Jump to content


Photo

Zeiss T* coated lens question, Flare?


  • Please log in to reply
16 replies to this topic

#1 Tim Carroll

Tim Carroll
  • Sustaining Members
  • 2165 posts
  • Other
  • Chicago, Illinois

Posted 15 July 2006 - 12:28 PM

As a follow up to the lens test I did a few months ago, I have done a more complete test of available lenses for the Arriflex 16S camera, and I followed Mitch Gross's advice and set the test up so that I was shooting with all the lenses wide open. And I got some results that have confused me.

To try to keep everything uniform, I set up a group of objects which included my Apple Cinema Display monitor, and listed which lens was used for each shot on the screen of the monitor in white letters on a black background.

Comparing the image captured with an old Zeiss 10-100 T3.1 zoom (I have no idea when this lens was last serviced), wide open, at the widest focal length 10mm, and an old Zeiss 8mm T2.2 prime (which was recently serviced and set properly) wide open, the letters on the screen were remarkably different.

The quality of these frame grabs don't show what I am talking about as well as the actual film does, but you hopefully can get the idea.

Here is the Zeiss 10-100 T3.1 shot.
Posted Image

Here is the Zeiss 8mm T2.2 shot.
Posted Image

On the actual film, you can see the woman's face in both shots is fairly sharp and it is on the same plane as the white letters. And the back wheel of the silver Porsche Spider model (behind the front motorcycle) is on the same plane as the white letters. But on the Zeiss 8mm clip, the white letters are really fuzzy, and on the Zeiss 10-100 clip, the white letters are fairly sharp.

Both lenses have the red T* for the Zeiss special coating, so I am trying to figure out why the one lens handles the letters better than the other. Does it have something to do with the light from the LCD Apple Cinema Display shining back into the lens, one handles it better than the other? I am really curious, because as I mentioned, on the actual film the difference between the two lenses is much more noticeable than in these screen shots. And I would think this could cause problems on a shoot.

Thanks for any and all assistance.
-Tim
  • 0

#2 Tim Carroll

Tim Carroll
  • Sustaining Members
  • 2165 posts
  • Other
  • Chicago, Illinois

Posted 16 July 2006 - 12:51 PM

Anybody?

-Tim
  • 0

#3 Laurent Andrieux

Laurent Andrieux
  • Sustaining Members
  • 1527 posts
  • Cinematographer
  • France

Posted 16 July 2006 - 05:56 PM

First of all, comparing a 8 mm to a 10 mm - I notice the 8 mm is defenetly wider than the zoom's 10 mm - is not that obvious... I mean that with short lenses, a 2 mm difference is a lot, not only for what is about it's viewing angle, but mainly for what is about the lens' carachteristics, esp. definition.

2. It looks like a general loss in definition and I would suspect the coating, but I'm not sure it's not a problem of steadiness of the camera.

3. The zoom shot looks overexeposed, compared to the 8 mm one. It may be a point.

For sure, it's not easy to tell from your stills... but I would never shoot a video screen as a test, myself. Also, one can ask himself where was the focus made ?

I notice the letters on the left flag "SCH" look pretty similar, but objects in the center look out of focus on the 8 mm shot.

I would make tests on both Foucaults charts and definition charts as to make an opinion. Also, you can test the coat in a typical flare situation (sun in the frame for instance) as to check this particular point.

Just a 2 cents before any body else answers...

Regards,
  • 0

#4 Charles MacDonald

Charles MacDonald
  • Sustaining Members
  • 1157 posts
  • Other
  • Stittsville Ontario Canada

Posted 16 July 2006 - 09:47 PM

I would make tests on both Foucaults charts and definition charts as to make an opinion. Also, you can test the coat in a typical flare situation (sun in the frame for instance) as to check this particular point.

Regards,

I would agree with Laurents observations. The ONLY item that looks sharper in the 8mm shot is the door of the VW Bug and that is because the zoom looks washed out.

A lens chart would be a good thing to use, with shots with the lens set narer and farther. Another clasic is the yardstick (meter stick) at a 45 degree angle to the camera. You foucs on the marking atthe center and see where the numbers get fuzzy.

For flare test, how about a dark bacjkground with a CLEAR Incandesent bulb ..(off to one side) any flare should turn up in the background or as a ghost.


