.

To start with, noise affects resolution, that is resolution is different in midtones and shadows. But let's pretend for a moment that we are trying to compare best possible resolutions from two (or more) cameras (one might want to extend this test to comparing resolution for different ISO sensitivity settings of the same camera and between different cameras). In this case, we need to overcome camera shake. Main sources of the shake are handshake, mirror slap, shutter slap, aperture slap, floor vibration, and wind. Floor vibration and wind affect both cameras and resolution targets. Shutter slap is one of most overlooked factors while acceleration of the shutter is reaching 7g. Only a very sturdy camera mount (I'm not saying "tripod" here, and that's on purpose - grip in a vice works better than most of tripods) can help dumping it. The mount being the same and tightened for all the tests this factor affects cameras under the test in a pretty uniform way and does not skew the test results too much. Of course remote control, MLU, aperture preset, VR/IS off, careful manual focusing etc. are necessary prerequisites, especially when comparing cameras with different sensor sizes. To overcome shutter slap the trick is to use bulb shutter setting and expose using an external flash. Given all the factors, in home basement environment (check that your air conditioning is off) it is hard to get resolution numbers with an error of less than 10%.

With current advance of digital there is no need to shoot full resolution targets anymore - a black slanted edge positioned over a white field is enough for analysis.

To obtain graphs and numbers characterizing resolution you can use measuring SFR through ISO 12233 slanted edge methodology. You can read about it first: search for SFR MTF Peter D. Burns (Kodak), and do not miss http://oatao.univ-toulouse.fr/303/1/estribeau_303.pdf

Free implementations are here:
http://www.imagescienceassociates.com/m ... ode=ISA001 (User Guide is a valuable reading)
http://rsbweb.nih.gov/ij/plugins/se-mtf/index.html
DirkVermeirre
moderator
Posts: 50
Joined: Fri 29 Aug 2008 01:02

Re: Comparing resolution

Post by DirkVermeirre »

Hi Iliah,

The camera that had the best tonality (especially in black and white) for me was the Fuji S5. But only with 100 and 200 Iso (I didn't like the higher Iso results, the previous Fuji models had better quality in that field then the S5). I agree that noise affect resolution. But for very big normal prints I think that adding fine grain (from a drum scan from a large format negative from a photographed light gray area) results in (for our eyes) bettere resolution, more sharpness and contrast (I think that this is called microcontrast). And yes floor vibration is often underestimated. I remember from a friend with a large studio with large format camera's that his images were ruined when a tram drove in the front of his studio (almost 50 meters away from the conctrete floor in the studio)!

With kind regards,

Dirk
Iliah Borg
AFFILIATED MEMBER
Posts: 20
Joined: Tue 16 Feb 2010 17:53

Re: Comparing resolution

Post by Iliah Borg »

Dear Dirk,

Yes, I was shooting sandpaper to use it as an intensifier of apparent details. Now I just scan it.
pam.meier
moderator
Posts: 421
Joined: Thu 28 Aug 2008 18:54

Re: Comparing resolution

Post by pam.meier »

I was shocked to find that high megapixel cameras are much more prone to camera shake when using studio strobes. I keep ruining a few shots in a session with the A900 while all my D3 images don´t show any blur.
If I can I prefer using a tripod, but with people photography that is very awkard.
Iliah Borg
AFFILIATED MEMBER
Posts: 20
Joined: Tue 16 Feb 2010 17:53

Re: Comparing resolution

Post by Iliah Borg »

Contrary to conventional practice I do not use studio strobes, they are slow and it is close to impossible to control exposure time using those. For shots where the colour reproduction is not a paramount HMI light serve well, for the shots where it is important to get extremely accurate colour I use SoLux bulbs.
pam.meier
moderator
Posts: 421
Joined: Thu 28 Aug 2008 18:54

Re: Comparing resolution

Post by pam.meier »

So what shutter speeds can you use with those lights?

