Jump to content
Daniel D. Teoli Jr.

Why are cine' lenses so big compared to still lenses plus gears?

Recommended Posts

Up till now I've used still lenses for my video work. I bought my first cine' lens, a 35mm Rokinon. (BH sent the wrong one, I ordered a DS and they sent an AS, and it looked 'open box' but that is beside the point.)

This 35mm is a monster compared to a 35mm still camera lens. Why are cine' lenses so big compared to still lenses?

Another question is...why do they put gears on the focus and T stop rings? I prefer the feel of the still camera rings, but maybe I'm missing something with the gears.

Thanks

Share this post


Link to post
Share on other sites

The gears are so that the lenses can be used with a follow focus.

I’ve  used the Rokinons once or twice. I’ve never thought of them as being much larger than a stills lens. Certainly, compared to a master prime or an S4 they are tiny.

  • Upvote 1

Share this post


Link to post
Share on other sites

i've got a rokinon cine ds (believe the 20mm) and a rokinon still lens (believe 14mm) and they are pretty much identical in size.  one marked in T stops, the other in f.  one aperture changes click, the other doesn't.  one has aperture markings on both sides, other doesn't.  one has focus & aperture rings geared for motorized manipulation, the other doesn't.  one had built in lens hood, the other doesn't.  but length, diameter, and weight are pretty much the same.

 

Share this post


Link to post
Share on other sites

I will look up 'follow focus', but the lens is huge. Compare it to a ZM 35mm. The Rok is like a telephoto or zoom in length.

OK...I see what it is...

 

Marking the ring is a cool idea. I thought the puller read the focus dial. They just need to look for the marks. 

Do they have some sort of machine for the T stops as well? Or do they just try to match the looks and put a gear on the T stop ring?

Here is a ZM 35mm. Manual lens. I think it is under 3 inches. (But do not have in front of me.) My f2.8 35mm ZM is +/- 2-1/8 inches long. (W/O hood.)

Fair Use photo

 

1488899145_361549 lr.jpg

Edited by Daniel D. Teoli Jr.

Share this post


Link to post
Share on other sites

That's not a lens.. this is a lens!

1082319913_lenssizecomparison.jpg.72e1b1689d8451e82542d46363e4e980.jpg

This Primo is only a 50mm, you should see the 200mm! 

 

  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites

Cinema lenses are generally larger diameter to minimize the smallest available depth of field.
That is, maximizing the de-focussing effect of parts of the subject not in the plane of optimal focus.
You could have two lenses with the same focal length but one with a front element one-quarter  the diameter of the other.
The smaller one will behave more or less like the larger one stopped down by  the iris, that is, with a wider depth of field.
Cinematographers mostly use lenses with large front elements simply because there is no other practical way of capturing the shallow depth of field required for conveying the illusion of  closeness to the subject.
The only other way would be to use larger format film such as 65mm. That was one reason for the popularity of the Canon 5D - its oversize image sensor. Otherwise it's not a particularly great video camera.

Share this post


Link to post
Share on other sites

Older cinema lenses can be quite small.  Even too small for accurate focus markings and following focus.

Cinema lenses started getting larger to accommodate larger focus rings and accurate focusing.  Lens sets were designed so that all had the same size front to make changing lenses faster also (fits in the same matte box quickly).

Newer cinema lenses are more complex designs and have gotten much bigger and heavier.  As have still camera prime lenses.  And not all large prime cinema lenses have apertures larger than T 1.8 🙂

I have an older set of Ziess primes made for the Arri BL I or II and Arri IIc.  They are maybe 1.25 inches in diameter and maybe 2 inches long.  And they are not so easy to focus as the rings are so small 🙂  And they don't have gears on them either...

  • Upvote 1

Share this post


Link to post
Share on other sites
12 hours ago, Dom Jaeger said:

That's not a lens.. this is a lens!

1082319913_lenssizecomparison.jpg.72e1b1689d8451e82542d46363e4e980.jpg

This Primo is only a 50mm, you should see the 200mm! 