Since NOTHING is really sharp on the 8mm Shot, I would be windering if the lens has been tampered with and needs some TLC>
  • 0

#5 Tim Carroll

Tim Carroll
  • Sustaining Members
  • 2165 posts
  • Other
  • Chicago, Illinois

Posted 16 July 2006 - 10:25 PM

Thanks Laurent and Charles.

I have used that 8mm a number of times since it was serviced, and it was always tack sharp, but it was always stopped down to 5.6 or so. Maybe it is just not a sharp lens wide open.

I will do some test charts with it to see how it does. The monitor in the shot is an LCD monitor, and I thought it would not be a problem, but the white letters on the black background really blurred out with the 8mm lens, compared to how it behaved with the 10mm end of the zoom lens.

Thanks again for the help,
-Tim
  • 0

#6 Mike Welle

Mike Welle
  • Basic Members
  • PipPipPip
  • 125 posts

Posted 16 July 2006 - 10:57 PM

I don't see a huge problem. But if it is a concern you can either solve the problem practically or ideally. Practically costs nothing. For the time being, simply stop the lens down and don't worry about it. Ideally, when the time and funds are present, you could investigate the flange focal depth to ground glass vs. gate (you may have already done this). If this is possibly the issue and, there is any discrepancy, depth of focus could be an issue--especially given the fact that you are using a somewhat extreme wide angle lens (8mm) at a wide open setting T2.1. I've had problems in the past with my Optar Illumina 8mm T1.3 because the lens was wide open and the flange focal depth/ground glass distance of my camera was set incorrectly. I didn't have major problems with 12mm, 16mm, 25mm or 50mm because, as you know, depth of focus becomes less critical at these focal lengths. And once these lenses were stopped down even a little, depth of field overcomes the problem. I don't use any ND when I'm filming (but I plan to in the future when I get around to it)--so that may account for middling T-stops. The 9.5mm and 8mm really test the optical setup of the camera, especially when wide open. Of course, didn't you say in an earlier post that your camera was set up perfectly in terms of flange focal depth?

All the best,
Mike Welle

Anybody?

-Tim


Edited by Mike Welle, 16 July 2006 - 10:59 PM.

  • 0

#7 Tim Carroll

Tim Carroll
  • Sustaining Members
  • 2165 posts
  • Other
  • Chicago, Illinois

Posted 16 July 2006 - 11:09 PM

Thanks Mike,

This was not shot with my regular Arriflex 16S camera. It is another one that I just picked up and need to do some work on. You may be right, it could be the FFD and GG are off, but I was surprised that if it were as off as the images indicate, that it did not effect the 10mm any more than it did. I certainly understand depth of focus, but I don't know as much about lens optics as I wish I did. I did not believe the 8mm was that much different from the 10mm as far as it's depth of focus goes, when they are both wide open. I got to find a chart that gives all that information.

Thanks,
-Tim
  • 0

#8 Daniel Stigler

Daniel Stigler
  • Basic Members
  • PipPipPip
  • 177 posts
  • 1st Assistant Camera
  • Berlin, Germany

Posted 16 July 2006 - 11:23 PM

I also think the problem might be the open aperture. A while ago i worked on a low budget show. We shot with an SRI and old Zeiss Standard primes. The stuff we shot at day (T4 - T8 i would say) came out fine, while everything shot at night (T2) looked pretty soft. I'm sure this wasn't due to poor focusing...
  • 0

#9 Tim Carroll

Tim Carroll
  • Sustaining Members
  • 2165 posts
  • Other
  • Chicago, Illinois

Posted 16 July 2006 - 11:37 PM

Found the depth of focus formula in the American Cinematographer Manual. The difference between the two lenses above, the 10mm at T3.1 has a depth of focus of plus or minus .015 mm and the 8mm at T2.2 has a depth of focus of plus or minus .009 mm, so it is significantly different.

More testing will tell.

Thanks all,
-Tim
  • 0

#10 Stephen Williams

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

Posted 17 July 2006 - 03:55 AM

Hi Tim,

One thing to remember with a zoom set to 10mm v a 10mm prime is the image sizes will not match unless you back off the camera with the zoom. You have to measure to the lens front nodal point not the film plane!
DOF is also calculated from the front nodal point.

