Disambiguation: FoV & DoF on 4k & 2k

[Ryan E. Walters]

I just put together this blog post that I thought others might find useful / informative:

Lens Characteristics and shooting formats.

It has come to my attention after hearing some things said on various sets, as well as reading posts online that there is some confusion out there when it comes to lens characteristics and shooting formats especially when it comes to shooting 4k and 2k on the Red One. For some reason there seems to be the perception that when you switch from shooting in 4k to 2k using the same lens, you automatically get the same results as sooting on a lens that is twice as long. To help provide some clarification on the matter I decided to shoot a series of tests to provide a visual example of what is actually going on when you choose various shooting formats and how that impacts the final image.

Read the complete posting here: Disambiguation: FoV & DoF on 4k & 2k

 

[Birns and Sawyer]

Thank you for doing this, as it can be a tricky thing to explain without showing it. FoV characteristics do not follow DoF characteristics on different format sizes.

 

[Igor Ridanovic]

Excellent post Ryan. I forwarded it to RED L.A. User Group members. I could not post a "guest" comment on your blog.

 

[Stephen Strangways]

I commend your time and effort in putting this together, Ryan, but I'm afraid there are a number of problems with your article. I won't even try to correct your conclusions, because there are too many problems with your definitions, test setup, and understanding of lenses.

1) When you say "compression" and "The size of the individual picture elements and how close together or far apart they seem to be within the given picture." I think you are talking about perspective compression, but you are then confusing it with DoF later on.

2) You are either moving your camera position, or the wind is moving your subjects too much. It's throwing off your observations.

3) The focal length markings on your RED 18-50 are not exact. Doing the test again with 20mm and 40mm prime lenses would help, as they are usually much closer to their marked focal length.

4) "To do that, then a different set of lenses has to be used- a set designed for super 16mm." If you shoot 2K with a 40mm T3 lens designed to cover 16, or a 40mm T3 lens designed to cover 35, or a 40mm T3 lens designed to cover medium format... they all have the same focal length, they all have the same aperture, it's just the coverage (the largest format they will work with) that changes.

5) "If you switch lens formats from 35mm lenses to 16mm lenses, then this will be true, as smaller formats have greater depth of field." You're confusing lens formats and shooting formats. Different shooting formats (2K and 4K, 16 and 35) have different depth of field, given that you adjust for the same field of view. Different lens formats have the exact same depth of field on the same shooting format.

 

[Ryan E. Walters]

Thank you for doing this, as it can be a tricky thing to explain without showing it. FoV characteristics do not follow DoF characteristics on different format sizes.

Excellent post Ryan. I forwarded it to RED L.A. User Group members. I could not post a "guest" comment on your blog.

Thanks I'm glad that you found it helpful.

 

[Ryan E. Walters]

I commend your time and effort in putting this together, Ryan, but I'm afraid there are a number of problems with your article. I won't even try to correct your conclusions, because there are too many problems with your definitions, test setup, and understanding of lenses.

This was just meant as a basic example to show what is going on when changing formats and using the same glass on the Red One. To many times I have read, and heard on set that when you change formats the characteristics of the lens are changed as well. (It is usually said as "Now that I am in 2k I have a 40mm lens" Even though it may only be at 20mm.) This was not meant as the end all be all example. I even provided my own definition to help the reader understand what I was referring to, and meant be the terms. Could it have been done more scientifically? Sure- there are to many variables in this test to call it scientific by any means.

1) When you say "compression" and "The size of the individual picture elements and how close together or far apart they seem to be within the given picture." I think you are talking about perspective compression, but you are then confusing it with DoF later on.

Yes I was referring to to perspective compression, and not DoF. I'll have to go back through the write up, but I know the difference between then two- I may not have written it up clearly enough. So I'll check that out...

2) You are either moving your camera position, or the wind is moving your subjects too much. It's throwing off your observations.

