Why Bigger is Better (but not much else) S35mm vs. Vistavision Comparison

[Christoffer Glans]

So now lets talk about lenses as this is a big part of the story. I have come to believe that this is one of the key factors. By going to longer focal lengths on larger sensors we are minimising the geometric distortions we associate with short focal lengths. Shorter focal lengths are harder lenses to design and construct and Inevitably involve optical compromise. Assuming you want DOF equivalents you would stop down on the larger sensor again improving lens performance. [/IMG]

This is exactly the argument I'm trying to put forth. Just saying that you can match up a certain lens in FOV and DOF on s35mm and FF and call it a day with that it gives the same result is ignoring all the nuances that the step up creates. In another discussion I mentioned that my 35mm Canon CN-E looks way better than my 24mm. If I match up the look with closing down aperture on an FF I get a much sharper result with less distortion, but retaining the DOF and FOV of what the 24mm would give me on s35mm.

I find it interesting that in other discussions about viewable resolution, people argue that people will see a difference between 8K and 4K experiences and many of the same people who argue for this put forth the argument that we're not able to pick up the differences between FF and s35 when matching up lenses for the same image.

The thing is that this discussion has been going on before. When we went from 16mm sized digital sensors to 35mm digital sensors, the talk of matching up lenses went on for a long time. Now, everyone would argue about the benefits of 35mm sized digital sensors and not only because of how many pixels you can cram into them.

It's about widening your shots while keeping the superior quality of the normal focal length lenses. But there's a kind of sweet spot here as well as when talking resolution. I personally am of the opinion that viewers have a hard time distinguishing between 4K and 8K viewing if they view something as we normally view film and TV. Anything else is pixel peeping and highly trained eyes for the purpose of spotting that difference, which is at a great minority in the world. For the size of sensors, I don't think the benefits of medium format is that great compared to FF. FF in my opinion is the sweet spot where we can get the best quality image together with the fastest lenses, giving a sweet spot for both wide imagery and light throughput.

 

[Michael Tiemann]

Christopher,

Don't forget that because your CN-E lenses were designed to cover FF, you are "wasting" resolution of your 24. A Summilux 24 designed for a smaller sensor may well outperform your 35mm on a larger sensor. But at 6x the cost. Monstro gives me truly great quality from "affordable" lenses.

 

[Misha Engel]

I think its safe to say that the ACS tests and Panavision tests all agree with Mark’s results that assuming the camera and subject don’t move you can always match a smaller sensor shot with a larger sensor by increasing focal length and stopping down. You can also match larger sensors shoots with smaller sensors

Agree.

The rest is up to physical limits(Like diffraction) and the abilities, skills, patents, money, etc... of lens and sensor(camera) makers.(All within a certain bandwith ofcourse s35...IMAX).

Otherwise, when you have the budget, shoot Alexa65 and hope others will also build these size of sensors, or bigger.

 

[Jacek Zakowicz]

65mm, 70mm Imax are all fine but, if you consider practical application focused on optics (no pun ;-)) the sweet spot is really FF due to the still film legacy. For 100 years or so the optics have been perfected, they are the brightest, sharpest, most compact lenses available today.
You go medium format and the (lens) size doubles and the speed is cut by factor of 4, not to mention accessibility and cost. Marketing notwithstanding- FF (VV) is where it's at now. But the (bigger, more magnificent) race has't slowed down so far, so who knows....

 

[Björn Benckert]

This is exactly the argument I'm trying to put forth. Just saying that you can match up a certain lens in FOV and DOF on s35mm and FF and call it a day with that it gives the same result is ignoring all the nuances that the step up creates. In another discussion I mentioned that my 35mm Canon CN-E looks way better than my 24mm. If I match up the look with closing down aperture on an FF I get a much sharper result with less distortion, but retaining the DOF and FOV of what the 24mm would give me on s35mm.

I find it interesting that in other discussions about viewable resolution, people argue that people will see a difference between 8K and 4K experiences and many of the same people who argue for this put forth the argument that we're not able to pick up the differences between FF and s35 when matching up lenses for the same image.

The thing is that this discussion has been going on before. When we went from 16mm sized digital sensors to 35mm digital sensors, the talk of matching up lenses went on for a long time. Now, everyone would argue about the benefits of 35mm sized digital sensors and not only because of how many pixels you can cram into them.

It's about widening your shots while keeping the superior quality of the normal focal length lenses. But there's a kind of sweet spot here as well as when talking resolution. I personally am of the opinion that viewers have a hard time distinguishing between 4K and 8K viewing if they view something as we normally view film and TV. Anything else is pixel peeping and highly trained eyes for the purpose of spotting that difference, which is at a great minority in the world. For the size of sensors, I don't think the benefits of medium format is that great compared to FF. FF in my opinion is the sweet spot where we can get the best quality image together with the fastest lenses, giving a sweet spot for both wide imagery and light throughput.

