Home of Professional Cinematography since 1996

style="margin-bottom: 0"> 

class="style5" Testing HD Lenses

>Published : 17th May 2005

>Having followed the recent discussions on film/ video lens performance related to use on CCD and CMOS sensors, what do people consider the best way to test lenses for HD?

>Is it possible that using a projector may allow you to see edge to edge sharpness, contrast, MTF etc but may obscure sensor related aberrations?

>When I buy a lens I usually have the manufacturer handpick the best they have in stock. They test the lenses on a projector. Is it possible they could miss something?

>I always check out new lenses on a waveform. However with the introduction of SDI-DVI converters looking at lenses on a 23" display has been a revelation. It is very easy to see even the most subtle differences in quality. Often clarifying visually what the waveform is telling me electronically.


>John Chater
San Francisco

class="Paragraph">>What I am asking is can a projector tell you everything you need to >know about a lens destined for use on a CMOS or CCD sensor camera?

>There's always going to be some additional aberration introduced by the chip block, so projecting light through a lens isn't going to show you how the image will actually look.

>Putting an HD lens on a projector will show you how it will look if you put it on a projector, but not how it will look on an HD camera.

>Art Adams
Director of Photography
Film | HiDef | Video

>"Get the facts first, you can always distort them later." - Mark Twain


>Projecting the lenses is at least a place to start especially if you have lots to choose from.

>Getting a Rock Solid Bullet Proof Back Focus is also critical and must be done first.

>Hold the Lens up to the Light and look at the Elements for dirt or bubbles or scratches and color.
Look for any Glue holding the elements inside the lens with a Flashlight you should see none.

>OK I feel another off the top of my head list coming on...

>1) Roll through the Focus and Zoom and Stop looking for any kinks or binding feel it with your eyes closed and see with your fingers if there is any grit, that might be sticking. Feel for any dents in the followers and cams as you roll through and it shouldn't have any slop as you go back and forth.

>2) Put the lens on the Camera having properly tweaked the White Shading with no Lens so you start with an even Field. Mount the Lens and crank up the Chroma on the Monitor now shoot a perfectly evenly lit white Field. Paper or Wall white or light grey. Look at the Wide and look for color shifts, Corners edges top bottom etc. Most of this can be corrected with a Good White Shade but its good to know where your lens falls naturally.

>3) Slow Zoom, watch the Lens and the Shot (put an object in the shot, Flowers) look again for slight shudder or binding or skipping indicating a bad gear or dent in internal followers. Go back and forth check for Play and responsiveness.

>4) Zoom out wide, Rack Focus from end to end. Note how much breathing occurs, do the same on the Mid and Long end. Shoot a Siemens Star and look for good solid contrast between B & W when in "Focus" shouldn't be muddy

>5) Zoom out wide tilt camera up and down looking for distortion on the corners & edges and how acceptable the field consistency looks (Subjective)

>6) If its an ENG dial the Speed up to Fast and use the Motor to ZAP in and OUT and listen to how loud it is and does the Motor generate any Torque on the end that distorts the shot.

>7) This is Critical and a WFM Monitor must be used here. Close the Iris and slowly open to find where light begins to pass through f-16 or above or below. Now slowly open the Iris (FOV Wide, Med, Long) and look at the evenness of field. That is the flat line that is rising staying flat as it passes light through the lens to hit the CCD or CMOS sensor EVENLY.

>There will be a point on most lenses where they cannot evenly convey that light and fall off on the edges begins to occur, The Line Bends at the edges as it rises. Note where that point is at what F-Stop, now continue to open and note how bright the middle is compared to the edges. This "Fall off" or "Portholing" or "Vignetting" may be severe or may be very mild. You be the judge. Take the monitor and crank up the Contrast to be able to really see it there also. This is why Shooting WFO comes with Side effects that may be very objectionable especially if your Lens does not perform well at this.

>8) Pick a Good Middle stop f-5.6 lets say where you get a good even field. Now Slowly Zoom In and look for the Same Bend in the Field and note where it begins, Now continue and note the "Fall off" or "Potholing" or "Vignetting" at the very end of the Lens.