 

Jeeeesus!

Share this post


Link to post
Share on other sites

If I understand Keith's answer, which differs from but doesn't negate the validity of Bruce' answer, the larger lens allows more light to get in which in turn enables a wider minimum aperture (like T1.8)?

Share this post


Link to post
Share on other sites
19 hours ago, Keith Walters said:

Cinema lenses are generally larger diameter to minimize the smallest available depth of field.
That is, maximizing the de-focussing effect of parts of the subject not in the plane of optimal focus.
You could have two lenses with the same focal length but one with a front element one-quarter  the diameter of the other.
The smaller one will behave more or less like the larger one stopped down by  the iris, that is, with a wider depth of field.
Cinematographers mostly use lenses with large front elements simply because there is no other practical way of capturing the shallow depth of field required for conveying the illusion of  closeness to the subject.
The only other way would be to use larger format film such as 65mm. That was one reason for the popularity of the Canon 5D - its oversize image sensor. Otherwise it's not a particularly great video camera.

 

The size of the front element has no direct bearing on the depth of field, other than it just needs to be large enough to contain the entrance pupil. The entrance pupil is the image of the iris aperture seen through the front, which does relate to depth of field in the sense that it determines the maximum geometric aperture (f/2 for instance means that the entrance pupil is half the diameter of the focal length). So to achieve a relatively fast aperture - which is the main variable cinematographers use to control depth of field - the entrance pupil diameter (and thus the front element size) needs to be at least half as large as the focal length. That's why longer focal lengths generally have larger front elements - a 100 mm lens needs an entrance pupil (and a front element large enough to contain it) of 50mm diameter to be f/2.  But a 32mm lens only needs an entrance pupil and front element diameter of 16mm  to achieve the same speed.  

So while it's true that cinematography lenses are generally larger than stills lenses because (among other reasons) they often have faster apertures, a large front element is a correlation rather than a causation.

Different lens designs and different format coverage needs will result in lenses with different front element sizes, but two lenses of the same focal length set to the same f stop at the same distance from the subject will have more or less the same depth of field regardless of whether their front elements are different sizes. 

A Zeiss Ultra Prime has a larger front element compared to the same focal length Zeiss Standard Speed for example, but their depth of field characteristics are pretty much the same. Any perceived differences are due more to the optical design and how they defocus light than specifically the front element size. Simpler designs sometimes have slightly deeper depth of field towards the edge of frame for example.
 

Share this post


Link to post
Share on other sites
4 hours ago, Webster C said:

If I understand Keith's answer, which differs from but doesn't negate the validity of Bruce' answer, the larger lens allows more light to get in which in turn enables a wider minimum aperture (like T1.8)?

This is not the case.  My little Ziess lenses are T2.1 which is darned close to T1.8 🙂  I think the larger lenses with lots of glass are designed to eliminate a lot of imperfections, but not so much to allow more light to pass through.  All things being equal, of course T1.4 lenses are larger, but not so much larger as the modern lenses.

Share this post


Link to post
Share on other sites

A focussed image is made up of an infinite number of points of light that are produced by converging cone-shaped beams. At the point of maximum focus (the "focal plane")  the photons are most tightly converged; either side of that, instead of points of light they produce circles, and the overall result is what we called a "blurred" image. The further you get away from the focal plane, the larger the part of the "cones" that is intersected, resulting in larger circles and less focus.
If there is no iris fitted, the angle of these "cones" is specifically determined by the size of the  front element of the lens.

When an object is being imaged, photons from every point of light on its surface are going to strike the entire front surface of the lens, and eventually be re-converged  into the "cones" striking  the imaging surface. Obviously the size of the front element is going to determine the angle of these cones

For example, a lens 2 inches in diameter is going to produce cones with twice the angle of one with a 1 inch diameter. Therefore, the out-of-focus "circles" are going to grow in size twice as fast with the 2 inch diameter lens as they would with the one inch diameter lens. In other words it has a much shallower depth of field.

