Jump to content


Photo

Reverse diopter


  • Please log in to reply
20 replies to this topic

#1 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 01:25 AM

Can you flip a X2 diopter? If not, is there such thing as a X.5 diopter? The reason I ask is because of the larger image area possible in my Nikon lenses. If I squeezed the larger SLR image down to fit my Techniscope frame, might I get away with sharper resolution at maximum aperture like f1.2? What would it do to my depth of field? What does it do to the light collection properties (T-stop)? The additional glass would eat some light but the concentration of light energy into a smaller space would have some effect as well. What's your thoughts on this?
  • 0

#2 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 01:37 AM

I started thinking about this more. Wouldn't a .5 down-conversion be subject to the inverse square law? Not yet counting the light loss from additional glass, would that make the lens 2X2 brighter? Would that make the lens the equivalent of an f0.3? (f1.2/2=f0.6, f0.6/2=f0.3) Am I having a brain fart? With 2 stops gained at the lens, 1 stop pushed, and one stop bought back at the scanner could you shoot V3 at ASA 8,000? This can't be right, especially, the lens thing.
  • 0

#3 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 01:48 AM

If I haven't completely lost my freakin' mind, could you take a Hasselblad f1.8 lens and X.25 it and get an f0.112? A large format lens X.125 could get an exposure of a black hole? Someone fix my head before I think myself into a black hole.
  • 0

#4 Satsuki Murashige

Satsuki Murashige
  • Sustaining Members
  • 3510 posts
  • Cinematographer
  • San Francisco, CA

Posted 21 August 2008 - 01:57 AM

As far as I know, a diopter doesn't change your field of view - it only allows the lens to focus closer. There is a +1/2 diopter but what it does is convert infinity focus on the taking lens to 2 meters, or approximately 6ft. A +1 diopter would convert infinity focus to 1 meter, and a +2 diopter would convert infinity to 1/2 meter, etc.

I think you're thinking more along the lines of converting a larger format lens to a smaller format, which is a more technically involved process than just putting a diopter on a lens and would require the services of a lens technician. I don't really know much more about it, maybe someone more knowledgeable will chime in. Interestingly, there are a lot of zoom lens conversions between 35mm and 16mm, but you never hear of prime lens conversions, I wonder why... :unsure:

*Oh, and flipping a diopter doesn't have much effect, as far as I can tell. They're supposed to be used with the curved side facing out, wonder what happens when you flip it?

Edited by Satsuki Murashige, 21 August 2008 - 01:59 AM.

  • 0

#5 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 02:04 AM

Yea, I should have specified: I'm thinking of a behind-the-lens kind of thing. I got so busy digging on the light values, I forgot to mention that I had shifted my idea to the back. The Nikons have 1.8 inches of flange focal space to exploit.
  • 0

#6 Stephen Williams

Stephen Williams
  • Sustaining Members
  • 4708 posts
  • Cinematographer
  • Europe

Posted 21 August 2008 - 03:04 AM

Yea, I should have specified: I'm thinking of a behind-the-lens kind of thing. I got so busy digging on the light values, I forgot to mention that I had shifted my idea to the back. The Nikons have 1.8 inches of flange focal space to exploit.


Hi Paul,

Cooke Optics did this sort of thing all the time by replacing the back end of the lens, the 35mm Cooke 20-60 T3.1 became a S16 10-30 T1.6. They also can convert a 20-100 or 18-100 to a vert fast S16 lens.

Stephen
  • 0

#7 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 03:24 AM

I just woke up from a dead-sleep realizing I miscalculated the stops. 2X2 light increase on f1.2 is 2 stops and is f0.6, not f0.3.