Stephen
  • 0

#11 Laurent Andrieux

Laurent Andrieux
  • Sustaining Members
  • 1527 posts
  • Cinematographer
  • France

Posted 17 July 2006 - 06:56 AM

Stephen, do you mean you cannot use the film plane mark on the camera as to pull focus with a zoom or just with this particular one ? :blink:
  • 0

#12 Tim Carroll

Tim Carroll
  • Sustaining Members
  • 2165 posts
  • Other
  • Chicago, Illinois

Posted 17 July 2006 - 10:08 AM

Hi Tim,

One thing to remember with a zoom set to 10mm v a 10mm prime is the image sizes will not match unless you back off the camera with the zoom. You have to measure to the lens front nodal point not the film plane!
DOF is also calculated from the front nodal point.

Stephen


Stephen,

What is the front nodal point on a zoom lens? I am familiar with the nodal point on a lens being where the light rays cross as they pass through the lens, but you are saying that in a zoom there is a "front" nodal point and a "back??" nodal point? And how is one to determine where these are in the lens?

And does this mean that a zoom set at 10mm is not going to have the same circle of confusion and same depth of focus as the 10mm prime you were referring to? Would the depth of focus be greater, which would explain the image results above?


Thanks,
-Tim
  • 0

#13 Max Jacoby

Max Jacoby
  • Sustaining Members
  • 2955 posts
  • Other

Posted 17 July 2006 - 11:17 AM

The issue with zooms vs primes is that in order to calculate the depth of field, one needs to take the physical length of the zoom into consideration. With a prime lens the nodal point is very close to the film plane, but with a zoom it is further away and becomes an issue.

To give a simplified example, if your focus is set at 4 feet, but the zoom is 1 foot long, then you will get a more accurate depth of field estimate if you calculate it for 3 feet.

A related issue is that of focal lenght that Stephen mentions. Because zooms are longer than primes, you will get a narrower field of view if you switch from a prime to a zoom of an identical focal lenght and leave the camera in the same position.
  • 0

#14 Laurent Andrieux

Laurent Andrieux
  • Sustaining Members
  • 1527 posts
  • Cinematographer
  • France

Posted 17 July 2006 - 12:51 PM

And this is why you have different settings on the the tables, according to the fact it's a zoom or a prime lens.
  • 0

#15 Stephen Williams

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

Posted 17 July 2006 - 05:06 PM

Stephen,

What is the front nodal point on a zoom lens? I am familiar with the nodal point on a lens being where the light rays cross as they pass through the lens, but you are saying that in a zoom there is a "front" nodal point and a "back??" nodal point? And how is one to determine where these are in the lens?

And does this mean that a zoom set at 10mm is not going to have the same circle of confusion and same depth of focus as the 10mm prime you were referring to? Would the depth of focus be greater, which would explain the image results above?
Thanks,
-Tim



Tim,

Some books mention front entrance pupil. I was not sure how to spell entrance! I am not on my PC with an English spell check at the moment. Zeiss will have the data for your lens.

A zoom set at 10mm will be more enlarged than a prime lens if you just change the lens. The DOF will be less in this case because of image magnification and could partially explain the DOF issue.

Stephen
  • 0

#16 Stephen Williams

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

Posted 18 July 2006 - 06:07 AM

Tim,

The front nodal point moves as you zoom! Sorry I forgot to mention that.

Stephen
  • 0

#17 Peter Egan

Peter Egan
  • Basic Members
  • PipPip
  • 24 posts
  • Other

Posted 19 July 2006 - 07:13 AM

Hi Tim,

Sorry, nothing substatial to add to the topic, just wanted to say I'm looking forward to seeing the full tests. Do you know when you'll be posting them?

Cheers,
P.
  • 0


CineTape

Broadcast Solutions Inc

Abel Cine

CineLab

Gamma Ray Digital Inc

Tai Audio

Media Blackout - Custom Cables and AKS

The Slider

Opal

Glidecam

Metropolis Post

FJS International, LLC

Wooden Camera

Willys Widgets

Technodolly

Aerial Filmworks

Rig Wheels Passport

Visual Products

rebotnix Technologies

Paralinx LLC

Ritter Battery

Gamma Ray Digital Inc

Metropolis Post

CineTape

rebotnix Technologies

Wooden Camera

Visual Products

Technodolly

Ritter Battery

Media Blackout - Custom Cables and AKS

The Slider

CineLab

Rig Wheels Passport

Paralinx LLC

Willys Widgets

Abel Cine

FJS International, LLC

Tai Audio

Opal

Broadcast Solutions Inc

Aerial Filmworks

Glidecam