If you interpose an iris somewhere between the front element and the focal plane, it is going to have the effect of reducing the diameter of the front element, reducing the angle of the cones, and consequently  increasing the depth of field.
If you reduce the size of the aperture down to the size of a pinhole, then the angle of the cones will become extremely small, the photons will be almost parallel and it doesn't really matter where the image sensor is located; you have "infinite" depth of field.
Which is exactly what you get with most small consumer cameras.

Share this post


Link to post
Share on other sites
On 9/6/2019 at 2:33 PM, Keith Walters said:

When an object is being imaged, photons from every point of light on its surface are going to strike the entire front surface of the lens, and eventually be re-converged  into the "cones" striking  the imaging surface. Obviously the size of the front element is going to determine the angle of these cones

For example, a lens 2 inches in diameter is going to produce cones with twice the angle of one with a 1 inch diameter. Therefore, the out-of-focus "circles" are going to grow in size twice as fast with the 2 inch diameter lens as they would with the one inch diameter lens. In other words it has a much shallower depth of field.

Apart from assuming the front element is the field stop in every complex lens and ignoring the presence of an aperture and the role of focal length, you are confusing depth of field (which is in the object space - actors, scenery etc) with depth of focus (which is in image space - the sensor or film plane). They're not the same thing.

The only things in a lens that affect depth of field are the aperture stop and the distance to subject. Yes, the front element needs to be a minimum size to contain the entrance pupil, which together with focal length determines the f stop. But a front element can be and often is larger than the entrance pupil, with no affect on the depth of field.

Retro-focus wide angle lenses often use large curved meniscus lenses as a front element, with the entrance pupil quite small within their diameters. The large angle of view of a wide angle lens means the entrance pupil needs to be visible through the front right to the edge of that angle, which necessitates a large front element. It doesn't give the lens a shallower depth of field though, only the aperture itself and the subject distance will affect that.

A lens that covers a larger format or is designed to minimise vignetting or better control certain aberrations might use a front element much larger than the entrance pupil, but that also doesn't make the lens any faster. 

Longer focal lengths need larger front elements to maintain the same speed as shorter focal lengths, so a large or small front element may simply be the consequence of the focal length.

A zoom lens will often have a very large front element, but may be quite slow compared to a much smaller prime lens that is faster and capable of much shallower depth of field.

In short, the size of the front element plays no causative role in the depth of field. 

 

On 9/6/2019 at 2:33 PM, Keith Walters said:

If you interpose an iris somewhere between the front element and the focal plane, it is going to have the effect of reducing the diameter of the front element, reducing the angle of the cones, and consequently  increasing the depth of field.

This is incorrect. If an iris had the effect of "reducing the diameter of the front element" it would be acting as a field stop and as you stopped down it would eventually cause vignetting. An iris will be positioned within a lens such that it does not act as a field stop to limit the field of view, but only acts to limit the intensity of light. An iris can be fully stopped down, and depending on the angle they hit the front element, some light rays will still pass from the very edge of the front element through the centre of the aperture stop and on to the image plane. But the rays that would have traced a path from anywhere on the front element through the outer edge of the aperture stop will now be blocked. The angle of the cones hitting the sensor is a function of that stop (or more accurately the exit pupil) not the front element, and it affects depth of focus rather than depth of field. 

  • Upvote 2

Share this post


Link to post
Share on other sites

What Bruce said — larger barrels make it easier to space focus distance marks further plus longer rotations make focus racks and pulls smoother.  Plus it’s probably easier to design out imperfection and aberration. I have the new mirrorless Nikon Z6 and despite the idea that getting rid of the mirror shutter should have allowed more compact lenses to be built, the new Nikon f1.8 Z lenses are larger than their old ones of similar speeds. Mainly because they wanted to make them as optically perfect as they could.

Old cine lenses like Zeiss Super Speeds are not that large despite being fast, but they also aren’t very sharp wide-open. And pulling by distance marks are harder — if the actor is 6 feet away and leans back in a close-up by four inches, try to find 6’ 4” on some of those old small barrels... Look at that Zeiss still lenses at the top of the page, the last distance marked is 5 feet with infinity marked just a tiny space past that.