Wrong,

The reson your 24mm looks less good is that you also crop in on the image circle, you can not use two FF lenses for these comparisons without clouding the result. You need lenses that are built for the format.

For example a s16 24mm lens will piss on most FF 24mm lenses if you shoot s16.

When you don't use the whole realestate of the lens on the s35 camera that makes the picture look more muddy and together with less pixels things are getting even worse. Also you got the canon lenses. I think your canon is based on the 24mm L series. Not sure about this / just my assumption. The L24mm is not the best of lenses, the canon SSC Aspherical 24mm is the good one and I doubt they used it on your set.

So the small nuances you see possibly it´s not the size of the sensor that the creator of those nyances.

Using your lenses and gear to measure things is never bad. But its important to understand that the results can be poluted or simply just accurate in that one scenario. If you want small sensor lenses to compare to try the Zeiss Digiprimes.

I think it´s important to separate the comparison between the qualities of lenses ans the physics of things. If you go by the lens quality, well then your thesis only works for one given lens.

Now it´s quite easy to argue that there is a huge amount of good FF glas out there. Thats what has been the aim for pocket cameras since leica invented that light meter. But you also need to consider cinema glas has mostly been s35 or even as small as accademy and where developed and built for those smaller formats.


So yes, its ok to say earth is flat and as long as you are not going to space that assumption will not ruin your day.

But don't take me wrong, I love the larger sensors and Monstro according to my testings outperforms both Dragon and Helium on s35mm even though it´s doing 6k vs 8k. The sensor is simply that much better. Then there is a lot of FF glas that I love, my Cinemascopes included, and for FF glass the others can not compare.

 

[Michael Lindsay]

I am pretty sure of who makes them but I don't think it is super relevant or useful to relay.. some people are upset they are not made in Germany but neither are the PL Sumicrons? Who makes angenieux ez series lenses? they are not made in France?

Surely what matters the most is the what they are not the who made them... questions of material provenance are less important to me than questions of function or even design provenance.

So who is making them?


Matt Duclos writes: "I don’t have any firm information on the origins of this glass, but it’s not too difficult to deduce a few clues… "

Dotn know what he means with that, but I guess he knows.

 

[Jon Dishler]

In HD format the Alexa LF is only 1.13x field of view wider than the Helium in HD format. The resolution disparity is 400%. Considering this is the new thing out there, it is sort of interesting. The Monstro is of course a lot bigger/better.
Thanks for the chart Phil.

http://phfx.com/tools/formatCompare..._typeTech=ARRI|Alexa LF&formatTwo_format=16:9

 

[Jacek Zakowicz]

I am pretty sure of who makes them but I don't think it is super relevant or useful to relay.. some people are upset they are not made in Germany but neither are the PL Sumicrons? Who makes angenieux ez series lenses? they are not made in France?

Surely what matters the most is the what they are not the who made them... questions of material provenance are less important to me than questions of function or even design provenance.

I think it is very much relevant. Changing manufacturers may mean that the only thing left on the lens is the engraved name. Manufacturers develop style and look, feel, fit and finish that is very difficult and time consuming to replicate by complete new facility in a different part of the world and different culture. Not to mention pricing pressure...

 

[Björn Benckert]

I am pretty sure of who makes them but I don't think it is super relevant or useful to relay.. some people are upset they are not made in Germany but neither are the PL Sumicrons? Who makes angenieux ez series lenses? they are not made in France?

Surely what matters the most is the what they are not the who made them... questions of material provenance are less important to me than questions of function or even design provenance.

I assumed they where Zeiss but it does not say Zeiss on them, So I guess not. Why not say where you think they are from? I'm curious as it's a pretty big lineup, not many mange to make such.

 

[Michael Lindsay]

I think it is very much relevant. Changing manufacturers may mean that the only thing left on the lens is the engraved name. Manufacturers develop style and look, feel, fit and finish that is very difficult and time consuming to replicate by complete new facility in a different part of the world and different culture. Not to mention pricing pressure...

I don't think ARRI has tried to replicate as much as you might imagine .... the lenses are def an aesthetic and ergonomic change of direction for ARRI... I felt it was big news but it seems I was wrong.

 

[Misha Engel]

Looking forward to independant reviews on those signature primes, when they have 1% of the sharpness of the optics used in current ASML machines they will be the sharpest cine lenses of today.
It would be fun when they are made by Samyang.