So, Stephen, you're saying that a behind the lens X.5 already exists? It's doable? Who could make a X.5 for an SLR lens? I assume I'd have to go back to a rack-over camera as well unless I could find someone to grind a new back element like you mention. Gosh, my Som Berthiot zoom could be made an f1.8, 16-75. That's darn usable!
  • 0

#8 Stephen Williams

Stephen Williams
  • Sustaining Members
  • 4708 posts
  • Cinematographer
  • Europe

Posted 21 August 2008 - 03:47 AM

I just woke up from a dead-sleep realizing I miscalculated the stops. 2X2 light increase on f1.2 is 2 stops and is f0.6, not f0.3.

So, Stephen, you're saying that a behind the lens X.5 already exists? It's doable? Who could make a X.5 for an SLR lens? I assume I'd have to go back to a rack-over camera as well unless I could find someone to grind a new back element like you mention. Gosh, my Som Berthiot zoom could be made an f1.8, 16-75. That's darn usable!


Hi Paul,

I don't know if any are commercially available but I think it's doable, the FFD would probably stay the same like with a x2 extender.
Calling Chuck Colburn!

Stephen
  • 0

#9 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 12:22 PM

Twice the FOV. Two more stops speed. Twice the resolution in both H and V. If I had to spend only once and get a behind-the-lens widget that worked on all my Nikkies... that would be just fine. Who do you think can do this kind of thing? Should I go to one of the big lens companies first?

With twice the H&V resolution on an already sharpish Nikon, how would that, realistically, stack up to the best Zeiss on a lens projector?
  • 0

#10 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 12:31 PM

I haven't done the calculations yet. But, this could work fine for both the 2-perf and 3-perf frame. That's a lot of cameras now able to shoot better with somewhat cheaper lenses.

Where's the resounding "Thump" from all the lens-hungry RED crowd?
  • 0

#11 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 21 August 2008 - 01:30 PM

Diopters, in this context, are thin single element lenses of relatively long focal length. The diopter power is the inverse of focal length, in meters. So, a +2 diopter has a focal length of 500 mm, a +0.5 would be 2000 mm. The convenient approximation that works OK for these long focal lengths is that the diopter numbers are simply additive. Stack up the +0.5 and +2, and the effect is that of a +2.5 diopter lens (400 mm). (Put the higher power outboard, as it's more convex, and might touch the other lens if you did it the other way.)

The same diopter math is used for both photographic accessory lenses and eyeglasses. I've even used my reading glasses with a little autofocus digital camera for some quick, non critical diopter shots.

Putting a diopter on a camera lens results in a combination of slightly shorter focal length, but of course, the actual lens remains in the same place. Therefore, the focus scale no longer applies, and you can focus on objects that are closer than the mechanical limit of the focus ring without the diopters. Diopters don't change the exposure significantly. You can even use split diopters that shift the focal plane for part of the frame, sort of like bifocal eyeglasses. For instance, you could hold a person's face on the far side, and a small object in their outstretched hand on the close side.

The same theory works with concave lenses, which have negative focal length, and negative diopter powers. Some people need negative diopter powers for their eyeglasses, but there's no real use for them in photography. They'd just move the close focus limit farther away, giving you less range between there and infinity.



-- J.S.
  • 0

#12 Patrick Neary

Patrick Neary
  • Basic Members
  • PipPipPipPip
  • 873 posts
  • Cinematographer
  • Portland, OR

Posted 21 August 2008 - 02:43 PM

Twice the FOV. Two more stops speed. Twice the resolution in both H and V. If I had to spend only once and get a behind-the-lens widget that worked on all my Nikkies... that would be just fine. Who do you think can do this kind of thing? Should I go to one of the big lens companies first?

With twice the H&V resolution on an already sharpish Nikon, how would that, realistically, stack up to the best Zeiss on a lens projector?