The wind was blowing - but I tried my best to shoot when the wind was at its low point. Again it was not meant to be a scientific test- just one to show that a mere change of the resolution does not magically change the characteristics of the lens.

3) The focal length markings on your RED 18-50 are not exact. Doing the test again with 20mm and 40mm prime lenses would help, as they are usually much closer to their marked focal length.

Yep, they are not exact. The iPin data was helpful in at least telling me where the lens read as 20mm and 40mm. It also allowed me to ensure that the exposure was exact all the way though the test. Sure primes would be better- but would that then change the results? I do not think so.

4) "To do that, then a different set of lenses has to be used- a set designed for super 16mm." If you shoot 2K with a 40mm T3 lens designed to cover 16, or a 40mm T3 lens designed to cover 35, or a 40mm T3 lens designed to cover medium format... they all have the same focal length, they all have the same aperture, it's just the coverage (the largest format they will work with) that changes.

Yes Exactly! What I was trying to convey, is that in order to get the same lens characteristics of a 40mm lens when shooting in 2k you need to use a lens that cover that format. If you use a lens designed for 35 that is at the 20mm focal length and just crop in to 2k the only thing you are changing is the FoV. The FoV is now 40mm- every other aspect of the lens remains the same.

5) "If you switch lens formats from 35mm lenses to 16mm lenses, then this will be true, as smaller formats have greater depth of field." You're confusing lens formats and shooting formats. Different shooting formats (2K and 4K, 16 and 35) have different depth of field, given that you adjust for the same field of view. Different lens formats have the exact same depth of field on the same shooting format.

Yep- that is what I meant to say- but I missed it on that one. Thanks for the clarification.

Obviously I have some clarification to make- as I feel like you missed the overall point I was trying to make which was this: Keeping the lens the same and changing shooting formats only changes the FoV, everything else about the lens stays the same.

 

[Stephen Strangways]

Yes I was referring to to perspective compression, and not DoF. I'll have to go back through the write up, but I know the difference between then two- I may not have written it up clearly enough. So I'll check that out...

It's your 2K @ 20mm and 4K @ 40mm images where I don't think the perspective compression changed. Wind movement aside, each element is not closer together in terms of up/down/left/right, or in relative size.

The wind was blowing - but I tried my best to shoot when the wind was at its low point. Again it was not meant to be a scientific test- just one to show that a mere change of the resolution does not magically change the characteristics of the lens.

I might be more picky about your images than you are, and placing more weight on them influencing your conclusions than they actually did for you.

Yep, they are not exact. The iPin data was helpful in at least telling me where the lens read as 20mm and 40mm. It also allowed me to ensure that the exposure was exact all the way though the test. Sure primes would be better- but would that then change the results? I do not think so.

It's the difference in FoV you have between the 2K @ 20mm and 4K @ 40mm images, that I think should not exist (in theory) so I think the /i data is off.

Yes Exactly! What I was trying to convey, is that in order to get the same lens characteristics of a 40mm lens when shooting in 2k you need to use a lens that cover that format. If you use a lens designed for 35 that is at the 20mm focal length and just crop in to 2k the only thing you are changing is the FoV. The FoV is now 40mm- every other aspect of the lens remains the same.

Well then, to answer "So are you confused yet?" I would have to say yes! I'm still confused by what you wrote here. I think we understand the same thing, but are speaking completely different languages to describe it. Here's what I interpret from your paragraph above:

If you have a 40mm lens on 4K, and want the same FoV on 2K, you need to use a 20mm lens (whether designed to cover 35 or 16 doesn't matter)

By the same token, if you have a 40mm lens on 4K, and want the DoF on 2K, you stick with the 40mm lens, but you've got a different FoV.

When you say "all the characteristics" there's a problem, because you can't possibly have all the characteristics on a different format... you can have DoF or FoV stay the same, but you can't have both.