Oh, by the way, Iliah, my D3 came back from having its aa-filter removed a few days ago.
User avatar
Stany Buyle
moderator
Posts: 2890
Joined: Sun 30 Nov 2008 15:05
Active Member?: Yes
Post pictures?: Yes
Contact:

Re: Comparing resolution

Post by Stany Buyle »

Iliah Borg wrote:With the introduction of high megapixel count sensors resolution and noise are two of most hot topics.
...
In this case, we need to overcome camera shake. Main sources of the shake are handshake, mirror slap, shutter slap, aperture slap, floor vibration, and wind. Floor vibration and wind affect both cameras and resolution targets. Shutter slap is one of most overlooked factors while acceleration of the shutter is reaching 7g. Only a very sturdy camera mount (I'm not saying "tripod" here, and that's on purpose - grip in a vice works better than most of tripods) can help dumping it. The mount being the same and tightened for all the tests this factor affects cameras under the test in a pretty uniform way and does not skew the test results too much. Of course remote control, MLU, aperture preset, VR/IS off, careful manual focusing etc. are necessary prerequisites, especially when comparing cameras with different sensor sizes. To overcome shutter slap the trick is to use bulb shutter setting and expose using an external flash. Given all the factors, in home basement environment (check that your air conditioning is off) it is hard to get resolution numbers with an error of less than 10%.
Good morning Iliah,
In relation to your topic, was an interesting article in the latest number of Chasseur'd'Imagesthere (more like a paragraph in the D7000 review) insinuating that the obtained resolution in D7000 is negatively influenced by the little, lightweight body which cannot withstand the mirror slap and vibration like big bodies do.
Kindest regards,
Stany
kindest regards,

Stany
I like better one good shot in a day than 10 bad ones in a second...
http://www.fotografie.cafe
Steve Bingham
moderator
Posts: 91
Joined: Tue 5 Oct 2010 18:06
Active Member?: Yes
Post pictures?: Yes
Location: Payson, Arizona, USA

Re: Comparing resolution

Post by Steve Bingham »

I remember seeing some vibration graphs showing actual vertical oscillation from a pin point light source and a timing chart underneath the graph. I used to shoot with a very big and heavy Pentax 6x7 - and heavy glass (later renamed Pentax 67) which was known for its mirror slap vibration - in spite of valiant efforts by Pentax to dampen it.

This kind of puts the bigger, heavier camera equal theory (= less shutter slap vibration) in question. More to the point would be the weight of the mirror and the dampening mechanism used. Modern DSLRs with their extreme FPS would probably go with lighter mirror mechanisms and less restrictive dampening. In short, I think the theory is just that - a theory.

Pulling from a rusty memory, I believe the "danger zone" was from 1/15 to 1 second. Longer exposures averaged out the mirror slap and shorter exposures minimized it. The newer Pentax 67 came out with MLU which helped . . . then we are left with shutter curtains starting and stopping (small movement indeed, but still there). I suppose this might apply to modern DSLR 35mm bodies - to a lesser degree. Something to be said for mirror-less.

I am still amazed at the sharpness I get from my old Olympus E-10 files (4 mp) - which had no mirror slap and an electronic shutter. :D
DiederikVP
AFFILIATED MEMBER
Posts: 457
Joined: Fri 18 Dec 2009 23:27

Re: Comparing resolution

Post by DiederikVP »

One of the best articles on this blog. Thanks for posting.
Steve Bingham
moderator
Posts: 91
Joined: Tue 5 Oct 2010 18:06
Active Member?: Yes
Post pictures?: Yes
Location: Payson, Arizona, USA

Re: Comparing resolution

Post by Steve Bingham »

Here are some recent tests on my Nikon 85mm f1.4G. Cement floor, rock solid tripod and head, MU, etc. You can view at 100% or download and look at 200%. The rest of the tests to follow as time permits. http://dustylens.com/lens_tests.htm
Post Reply