  • Upvote 2

Share this post


Link to post
Share on other sites

What David said!

I'm shooting now with the old Ziess Super speeds from the 1990's.  I don't use them wider than T 2.5.  And they ARE hard to focus. I feel for my poor focus puller...  But it's a small film, with small money, and... thus, small lenses 🙂

But they do take very nice images!

When I used the even much smaller Ziess standard mount/speed primes years ago, it was with my Arri BL 1 enclosed in a lens blimp, which had a very wide diameter gear on it.  Using the Preston follow focus, the focus marks could be made much wider apart, and they worked pretty well.  And in dailies, no one ever noticed a difference with the Panavision Primo primes from the "A" camera.  Not saying there wasn't a difference, but it was never noticed 🙂

Share this post


Link to post
Share on other sites

There is another factor that has so far been overlooked. 

Manual focus stills lenses are only designed for a person's hand to turn the focus ring slowly, by small amounts.  (This also applies to older cine lenses like standard speeds, which were primarily focused by hand via a follow focus.)

Modern cinema lenses need to be robust enough to handle the very high torque of remote lens motors, turning the lens 300 degrees from infinity to minimum focus in a fraction of a second, possibly hundreds of times per day, day in and day out.  

Repeated high torque focus racks like this would quickly destroy the mechanical parts most stills lenses (and it isn't good for cine lenses with older mechanical designs either).  When you add in the increased complexity for other reasons, e.g. breathing correction, and the need for quick, easy maintenance, it's no surprise that the construction of the lens becomes larger.

  • Upvote 1

Share this post


Link to post
Share on other sites
22 minutes ago, Daniel Klockenkemper said:

There is another factor that has so far been overlooked. 

Manual focus stills lenses are only designed for a person's hand to turn the focus ring slowly, by small amounts.  (This also applies to older cine lenses like standard speeds, which were primarily focused by hand via a follow focus.)

Modern cinema lenses need to be robust enough to handle the very high torque of remote lens motors, turning the lens 300 degrees from infinity to minimum focus in a fraction of a second, possibly hundreds of times per day, day in and day out.  

Repeated high torque focus racks like this would quickly destroy the mechanical parts most stills lenses (and it isn't good for cine lenses with older mechanical designs either).  When you add in the increased complexity for other reasons, e.g. breathing correction, and the need for quick, easy maintenance, it's no surprise that the construction of the lens becomes larger.

Hadn't considered this, but it makes a lot of sense.

Edited by M Joel W

Share this post


Link to post
Share on other sites

Over the years the size of cine lenses has been somewhat influenced by trends as well as functional requirements.

Early lenses for cinema were generally quite small, and often limited to more "normal" focal lengths. It took until the 50s for example for an acceptably sharp cine lens under 20mm to be released (for 35mm format), and telephotos tended to be relatively slow and not that large. Most lenses were around the size of Cooke Speed Panchros or smaller. Bausch and Lomb  Baltars (and later Super Baltars) were about as large as they got. (The exception being early anamorphics like the B&L CinemaScopes which have never been equalled in size!)

As zooms came into use during the 60s, lens supports started to be required, but the use of big heavy blimped cameras meant it wasn't such a big deal.

The introduction of quiet sync sound cameras in the 60s and 70s ushered in a new style of handheld cinematography which favoured smaller lenses, at the same time as high speed lenses like Canon K35s and Zeiss Super Speeds were introduced. Super Speeds are about as compact as a T1.3 cine lens set can get while maintaining unified front sizes. 

During the 90s and 00s, a vogue for large, extremely well-corrected lenses took over, starting with the  Panavision Primos. Both Zeiss and Cooke released large barrelled lens lines - Ultra Primes and S4s - that evolved into the even larger high-speed Zeiss Master Primes and Cooke S5s, which average 2 to 3 kgs, have 110 or 114mm fronts and lengths of 15 to 20 cm or more. The designer of REDs ill-fated Red Pro Primes admitted that he was specifically asked to make the lenses bigger by Jim Jannard, to make them look more like "proper cine lenses".