Hi-

I think there might be a flaw in logic here, in that I can't imagine that adding optical elements to an existing lens could increase its resolution. Usually it's the opposite because you're adding more optical stuff that the light now has to travel through on its way to the film.
  • 0

#13 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 02:53 PM

A Techniscope frame only uses the middle of the SLR lens. It exploits only 1/4 of the len's total, usable surface areas and full resolving power. By reducing the len's SLR sized image at the back of the lens with a reground back element or add-on back element you are squeezing the larger image area of the SLR image into the smaller area of a Techniscope frame. Of course, an add-on element will reduce some of that gain in resolution. The more expensive option of regrinding and re-coating the back element would not suffer from this. The full resolving power of the lens could be used on that tinier frame. The full jump in resolution only occurs in the same SLR lens before and after modification. The number of elements, quality of glass and quality of manufacture would determine the amount of compromise to the resolving power of an add-on unit.

I post this as if I really know what I'm talking about. This is actually just my best theoretical guess. What I don't know is if one intemediate device can accomodate all the differing beam patterns from all of my Nikon lenses. I suspect that regrinding the back element of each lens might be the only quality solution. If I can't get a specific recommendation from someone here on a good lens cutter, I'll start emailing the big-boys in the lens world for some quotes. (Now, off to knock over some convenience stores)
  • 0

#14 Leo Anthony Vale

Leo Anthony Vale
  • Basic Members
  • PipPipPipPip
  • 2010 posts
  • Other
  • Pittsburgh PA

Posted 21 August 2008 - 03:40 PM

So, Stephen, you're saying that a behind the lens X.5 already exists? It's doable? Who could make a X.5 for an SLR lens? I assume I'd have to go back to a rack-over camera as well unless I could find someone to grind a new back element like you mention. Gosh, my Som Berthiot zoom could be made an f1.8, 16-75. That's darn usable!

Paul:
I believe Cooke completely replaces the rear section of the 35mm zooms to convert them to 16mm.
You'd have to do something similar with the Pan Cinor, replace the rear section. & you ought to wind up with a fast 16mm zoom.

As to what you're actually looking for:
http://www.vantagefi.../index_12.shtml

Hardly a universal solution. I think it slightlyreduces the overall length of the lens, just as a tele-ectender slightly lengthens it.

While on the Vantagefilm site also found this new item:

http://www.vantagefi.../index_43.shtml

---El "Wet Blanket" Pedante
  • 0

#15 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 21 August 2008 - 05:40 PM

Great links. Thanks, Leo. 5 elements sounds pretty complicated. Unavoidable, I assume. I'll give them a shout for pricing. It looks like one end is in Nikon mount already. I wonder what's the shortest lens it can handle. I could wait for them to grind up a fresh run if they could do me a X.5 version. The X.7 is probably for a larger full aperture. I could use a smaller frame model for my Techniscope frame and enjoy the inherent benefits.
  • 0

#16 Patrick Neary

Patrick Neary
  • Basic Members
  • PipPipPipPip
  • 873 posts
  • Cinematographer
  • Portland, OR

Posted 21 August 2008 - 07:44 PM

wouldn't that be sweet to have Nikkors 20mm to about 135mm all at a stop faster!
  • 0

#17 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 23 August 2008 - 06:11 PM

I did a little comparison of an 8-perf, 35mm frame and circle of coverage and a 2-perf, 35mm frame and circle of coverage. My initial assumption of a 2-perf frame being 1/2 of an SLR, 8-perf frame was incorrect. Below, is a graphic comparison which indicates that a 2-perf frame is X.6 to .7 of the SLR frame. That makes the Vantage adapter more accurate to even a 2-perf frame and a gain of only one stop probably the most practical for an adapter. Rear element regrinding could probably go closer to .65 or so and buy 1.5 stops at most.

I'll have to do some actual math (darn the left half of my brain!) to verify this info.