You say as much in the paragraph above that in your article, but I think the summary confuses things on two points: "all of the characteristics" and "So if using a 20mm lens in 4k, in 2k the FoV is now 40mm"

I know you're trying to keep it simple, but I think it's making things worse. If you said 'So if using a 20mm lens in 4K, in 2K the FoV is now the same as if you used a 40mm in 4K.' It's that frame-of-reference problem that you're trying to clear up to begin with.

To get more technical than I think you want to, you could say that using a 20mm lens in 4K, it has a FoV of 34.36 degrees, and in 2K the FoV is now 17.57 degrees.

Obviously I have some clarification to make- as I feel like you missed the overall point I was trying to make which was this: Keeping the lens the same and changing shooting formats only changes the FoV, everything else about the lens stays the same.

That's perfect! Say that! :thumbsup:

BTW, thanks for taking this as constructive criticism, because that's how I meant it.

 

[keicol]

4k and 2k focus

4k and 2k focus

Is the apparent softness I observe when shooting 2k the difference between 4k resolution and 2k? Even though I'm viewing both formats on a Panasonic 720p field LCD monitor? When I switch to 2k for off speed shooting there's an embarrassing softness on wide shots where you just don't quite feel that the shot is in focus, the focus seems to be somewhere between 20 and infinity on the lens while trying to focus in on an object at 15 ft. When shooting 4k the lenses focus perfectly. I've even tried fine tuning the back focus to sharpen a wide 2k shot with no real luck. I see this especially when using the 14mm, 18mm, 21mm and 25mm Cooke S4 Primes at 2k. Come to think about it I've had soft looking shots at 2k 120fps using the new Zeiss S16 low distortion primes.

I've seen this on different cameras with different lenses but only at 2k high speed. Is it just my tired eyes or has anyone else experienced this? I admit looking at an image on even the 24" Panasonic 720p field monitor leaves a lot to be desired.

thanks, kc

 

[Ryan E. Walters]

It's your 2K @ 20mm and 4K @ 40mm images where I don't think the perspective compression changed. Wind movement aside, each element is not closer together in terms of up/down/left/right, or in relative size.

Point taken- I think I'll set up a more controlled environment and shoot it again. I also think by introducing this point / thought I may have unintentionally encumbered the message I was trying to convey. (A 40mm lens is closer to a "normal" lens on 35mm, so it is not the best suited to show off some of the perspective compression effects that occur on a telephoto lens.)

I might be more picky about your images than you are, and placing more weight on them influencing your conclusions than they actually did for you.

Agreed- I was just looking for a quick and dirty example to show what was going on- not a text book test with text book results.

It's the difference in FoV you have between the 2K @ 20mm and 4K @ 40mm images, that I think should not exist (in theory) so I think the /i data is off.

Yep.

Well then, to answer "So are you confused yet?" I would have to say yes! I'm still confused by what you wrote here. I think we understand the same thing, but are speaking completely different languages to describe it. Here's what I interpret from your paragraph above:

If you have a 40mm lens on 4K, and want the same FoV on 2K, you need to use a 20mm lens (whether designed to cover 35 or 16 doesn't matter)

By the same token, if you have a 40mm lens on 4K, and want the DoF on 2K, you stick with the 40mm lens, but you've got a different FoV.

When you say "all the characteristics" there's a problem, because you can't possibly have all the characteristics on a different format... you can have DoF or FoV stay the same, but you can't have both.

You say as much in the paragraph above that in your article, but I think the summary confuses things on two points: "all of the characteristics" and "So if using a 20mm lens in 4k, in 2k the FoV is now 40mm"

I know you're trying to keep it simple, but I think it's making things worse. If you said 'So if using a 20mm lens in 4K, in 2K the FoV is now the same as if you used a 40mm in 4K.' It's that frame-of-reference problem that you're trying to clear up to begin with.

To get more technical than I think you want to, you could say that using a 20mm lens in 4K, it has a FoV of (x) degrees, and in 2K the FoV is now (x/2) degrees.

I haven't bothered to solve (x) right now, because I can't get iSee4K running.