 

[Les Hillis]

Just throwing another data point out there, using the Helium sensor -



(Interesting to note how the minor difference in parallax views here enables a 3D effect when using the cross-eyed viewing technique :smile:.)

I know comparing a 50mm to a 100mm lens isn't the same as making a comparison between wider lenses, but was still surprised how easy it was to match them on a superficial level.

I'm still seeing some more subtle differences between them though. In the bokeh and out of focus areas for example there seems to be something going where both the foreground and background blurriness can't be matched at the same time.

There's a lot going on here (a lot of inter-related variables), but I haven't got much more to say about it. I am seeing a difference between the formats, but I actually like the way they both render the 'same' image.

For me personally, the higher 8K resolution trumps a lot of other concerns.

 

[Björn Benckert]

Just throwing another data point out there, using the Helium sensor -



(Interesting to note how the minor difference in parallax views here enables a 3D effect when using the cross-eyed viewing technique :smile:.)

I know comparing a 50mm to a 100mm lens isn't the same as making a comparison between wider lenses, but was still surprised how easy it was to match them on a superficial level.

I'm still seeing some more subtle differences between them though. In the bokeh and out of focus areas for example there seems to be something going where both the foreground and background blurriness can't be matched at the same time.

There's a lot going on here (a lot of inter-related variables), but I haven't got much more to say about it. I am seeing a difference between the formats, but I actually like the way they both render the 'same' image.

For me personally, the higher 8K resolution trumps a lot of other concerns.

Again, as long as your 50mm lens is not a s16 lens the observations you made is possibly due more to the fact that you "crop into a image circle" than anything else. And if it is a s16 lens then, yes the difference can be due to different lense design.

This thread is now 20 pages long. From skimming it trough I think that we can agree that the differences seen between swapping between s35 and VV while compensating with focal length and speed can be described as far less than what you see between two different lenses. Something like swapping from a cooke 50mm to a Canon 50mm creates a bigger difference than swapping sensor size and compensating with focal length and speed.

And we still wait for someone to describe this difference between formats clearly. So far it has not really happen. Only tests using different lenses. And well, a different lens is a different lens and it will always render a different result. But it would be great if we could come down to some sort of consenuss of what the difference in format brings to the table.

So far I got the following from following this thread about larger sensors:

1. More shallow depth of field with larger sensor.

False. There is to date no large format lens that can create a shallower depth of field then what can be created using a s35 lens with the same FOV

2. A change of pespective/ parallax. Compression of the scene, Magnificaton or what ever you want to call it.

False perspective change is due to position of camera nothing else, no matter what your sensor size you use, you can find a focal length and Fstop to match, Normally the parallax change is due to the fact that altered nodal point needs to be taken into account if not, then you simply have moved the camera, and thats why the perspective, parallax, compression of the scene, level of magnification or what ever fancy name you want to call it, got changed.

3. Smoother Defocus Rolloff.

False, This is built into the design of the lens does not really have anything to do with the size of things.

4. Different panning speeds needed.

False, panning speed possible before getting judder problems is related to the FOV not the sensor size. Then also rolling shutter is related to the number of pixel lines and redraw time, but not directly related to sensor size.

5. The wider lens you got the more fisheye it will get so when compensating sensor size with a wider lens you image becomes more bevelled.

False, the difficulty with keeping a lens rectilinear is related to the FOV not really to the focal length, So it's just as difficult to make a 35mm FF lens rectilinear as it is to make a 50mm s35 lens rectilinear.

6. More resolved resolution.

True.


So 1,2,3,4,5 is as I see it, Is simply false. I would be happy to be proven wrong on these 5 points with a valid test that is not mudded by cropped image circles or moved nodal points, etc. Or more preferably a nice diagram and explanation of what happens when light goes trough the lens would do even better.

6. Yes this is true. You can capture images that simply looks better with a larger sensors due to the fact that it´s difficult to create sensors with high density photosites. So the VV array looks better than the s35 array when both capturing 8k. And CoC comes in to play at a certain point so there are still quite a few reasons why having the larger sized imager matters.

And to me the character that large and thick glas brings to an image has charm. Big anamorpics like the cooke and the hawks has this thick glass look that I find appealing and old big zooms like the Angies and the K35 zooms has the same type of character, that to me looks something like looking out trough a thick stack of glas, which makes it cinematic in a great way that small little lenses, no matter what format, can newer really match. Kowas will look anamorphic but they don't get that looking out of a periscope kind of feel like larger anas like Hawk, Cooke or PV brings to the table. But as I see it that is more, I think, me that like those kind of lens distortion effects caused by thick glass, more than a certain size of the sensor. And sure the oval DOF,flares and edge softens that comes with anas.

