Full frame "look"; real or a myth?

Gary Huff said:
The FF "look" is shallow DOF at wider angles. Everything else is the Emperor's New Clothes.
Click to expand...

Not true, the optical compression that comes from using longer focal lengths for equivalent FoV also plays a LARGE part in the different looks between different imaging formats.
 
Grug said:
Not true, the optical compression that comes from using longer focal lengths for equivalent FoV also plays a LARGE part in the different looks between different imaging formats.
Click to expand...

This is completely wrong. Coming from 30 years of professional photography, where a 165mm Super Angulon lens is an ultra-wide on an 8x10 Sinar camera ( largest camera format I've shot with is the 20x24 inch Polaroid camera ), there is ZERO difference between equivalent FOV when shooting different formats.

If you have properly matched the FOV with different film/sensor formats by using the appropriate focal length lens, then all of these formats are shooting from the SAME position ( SAME distance from camera to subject ), and the resulting image is IDENTICAL. ( provided you've matched the DOF F-Stop for each format )

If you find this hard to believe, put your camera on a tripod and take a still photo of a person using a wide-angle and a telephoto lens, but do not move the camera, and keep the F-Stop the SAME. Now take these two photos into your photo editing software and crop the wide-angle shot to match the same image area as the telephoto shot, and lastly adjust the size of the two photos so they display as the same size pixel size. ( cropped wide-angle photo and the telephoto photo ) If you've done this correctly you will notice that the two shots look absolutely IDENTICAL ( the cropped photo will have more grain and less detail ) but the FOV, the DOF, and the amount of image compression will be exactly the SAME.

If you want to introduce optical compression in an image, move your camera farther away and then use a longer focal length lens to achieve the same FOV you had with a shorter lens at a closer distance. Almost every fashion photographer will shoot models this way, to eliminate distortion, and to compress or "flatten" the image. ( farther away + longer lens = compressed image )
 
My answer to jamani is, yes, there is definitely a (quantifiable/qualifiable) FF look that no other camera offers. For me, after 40 years of shooting 8mm, 16mm, 35mm, 6x6cm, 6x9cm, 4x5, 5x7, 8x10, 2/3, 1/2, 1/3, APS, M4/3, S35 and other formats, it is a great feeling to pick up a MkIII and see exactly what I expect to see from a 17mm, 50mm or 105mm lens without having to apply yet another formula.

I prefer the simpler explanation from Kohli and Optitek and, though I appreciate and agree with the knowledge base offered by TheDingo, never in my life did I think we would ever be talking about the FOV of a 165mm Super Angulon lens on a Sinar 8x10 view camera!

Regards,
Scott
 
photochris said:
IMO the FF look is less about DOF (though that's also a part of the mix for many) and more about tonal gradation and better low light performance - FF sensors have a more natural look, highlight rolloff is smoother and less posterization and superior low light performance - generally speaking of course. In terms of detail there's no advantage, that's well documented.

Whether the 5d3 is worth the price vs. a 60d, 8string nailed some of the high points - the bitrate is dramatically improved on the 5d3, bigger/better LCD, headphone jack, better controls and so on. Superior image, better audio and the 5d3 has the clean HDMI out coming - those are worth the difference to me.
Click to expand...

Well the better image quality comes from a better processor in the more expensive camera. the chip size has nothign to do with it.

That said, the FF sensor look is VERY real. To achieve the same FOV you are using much longer lenses, and thus changing the way the lens interprits the scene. It is beautiful.

To reply to TheDingo, yes you are correct but only if you stop down to match DOF. If you keep both lenses at a F2.8 the dof characteristics will be different between the different cameras.

I think it has some applications in motion, but its better suited to Stills. I still far prefer the S35 frame size for motion, though i dislike it greatly in stills (aps-C).
 
I think TheDingo is right, except for the bit that Timur Civan corrected.