Attached Images

  • 8_down_to_2_perf_frame_measurements.jpg

  • 0

#18 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 23 August 2008 - 06:21 PM

2-perf width {0.868"} / 8-perf width {1.485"} = 0.5845". So, a X0.6 (X0.5845, specifically) adaptation. A regrind would be optimum. A 50mm SLR would drop to 29.225mm lens length and field of view. An f1.2 would drop to an f0.7014 speed. Still, not a shabby way to buy width, speed and resolution out of an already cost effective lens. Now, to get some quotes on grinding and have, yet, another heart attack.
  • 0

#19 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 25 August 2008 - 04:38 PM

My previous calculations weren't the best due to 2 factors: 1) They were based on 8-perf, Vista Vision frame measurements and not 235 SLR frame measurements. 2) They were based on frame width and not the diagonal of the frame measurements.

Here are the adjusted numbers using 235 vs Techniscope, diagonal numbers:

235: 1.417"W, .945"H, 1.703"D (1.339"Area, 4.724"Perimeter).

T-scope: .868"W, .373"H, .94475"D (.32376"Area, 2.482"Perimeter).

(Academy: .866"W, .630"H, 1.0709"D (.54558"Area, 2.992"Perimeter)).


.94475"D / 1.703"D = .554756 Diagonal conversion factor.

So, an f1.2 lens reduced by .554756 = f.6657 (roughly, f.6+1/5). The FoV would be roughly the same on a T-scope frame as an unmodified 50mm SLR lens on an SLR camera viewfinder. I'm uncertain what a X.5547, 50mm lens would look like compared to a 50mm cine lens in terms of FoV. For some reason, I can't seem to brain that at the moment. Would it look like a 27.75mm cine lens?

Please, double check my math. I got the D, A, P calculations here:

http://www.analyzema..._rectangle.html

I got the frame measurements here:

http://en.wikipedia....ki/Film_formats

http://en.wikipedia....of_film_formats
  • 0

#20 Paul Bruening

Paul Bruening

    (deceased)

  • Sustaining Members
  • 2858 posts
  • Producer
  • Oxford, Mississippi

Posted 25 August 2008 - 07:23 PM

A few more numbers for other formats (assuming a 50mm, f1.2 Nikon lens):

35mm Academy- .866"W, .630"H, 1.0709"D, X.62883, f.7545

S35mm full frame- .980"W, .735"H, 1.225"D, X.7193, f.863

S35mm at 1:1.85- .980"W, .5297"H, 1.1139"D, X.65408, f.7848

S35mm at 1:2.40- .980"W, .4083"H, 1.06165"D, X.62339, f.748

S16mm- .493"W, .292"H, .572986"D, X.33645, f.4073

S8mm- .245"W, .166"H, .29594"D, X.173775, f.2085

I don't know if S16 and S8 can be converted since they have differing FFD from 35mm. I just thought it would be cool to consider how fast a lens could be on any film camera. My Sekonic L-558 only goes down to f.5 anyway. Even if the glass could be cut to an image as small as S16 and S8, what would the shorter FFD do to add more speed? Is distance a factor in inverse square law, back-of-lens calculations? If so that might add another stop to the speed. Would that put the S8 speed down to f0.10 or something? I don't know. But that would be a fast-ass lens.
  • 0


FJS International, LLC

Aerial Filmworks

rebotnix Technologies

Visual Products

Opal

Willys Widgets

Glidecam

Media Blackout - Custom Cables and AKS

Abel Cine

Gamma Ray Digital Inc

Paralinx LLC

Broadcast Solutions Inc

Rig Wheels Passport

Metropolis Post

Technodolly

Wooden Camera

CineTape

The Slider

Ritter Battery

Tai Audio

CineLab

Ritter Battery

Wooden Camera

CineTape

Technodolly

rebotnix Technologies

Rig Wheels Passport

Paralinx LLC

The Slider

Aerial Filmworks

CineLab

FJS International, LLC

Glidecam

Tai Audio

Gamma Ray Digital Inc

Metropolis Post

Broadcast Solutions Inc

Willys Widgets

Abel Cine

Visual Products

Media Blackout - Custom Cables and AKS

Opal