>9) Put the 2 together, Long end of the Lens, WFO, and look at how well your Glass holds up and conveys an even flat field of consistent level of light hitting the Capture medium.

>10) Shoot an Absolute Black Source (Cavity Black of DSC Labs Test Chart) Now open wide and see if the Flaring of the excessively bright scene lifts the Black. It should not. Great Internal Paint is Critical in minimizing this.

>11) Back to Semins Star or resolution chart and go WFO looking for Coloured edges on the Black Chips

>How much and what Color. Now rack focus back and forth through the chart, See any Color shift. That's "Chromatic Aberration" where light from all three colors is not passed through the elements evenly especially difficult on the edges when WFO. Look for good Center to edge Sharpness, Shoot Newspaper Classified Ads.

>NOW good news is that a good portion of the problems can be corrected by White Shading and Flare adjustments creating an even color field. But this won't be able to cure all the Portholing or Chromatic Abb or Center to edge Sharpness inconsistencies. Remember that you will always be choosing a Lens from the best available.

>The persons that came before you did the same so you will always be getting the best of the Leftovers, Always.

>So if you go through this process at a rental house Keep track By serial number so you don't get stuck later with previously rejected glass. There are other ways of testing but this is what I have founds works and can be done in Prep without lots of Diagnostic equipment.

>B. Sean Fairburn
Director/ Cinematographer
Role Model Productions LLC

>John Chater asks about testing HD lenses with a projector:

class="Paragraph">>...what do people consider the best way to test lenses for HD?....Is it >possible that using a projector may allow you to see edge to edge >sharpness...

>I'm thinking you better have a damn good projection source, like at a factory because in my opinion anything less would be akin to what we experience in theatres! Hundred thousand dollar projectors that are dirty, misaligned, out of focus and out of parallax.

>I like Sean Fairburn's mantra...test...test and then test some more!

>You can't beat testing a lens in real life situations and if you're bound for film-out world, then you gotta find a way to do a test film-out and get it projected at a creditable screening room where everything is calibrated in stone!

>Otherwise trust the company whose selling you a $30,000 HD zoom and hope they'll stand by any visual anomaly.

>A sidebar would be that since there are only a few lens makers out there I think it's more a matter of personal preference then a lens that's bogus!

>Allen S. Facemire-DP/Director
SaltRun Productions,inc. Atlanta/Norcross, GA
Creativity Just Flows Here...

>Let me also Add that ...

>Once you set your Back Focus via screw or friction ring grab the Back Focus ring and see if it moves, It should not.

>B. Sean Fairburn
Director/ Cinematographer

>Allen S. Facemire writes :

class="Paragraph">> I'm thinking you better have a damn good projection source.

>I'm not really asking about projection quality or questioning the value of testing.

>What I am asking is can a projector tell you everything you need to know about a lens destined for use on a CMOS or CCD sensor camera? Will a lens selected from testing a on projector necessarily end up being the best looking lens on a HD camera?

>Are the projectors used by video lens manufacturers different from the optical ones used to test film lenses?

>John Chater
San Francisco


>When you project an HD lens on a lens project you need to put a glass block to simulate the prism effect otherwise the projector is not valid for HD lenses.

>Michael Bravin
Chief Technology officer
Band Pro Film & Digital

>Let me also add a few things …

>1/. Check the focus mechanism for backlash. Focus on something in the middle of the focal range and then focus to infinity and back to your mark and then minimum focus distance and then back to your mark. The lens should hold it's focus at your mark. If not then the focus mechanism has backlash or slop, not good.

>2/. At night point the lens at a car headlights coming towards you. Look for loss of contrast and light bouncing around inside the lens. Flare-Veiling-Narcissism these are the names for what happens when the lens does not control for internal reflections also not good.

>Michael Bravin
Chief Technology Officer
Band Pro Film & Digital

FYI : Century Optics Test Projector has a HD B4 mount available, with the
correct glasses (two), per the Sony specification.

Ken Robings
Engineer, Schneider Optics, Century Division
Van Nuys, CA