Sensor size and look..

[Morgan_Moore]

There has been a liitle chat going on on another board (beginners) about format size and 'look' of images

ie how out of focus renders dependant on chip size - possibly 'the 3d look'

I did a little test that some may be interested to have a look at..

Format Test

One will note that I draw no firm conclusions from the test

S

[John_Black]

I think the biggest difference is the crop factor and how that influences lens selection.  I love the Mamiya 200mm F2.8 APO on the AFDII/P25.  The DOF is very thin and the lens is quite sharp wide-open - a fun combo.

Going to the extreme with a 40D, now I need to back-up 20 steps are so to get a useful framing.  DOF increases significantly and now the F2.8 view looks boring.  On the 5D or 1Ds2 the 200 APO is a bit more interesting, but it still doesn't have medium format FOV.  Shooting the 200mm on the P25 and having essentially 2X the FOV is what gives the lens its character (IMO).

For some people the negative size / DOF / FOV relationship won't matter much.   I think those are lucky people because as the sensor gets bigger, the price goes.

[Sheldon N]

Thanks for doing that test! Really interesting results.

There's quite a difference in the quality of the results as you go up in format size, even looking at the low resolution images.

Also, the difference between the H1 image at f/11 and the D80 image at f/2.8 is also very telling. Theoretically the H1 image would have more DOF, but the background blur is much more smooth and diffuse.

The D3 vs H1 images are more in line with what you'd expect, about a 1.5 stop difference required to equalize DOF. The qualitative results of the H1 are superior, though.

[EricWHiss]

Well Done!
and Thank You!

A picture is really worth a thousand words.

[Ray]

One will note that I draw no firm conclusions from the test
[a href=\"index.php?act=findpost&pid=178443\"][{POST_SNAPBACK}][/a]

Very wise, Morgan. There are a number of problems I've identified which make it difficult to draw a firm conclusion on a matter such as this.

I presume your Sinar 54LV sensor is approximately 48x36mm. If this is true, then we have approximately either a 1.5x multiplier or a 1.33x depending on which dimension we are using to equalize FoV with respect to the 3:2 aspect ratio of FF 35mm.

In the case of the cropped format D80, the multipliers are approximately 2.25x or 2x.

None of the F stops you've used correspond with any of these multipliers. Now that's not necessarily an issue. If you want the same look (because your large format camera has broken and you have to use the smaller format as back-up) then you do whatever it takes to get the effect you want.

However, it's very cumbersome and time consuming to test different formats with different lenses to determine what multiplier should be used and under what circumstances.

The nominal FL that appears in the EXIF data often doesn't appear to be accurate, for example a 50mm lens on the 40D should be equivalent to an 80mm FL on my 5D with 24-105 zoom but is in fact equivalent to 90mm. I wouldn't expect such discrepancies in F stop accuracy, but maybe there are similar errors there too.

Seems a bit of a mess to me.

[Morgan_Moore]

Very wise, Morgan. There are a number of problems I've
The nominal FL that appears in the EXIF data often doesn't appear to be accurate,[a href=\"index.php?act=findpost&pid=178602\"][{POST_SNAPBACK}][/a]

The Exif Data I havnt looked but it certainly is completely wrong on the sinar/blad combo

The Fstops I chose were just good old Fstops !

I wasnt really trying to match looks exactly - its just a demo of how the differnt chips look

The main question to me is .. in the days of the D3 and 1DS3, is there value in owning an MF system for 'general high quality photography' (like the editorial and medium end work I do)

Obviously fine art and architecture poeple need movements and super big files

It seems that given tthat given the faster lenses available for in DSLRs it is conceivable that the answer is NO

By looking at the D3 at F2 V H1 at 2.8 for example one could draw that conclusion IMO considering the price difference

I am still not completely clear however - I want to concentrate on the abiliy to isolate a sharp subject next time, ie I think the differences may be more apparent with a small aperture

I find myself using my blad at smaller apertures now than I used to - I am learning that the DOF seems to remain limited as seen on the F11 shot whille provideing a sharp subject

My main interest is generally a sharp person shown in relation to a soft and pleasant background that introduces context through hints rather than allowing the context (ie background) to distract from the person - others have a completely different agenda with thier photography

I have reasons to do with intimacy and natural perspective that I dont like to isolate the subject through the use of telephotos whose flattening effect I find most unpleasant

The test is not meant to be scientific in any way

S