Examples of Large Format shots that cannot be done with small

[Jim Galli]

Those are probably the most important focal lengths for professional use but like many of us, I'm an eccentric amateur and don't use those focal lengths. That's one reason I've never gotten an SLR and tilt/shift lenses.

Give me a 190mm Bausch & Lomb Tessar "junker lens" with its 18-bladed aperture any time. 190mm... now that's a focal length. Like fine wine.

BTW, when I talk about "brute force" this is what I mean.

[Alan Gales]

From my understanding the Foveon sensor is thicker because of the three layers. This thickness causes less light to be absorbed so the camera does not work well at high ISO's. At lower ISO's the cameras are super sharp, have very little to no trouble with Moire and have fantastic color.

With the Foveon sensor as mentioned above you have to buy a Sigma camera.

For downloading raw you have to use Sigma's software which is not the most user friendly.

This is just what I have read. Foveon sensors are definitely intriguing. I've considered trying out one of their less expensive non interchangeable lens cameras like the DP Merrills. I've seen some used ones on Ebay pretty reasonable if some of you want to try one out.

[barnacle]

Regarding the Foveon sensor, I would not say it needs a filter, more like it can be used if concern about moire patterns is an issue. The Sigma cameras actually do not use it and I do not remember ever seeing any moire in the images.

I prefer to speak in terms of lines pairs to avoid confusion. And agree with 600 lp/in I could extract the info at 1200 dpi but I do not think is sufficient. I feel more comfortable with at least 4 times sampling to get a decent capture.

You also mentioned "there's no point scanning at more than 2400 or 3600 dpi". I agree this is the case with regular lens and emulsions but it's not hard to find a few quite good, capable of resolving above 100 lp/mm in MF and 35mm. That together with high resolution emulsions (some copy aerial and microfilm) can easily show details beyond 4000 dpi.

It needs *some* mechanism to limit the input frequency to the Nyquist limit: whether that is a physical low pass filter, a limitation of the lens, or simply never seeing an image that is above the resolution of the sensor I don't know. I might speculate that the microlenses on some sensors fill that function. But without something there, there *will* be moire patterning - AKA aliasing.

Like you, I would feel more confident scanning at least four times over the line pair frequency, for the reasons I explained earlier. As I said, *mathematically* it works, but practicalities intervene. And indeed, there are smaller format lenses and specialist films which have higher responses, but I was referring to run-of-the-mill LF lenses and emulsions.

I think the overall thing to bear in mind is that the largely analogue process of film emulsion gives a continuous scale while digital systems have issues as they approach their limits.

And I'm arguing still that because of those limits, and because of the nature of the sensors, the image I posted a few pages back could *not* have been made with a digital sensor. Something similar, yes; something looking the same, without significant post processing, no.

Neil

[Jim Galli]

Maybe we can get the mods to change the name of this discussion to the "foveon sensor" thread.

[Vaughn]

Maybe we can get the mods to change the name of this discussion to the "foveon sensor" thread.

I'm cool with such discussions here, but I will bookmark this thread so that I can use it as an example of why digital discussions are not allowed on APUG.

[Jim Galli]

Sorry, very low "boredom" threshold. I should probably keep my big mouth shut and use the unsubscribe button and check back later.

[SergeiR]

Well. "Can not do" is strong sentence.
There are number of things that are easier

- control of focal plane and distortion - you can't do it with sub-medium format and even with medium format without utilizing view camera
- transition of tones. I have somewhere direct comparing of same scene between dSLR (35mm) and dMF(645) (same 14 bit depth) that shows how different it is even for those. Even more so as you go higher. For someone with love for complex lighting scenes and dramatic lighting - LF is best thing to deal with. Unfortunately not in color though.
- aesthetics, of course. LF camera sometime gets you to places where small camera (specially digital) - won't. Of course other way is true too.
- easiest ever lenses swapping and ability to experiment with any bit of glass you can project image with

And so on. (there are more)

[Jim Galli]

The title is Malford aviating in his Peugot. Focal plane shutter effect.

[Jim Galli]

Sorry guys, I can't find the delete/move button. I will get in touch with the moderator.

My bad. I should have had a bit more forbearance.

[fishbulb]

Sorry guys, I can't find the delete/move button. I will get in touch with the moderator.

In my opinion, Foveon sensor discussion is on-topic.

1 - We started talking about things you can only do with large format
2 - In theory, large format film provides more resolution and better color accuracy than digital cameras
3 - When scanning film the actual color at the specific pixel is measured, vs. digital sensors, where each pixel is either measuring red, green, or blue light depending on what color filter is over it - so arguably, for color images, only 1/4 of the real colors are being measured.*
4 - The idea that Foveon sensors have all three colors on each pixel, such that they can produce more "film-like" colors - albeit with some other caveats.

Which matters for comparing color film vs. color digital, I guess. But not really for b&w film vs. b&w digital.

* An example from an On Landscape article - the 4x5 has been scaled down to match the ~21mp of the Canon 5DII. Note the missing red berries in the digital image, due to the color filter array. The dots are there (resolution) but just not the right color (accuracy).