And still there's a fullframe look. But it's created by depth of field and resolution and detail, not compression.
(Except "resolution and detail" are there when everything else is equal, for example 36x24 film vs 60x45 film; definitely not 5D2 vs BMC)

About the camera-in-fixed-spot experiment: consider a 3m shooting distance, and two setups:
* 36x24mm sensor, lens at 100mm f/2.8 --> at focal plane the image is 72x108cm; DoF is 13cm
* 18x12mm sensor, lens at 50mm f/2.8 --> at focal plane the image is 72x108cm; DoF is 25cm
(if you want to use the same camera, you can just use a 50% center crop for the second shot)

So: field of view, compression, etc, are identical. But depth of field is not.

Math here (hidden, but it's there):
http://www.similaar.com/foto/dof/dof.php?f_x=1&f_r=1.5&f_mm=100&f_f=2.8&f_d=3
http://www.similaar.com/foto/dof/dof.php?f_x=2&f_r=1.5&f_mm=50&f_f=2.8&f_d=3

If you want to have the same DoF, with this 2x crop sensor, you have to put your lens at f/1.4, and push the ISO up two stops to get the same exposure
http://www.similaar.com/foto/dof/dof.php?f_x=2&f_r=1.5&f_mm=50&f_f=1.4&f_d=3

Now you have the same field of view, depth of field, compression, exposure, everything. Except the small sensor has pixels with 1/4th of the area and you are applying a lot of gain to them. But if this is not an issue (e.g. with the BMC), then the image will be basically the same (now, good luck finding a 22mm f/0.6 to match my 50mm f/1.4...).

Further proof and sample shots here:
http://www.similaar.com/foto/doftest/doftest.html
 
I stand corrected... A lot of great and detailed explanations here! To skip the theory, here's the test scenario described by TheDingo, shot with a Nikon D600 fullframe camera using 50mm and 85mm primes:
h6B8k.png
 
Yep. Now put the 85mm at f/6.8 (or closest you can), while compensating exposure with the shutter or ISO values, and you'll get nearly-identical images (everything the same, except for noise and detail)
 
Question:

Does a full frame sensor actually 'gathers' more light as well? In other words, is there a higher exposure when setting a full frame sensor at the same iso/aperture/shutter speed settings as a camera with the same settings and a smaller sensor?
 
I've never thought the look was about the normal and telephoto end, but the wide end?

Where things are more likely to distort etc.? Anyone have any thoughts on that?

I remember asking David Mullens on REDUSER about this very same thing and he also said that it didn't exist. Dunno, I still think there's a look associated with larger sensors that's not just DOF related--when it comes to the wider end.
 
Kholi said:
I've never thought the look was about the normal and telephoto end, but the wide end?

Where things are more likely to distort etc.? Anyone have any thoughts on that?

I remember asking David Mullens on REDUSER about this very same thing and he also said that it didn't exist. Dunno, I still think there's a look associated with larger sensors that's not just DOF related--when it comes to the wider end.
Click to expand...

Wider lenses are more likely to have distortions, regardless of which sensor you place them in front of, so that will cause a difference - but that's a flaw of the lens and is not endemic to the sensor size.
 
nothing said:
Wider lenses are more likely to have distortions, regardless of which sensor you place them in front of, so that will cause a difference - but that's a flaw of the lens and is not endemic to the sensor size.
Click to expand...

Right, but wouldn't you consider the ability to use a lens with a specific AOV on a larger sensor equivalent to the same on a smaller sensor part of the look?

I mean, say using a 24mm on a 5D versus a 12mm on a GH2. They should both have a very similar AOV, but the 5D's going to look different altogether due to distortion.

Maybe (likely) I'm just over-thinking it. This one's always sort of confused me, though.
 
Kholi said:
I mean, say using a 24mm on a 5D versus a 12mm on a GH2. They should both have a very similar AOV, but the 5D's going to look different altogether due to distortion.
Click to expand...