That's perfect! Say that! :thumbsup:

BTW, thanks for taking this as constructive criticism, because that's how I meant it.

Yes, I think we are understanding the same thing and saying it in completely different languages. I had wanted to keep it as simple as possible so that anyone could understand it- especially those people who are non-technical. But obviously I have a ways to go in clearing it up. As I did not mean to make it more confusing... (I do appreciate your honest feed back. )

 

[Ryan E. Walters]

Is the apparent softness I observe when shooting 2k the difference between 4k resolution and 2k? Even though I'm viewing both formats on a Panasonic 720p field LCD monitor? When I switch to 2k for off speed shooting there's an embarrassing softness on wide shots where you just don't quite feel that the shot is in focus, the focus seems to be somewhere between 20 and infinity on the lens while trying to focus in on an object at 15 ft. When shooting 4k the lenses focus perfectly. I've even tried fine tuning the back focus to sharpen a wide 2k shot with no real luck. I see this especially when using the 14mm, 18mm, 21mm and 25mm Cooke S4 Primes at 2k. Come to think about it I've had soft looking shots at 2k 120fps using the new Zeiss S16 low distortion primes.

I've seen this on different cameras with different lenses but only at 2k high speed. Is it just my tired eyes or has anyone else experienced this? I admit looking at an image on even the 24" Panasonic 720p field monitor leaves a lot to be desired.

thanks, kc

My first thought is that either the focus is actually off, or that the back focus is off. But if you say that you have checked both and they are correct, then that cannot be the case...

I've had good results when shooting in 2k- so I'm not sure what is going on here...

 

[keicol]

I've had good results when shooting in 2k- so I'm not sure what is going on here...

It's most noticeable while shooting VFX elements on stage when focus is critical. I'm usually using Cooke S4 14 to 21mm or the Zeiss Ultra 16 lenses from 6mm to 12mm.

Focusing on objects in the 15 to 20' range is most difficult. As I've said, I've even tried to loosen up the back focus to see if I could dial in better focus. It must be a result of shooting in the 100 fps range at 2k with a 180 degree shutter that just doesn't hold up like 4k at normal speeds.


Could it have something to do with compression?

 

[Ryan E. Walters]

The "Scientific" Images-

The "Scientific" Images-

So I spent some time shooting some more "scientific" images that better represent what I was talking about. (They are not as fun visually, but they do better illustrate the point I was trying to make. These are just quick and dirty stills, ungraded. I'll be grading them and adding them to the blog post.

All images shot with the Red 18mm - 50mm (A 35mm film lens) Focal lengths noted are that of the 35mm format-

4k @ 20mm:

2k @ 20mm:

4k @ 20mm: with the center marked

4k @ 20mm: showing just the center crop of the 4k image

2k @ 20mm: (Again for direct comparison)

4k @ 40mm:

 

[Ryan E. Walters]

It's most noticeable while shooting VFX elements on stage when focus is critical. I'm usually using Cooke S4 14 to 21mm or the Zeiss Ultra 16 lenses from 6mm to 12mm.

Focusing on objects in the 15 to 20' range is most difficult. As I've said, I've even tried to loosen up the back focus to see if I could dial in better focus. It must be a result of shooting in the 100 fps range at 2k with a 180 degree shutter that just doesn't hold up like 4k at normal speeds.


Could it have something to do with compression?

I have not shot VFX elements with the Red, so unfortunately I cannot speak form experience. But form what you have said, it sound like the problem is occurring in the in camera sampling of the 2k image. The glass you are using is definitely good- and if it is in working order, then the only variable left is the scaling / sampling of the sensor that the camera is doing at 2k.

Do the results change at all when you move the object back and forth? Say you were focusing on an object at 5' or at 30'- do the results get better as far as focus goes? And the 15' to 20' range is consistent regardless of the lens used?

 

[keicol]

Do the results change at all when you move the object back and forth?