In the last 10 years a range of lower budget cine lenses based on still lens designs emerged, personified by Zeiss CP.2s or Canon CN-Es, which are relatively short and lightweight, but quite wide-barrelled, to allow for the expanded focus scale. The optics themselves are usually compact, and not as well corrected as more expensive cine lenses. The mechanics are relatively robust, but I've found for instance that Zeiss CP.2s are much less able to withstand a knock than an Ultra Prime for example.

Possibly the most successful recent design would be Leica's Summilux-Cs, which are high speed yet compact and well-corrected, using modern glasses, cam focus mechanics and lightweight metals. With the current vogue for hand-held, gimbal or drone cinematography, more and more designs are paring back the size and weight, even while having to increase the coverage for full frame sensors. The Zeiss Supremes are significantly smaller and lighter than the Master Primes for example, though at the cost of speed and optical perfection. Even Arri's Signature Primes, which are close to Master Prime size and as well-corrected as modern lenses get, have substantially cut the weight down. Panavision's Primo 70 series are similarly large barrelled and well-corrected but extremely lightweight, especially considering they also contain focus and iris motors within the lens itself.

The lowest budget ranges like Rokinon (Samyang) or Sigma manage to keep the size down to under 100mm diameters and under 1 kg in weight, while providing full frame coverage and even very high speeds on certain focal lengths, but there are of course many trade-offs, not least in image quality and mechanical durability. They have more variety in their size and speed (which is not optimal for cinematography), and are not as well colour-matched across the range (though probably a step up from stills lenses). Rokinon's DS lenses are really not much better than their stills series with gears added IMHO, you probably need to move up to Xeens for reliable mechanics, but then the diameter increases.

  • Upvote 1

Share this post


Link to post
Share on other sites

OK, just got a cheap follow focus puller. ($127) Very happy with it. Pretty good fit, some slop with sloshing of grease. But for the $$, I guess it is fine having never used $$ ones. When I first tried the lens I grabbed the t stop ring thinking it was the focus ring. Screwed up the shot. But it was just a test, so no big deal.

 

DSC06176.thumb.jpg.a3a88a7524a13b1474521713fce2a22d.jpg

 

Lens shown has to go back to BH. They keep sending AS when I order DS. Not once, but 4 times! The stop options for the follow focus is very useful. I almost threw the long handle away as it was hidden in the box. I thought it only used the wheel.

PS...I always glue extensions on the shutter button of the digital cams. I do mainly still photography. I hate the shutter buttons they use. Can't feel them for shooting from the hip. Terrible design. This cam has a synthetic turquoise cabochon epoxied on the shutter release. I would prefer a black cabochon, but blue was what I had.

Edited by Daniel D. Teoli Jr.

Share this post


Link to post
Share on other sites

AS and DS are not mutually exclusive. The AS in the Rokinon lenses means Aspherical. Even the DS versions have "AS" written on the lens. It may actually be a DS lens.

Share this post


Link to post
Share on other sites

I've only used the non DS versions, but the 360 view on the DS version on the B&H site doesn't have "DS" on it either. It just has a "II" since the DS was the second generation of the rokinon cine lenses, it may just say "II" instead of "DS"

  • Upvote 1

Share this post


Link to post
Share on other sites

Yes, you are right about DS. I guess DS means dual sided controls. One of the lenses had dual controls and ll, so it is a DS lens. Other one does not. Why didn't Rok just mark things on the box? 

 

 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.



  • Metropolis Post



    Ritter Battery



    Abel Cine



    Gamma Ray Digital Inc



    Tai Audio



    Wooden Camera



    Visual Products



    Rig Wheels Passport



    Paralinx LLC



    New Pro Video - New and Used Equipment



    Glidecam



    G-Force Grips



    Broadcast Solutions Inc



    Serious Gear



    CineLab



    Just Cinema Gear



    FJS International



    Media Blackout - Custom Cables and AKS


×
×
  • Create New...