Not if that 12mm is properly designed for the micro4/3 sensor size - i.e. free of wide angle distortions (which are flaws, after all). There are rectilinear wide angles for many sensor sizes, this is not exclusive to FF.
 
Stranden said:
I stand corrected... A lot of great and detailed explanations here! To skip the theory, here's the test scenario described by TheDingo, shot with a Nikon D600 fullframe camera using 50mm and 85mm primes:
h6B8k.png
Click to expand...

Except didn't TheDingo say that the FOV & DOF would be the same? But the 85mm has clearly got much narrower DOF (as would be expected with a telephoto). Unless I missed something?

Those 2 photos are the same without moving the tripod? Because the distance between the edge of the lens and the film can look different in the 2 photos (maybe just an illusion because the lens is more out of focus in the second shot)?
 
simonpwood said:
Except didn't TheDingo say that the FOV & DOF would be the same? But the 85mm has clearly got much narrower DOF (as would be expected with a telephoto). Unless I missed something?

Those 2 photos are the same without moving the tripod? Because the distance between the edge of the lens and the film can look different in the 2 photos (maybe just an illusion because the lens is more out of focus in the second shot)?
Click to expand...

The small change in perspective is because I had the camera placed on a table, so there were some minor changes in position between the two shots :)

I can with a great amount of confidence say that the statement he made about DoF being same must either have been wrong or a misspelling on his side. Assuming I haven't misunderstood anything of his described test he does specifically mention and emphasize it to be the same: "... and keep the F-Stop the SAME."
 
He didn't write it very clearly, was easy to read both ways but pretty sure when he said "match the f-stop" he meant by calculation and "don't move the camera, and keep the f-stop the same" as a clunky way of saying don't move the camera and don't keep the f-stop the same. Samuel provided the calculated f-stop to be f6.8 or as close as you can get it. Would be very interested to see the same test with a third photo using this value.
 
I'm going to have to run some new tests on all this. My favourite photos that I've taken have all been 6x7 medium format shots, but with a lot of them I grabbed 'backup' shots with a 135 35mm camera at the same time - and although they were the exact same shots, the medium format version were in a whole other league 'look' wise. And I don't think it would have been the DoF, because mostnof the shots I'm thinking of were deep focus.
 
This is similar to why the IMAX format looks so interesting and wonderful to me. Whether people think it's a poor comparison or not, IMAX is like watching a super shallow (most of the time) and super detailed full-frame image with a FOV that's very wide. That's how I've always seen it. You look at The Dark Knight and The Dark Knight Rises and take a look at a lot of the IMAX footage and you'll see a lot of really wide FOV shots, lots of wide FOV and super shallow shots, etc. Crazy looking...
 
Stab said:
Question:

Does a full frame sensor actually 'gathers' more light as well? In other words, is there a higher exposure when setting a full frame sensor at the same iso/aperture/shutter speed settings as a camera with the same settings and a smaller sensor?
Click to expand...



in theory, yes, it is hit by many more photons (1.6x1.6x) and so it should give you cleaner high ISO
but that is if everything else is the same, which it usually isn't
and it's very easy to test: just shoot with both and you'll see the difference (5D2, 1 stop higher clean ISO than 7D)

And regarding lenses:
That's the big deal for me, but it's very personal: it depends on which lenses you have at your disposal.
I love vintage Leitz glass. These lenses were designed for full frame, and had issues on the wide angle (24mm and below).
So, my 35mm, 50mm, 90mm are awesome, the 135mm and 180mm are good but not so great, and the 24mm and 19mm would be pretty soft by today's standards. This means I want full frame. Because the lenses I want to use will shine on full frame, but leave me kind of handicapped on crop sensors.

Same for others: if you've got a great 24mm for full frame but all the 12mm lenses you can find for m43 are mediocre, well, then there's a different look for full frame: the one that your great lens gives you.
 
You must log in or register to reply here.
Back
Top