Closer items seem fine, it's when shooting background plates at distances more than 15 feet do we see this, and it's on every Red we've tried.

Maybe I should start a separate thread to see if there's been any others who've experienced this.

 

[Ryan E. Walters]

Closer items seem fine, it's when shooting background plates at distances more than 15 feet do we see this, and it's on every Red we've tried.

Maybe I should start a separate thread to see if there's been any others who've experienced this.

Yeah, starting a new thread would probably help you more ...

It is odd that changing the distance changes the results- as that would suggest that it my be an optic problem. But you've tried with difference lenses so maybe it's not ...

This is a wild guess: maybe what is going on here is that at that distance, and that focal length (The extreme wide end) when going to 2k the object is small enough that when it is sampled from the center of the sensor the processing is making it feel soft, which might also be affected by the high frame rate...

If I were you I would get a red drive or red ram and shoot at the highest speed possible in 4k, then change over to 2k and shoot at that same speed to see if the problem still persists. (Keeping at that "magical" distance.) I would also shoot a test in Redcode 28 and 36 to see if that effects anything. If you see no problems in either the 4k or 2k tests at those frame rates, then maybe it is limited to the higher frame rates (Above 30).

 

[Ryan E. Walters]

Disambiguated the Disambiguation ...

Disambiguated the Disambiguation ...

I just got done updating the article. Hopefully this has cleared up the issues, and more clearly shows what is going on when you change from 4k to 2k.

The Updated post:
Disambiguation: FoV & DoF on 4k & 2k

 

[keicol]

All your tests are done with foreground subjects. Can you look at background subjects at least 15 feet away with a 4k 20mm lens or wider? I would love to compare your findings with mine that suggest that the 2k version with the same FOV totally falls apart.

Thanks, Keith

 

[Ryan E. Walters]

All your tests are done with foreground subjects. Can you look at background subjects at least 15 feet away with a 4k 20mm lens or wider? I would love to compare your findings with mine that suggest that the 2k version with the same FOV totally falls apart.

Thanks, Keith

Yeah, no problem. But it will have to wait a bit- I should be able to do it this week. What specifically do you want represented in the image? Lens size, where the focus should be, etc...

 

[keicol]

Cool, thanks.
I'm looking for same field of view at 2k with a 4k shot at around a 32mm or wider. Try to match the width as close as can be gotten with the glass you have available for the 2k shot, I've used the Zeiss S16 lenses as well as the 14mm S4 at 2k and they all seem to be soft. Focus on the objects at 15 to 25' range and see if they can be focused on sharply. That's my problem when shooting background plates for VFX shots at off speeds which need to go to 2k to reach desired speeds. I'm finding it very difficult to focus on objects at those distances when that wide at 2k.

 

[Ryan E. Walters]

Cool, thanks.
I'm looking for same field of view at 2k with a 4k shot at around a 32mm or wider. Try to match the width as close as can be gotten with the glass you have available for the 2k shot, I've used the Zeiss S16 lenses as well as the 14mm S4 at 2k and they all seem to be soft. Focus on the objects at 15 to 25' range and see if they can be focused on sharply. That's my problem when shooting background plates for VFX shots at off speeds which need to go to 2k to reach desired speeds. I'm finding it very difficult to focus on objects at those distances when that wide at 2k.

So you can download the stills that I shot at the link below. They are nothing special visually, but hopefully they will help you in your evaluation. (Red 18mm - 50mm lens used.)

Focus was set on the middle brandy bottle (The text) which is 15 feet from the camera. The footage was shot in Redcode 36, processed in Red Alert using the highest debayer setting. I shot the 2k footage at 18mm at both 24 frames a second and at 75 frames. To help match exposure I increased the ISO to 800 for the 75 frames. (The lens was at T3 for all shots.) The 4k still was shot at 38mm, and I tried to match the framing as much as possible.

I hope this helps you: (66MB)
http://www.ryanewalters.com/downloads/ResTest.zip