Again, I really like what large formats bring to the table. Like this week we shoot 1.3 Hawk V-light lenses with full frame Alexas down in Bucharest. But for sure I will test the lenses also on monstro. Hope the longer V lights also cover full 8kFF and that gives a nice 3:1 aspect ration with a .3 pinch of anamorphic, Which in my mind must just look amazing. So as I see it, I think the fascination here is more about maxing out the image circle at hand and by doing so get character, than actually moving to a larger imager, except for what the added resolution brings.


And yes, it´s odd. All this talk about how little the resolution matters to most DP´s then when large format comes and it pretty much only gives just that, a higher resolution, then it´s all of a sudden a gift from god. I though 2,5k capturing was enough, was it not?

 

[Misha Engel]

That sums it up pretty well, lets shoot some pictures and have fun about Nolan and van Hoytema walking around with those big heavy IMAX camera's. It's not all bad news, a few people could pay their bills because they had to carry all that stuff around. Happy shooting,

Misha

 

[Björn Benckert]

That sums it up pretty well, lets shoot some pictures and have fun about Nolan and van Hoytema walking around with those big heavy IMAX camera's. It's not all bad news, a few people could pay their bills because they had to carry all that stuff around. Happy shooting,

Misha

They win on point 6. I would say even more so on film then digital these days.

 

[Misha Engel]

They win on point 6. I would say even more so on film then digital these days.

Look at yedlin, Arri Alexa 65 has more resolution than 70mm IMAX(scanned at 11 Mpixel).
Otus 85 has a resolving power of 85 Mpixel wide open and 107 Mpixel at F4.1 (Cine sensors are still strugelling with 36 Mpixel).
Vista Vision with a Helium gets to 67 Mpixel, so still not running out of resolving power on the lens side VV and FF wise.
Panasonic is bringing us even more Dynamic range in the future, film is dying. It's easier to find a needle in a haystack then a film lab these days.

 

[Björn Benckert]

Look at yedlin, Arri Alexa 65 has more resolution than 70mm IMAX(scanned at 11 Mpixel).
Otus 85 has a resolving power of 85 Mpixel wide open and 107 Mpixel at F4.1 (Cine sensors are still strugelling with 36 Mpixel).
Vista Vision with a Helium gets to 67 Mpixel, so still not running out of resolving power on the lens side VV and FF wise.
Panasonic is bringing us even more Dynamic range in the future, film is dying. It's easier to find a needle in a haystack then a film lab these days.

Yes but they like film, and probably more the way it deals with colors and over exposure than resolved resolution... And you also have that argument for film that "the silver is different on each frame" Simply that in comparison with a sensor the "pattern" of the negative is different from one frame to the other and that would give a sort of improved resolution. Where as on a sensor the photosites are basically fixed. Film would from this get some sort of super resolution when you see multiple frames after each other. I'm not buying much into that argument, even if its a commonly used one. But I think film with its artifacts like thick grain and odd colors looks strangely great most of times.


We did a 4k TV series recently where the DP decided to shot the intro on film and the rest on helium. Same lenses and everything else, and sure the film stuff looks like it's drawn on sandpaper in comparison and don't hold even half the resolution. But I still find it tasteful at times.
Then also as this was early color science of helium there is some color fringing and other stuff that simply would not happen on film.

In this extended trailer some of the shots where film. I think it's quite easy to see even when watching in low res. The clip is viewable in UHD on vimeo not sure how it shows here.

https://vimeo.com/243111020

 

[David Mullen ASC]

Ignoring the cost and practical issues, what I like about 65mm film photography is that you get all the beauty of color negative but the finer grain and increased resolution of digital. But I work at a budget level where I can’t ignore the cost and practical issues!

 

[Hrvoje Simic]

I think its safe to say that the ACS tests and Panavision tests all agree with Mark’s results that assuming the camera and subject don’t move you can always match a smaller sensor shot with a larger sensor by increasing focal length and stopping down. You can also match larger sensors shoots with smaller sensors but there is a limit on your ability to match DOF.

2. A change of pespective/ parallax. Compression of the scene, Magnificaton or what ever you want to call it.

False perspective change is due to position of camera nothing else, no matter what your sensor size you use, you can find a focal length and Fstop to match,

What happens between S35 crop of 25mm FL and Vista crop of 35 mm FL on near to mid distances:

What happens between S16 crop of 24 mm FL and FF35 crop of 50 mm FL on short distances:

 

[Hrvoje Simic]

3. Smoother Defocus Rolloff.

False, This is built into the design of the lens does not really have anything to do with the size of things.

Sensor size is just the crop.


This whole topic is related to the lens focal length you will use for the same AOV based on the format size/crop.

This is all related to the lens. Format just defines which lens you will use.