LF vs APS-C

[Wilson Cheung]

I've had a niggle in the back of mind about the perceived advantages of the larger formats over digital were worthwhile. So I did a little comparison with my 4x5 LF against a Sigma DP-1s with an APS-C sensor. The LF was with a Provia 100f with a Super Angulon 90mm f8 shot at f16 4s. Slide was scanned in Epson V700 at 3200dpi with a little sharpening and levels adjusted as it seemed a little dark. The DP-1 has an equivalent FOV to a 35mm 28mm lens at was at shot at f5.6 0.5s.

The full scene.




Down-sampled LF vs DP1 centre crop of DP1



LF centre crop vs upsampled DP1 centre crop.



My thoughts on this very unscientific experiment is that the DP1 represented the light in the scene better, It could be I underexposed on the LF although I did use the same reading from a light meter. I think a full frame sensor would do a lot better but I suspect a drum scan would hold out.

[Clive Gray]

You don't think the colour difference might just be down to Provia being balanced for daylight ? Which will not have been anything close to the light temperature in that shot at a guess to which your digital will have done an auto white balance adjustment.

To make that fair at least you would need to use a decent light temperature meter and appropriate correction filters.

[Wilson Cheung]

I didn't try to correct the colour balance when I took the film shot. I tried to set the white balance on both the film and digital shots in post. What I meant was that using the equivalent amount of exposure on both film and digital, the film one seemed to come out a lot darker and contrastier. Other than that I was mainly interested in how much more detail was retained in LF.

[Martin Jan Köhler]

Hi Wilson,

Concerning detail,
Tim Parkin's landscape magazine "On Landscape" did a thorough test of different camera systems: http://www.landscapegb.com/2011/12/big- ... -comments/
that could interest you.
Amongst them is a Nikon D3x (35mm full frame digital, bit more quality than APS-C, but not much in terms of 35mm-versus-large-format) and also a 4x5" camera.

Best regards,
Martin

[Joanna Carter]

I didn't try to correct the colour balance when I took the film shot. I tried to set the white balance on both the film and digital shots in post. What I meant was that using the equivalent amount of exposure on both film and digital, the film one seemed to come out a lot darker and contrastier. Other than that I was mainly interested in how much more detail was retained in LF.

Of course the film shot would appear darker. If you take a shot on film, in the wrong lighting conditions, then part of the films colour sensitivity will not get the correct amount of exposure. No amount of post-processing can make up for this as, unlike with digital RAW files, the information in part of the spectrum is missing.

[timparkin]

I didn't try to correct the colour balance when I took the film shot. I tried to set the white balance on both the film and digital shots in post. What I meant was that using the equivalent amount of exposure on both film and digital, the film one seemed to come out a lot darker and contrastier. Other than that I was mainly interested in how much more detail was retained in LF.

Provia has about 6 stops of dynamic range on an Epson scanner whereas the typical digital compact has about 10 stops of dynamic range.

Here's my reblance to get the white balance closer. The screen is a good reference.



The tonality will be different - Provia is cooler in the shadows (as is Velvia for that matter) hence the colour in the shadows under the desk. Oh - and it's cyany in the highlights typically too (and everywhere else in my opinion!)

[Wilson Cheung]

Thanks for the link Martin will have a read of that. Good point Joanna, I never took heed in the past of when film is noted as daylight type so I shouldn't have dismissed Clive's comment my apologies. Tim that looks better thanks for the tip and also the bit about the dynamic ranges, I guess that is why the Provia has more contrast.

[dave_whatever]

On a tangent, does your powerball not have a rpm counter?

[Wilson Cheung]

Hehe, yes when you let it run down, it shows you the highest RPM achieved during that run.

[dave_whatever]

Nice one, those things are addictive.