Cinematography Mailing List - CML

Lens Testing


Hey All:

What kind of lens test do ya'll shoot when prepping for a feature? I usually do a reg test and a rack leader, but I'm wondering what other procedures are in existence. Thanks.

Best Wishes,
Michael Bratkowski
All round Camera Bloke
Los Angeles, CA

>>What kind of lens test do ya'll shoot when prepping for a feature?

When I get a final list of the focal lengths the DP wants, I ask the rental house to collect three of each focal length for me for when I get there to do the prep.

I then shoot a lens sharpness test, at whatever T stops the DP wants, often T2.8 and wide open, for example, of a test chart or several charts taped to the wall, so that there is something dead center and something in each corner and in other places in the frame, for each of the three lenses of that focal length, one after the other, all labelled with the serial number of the lens on a card on the wall so it is visible in the frame.

Instead of shooting a single chart, changing the distance for each focal length (although some AC's do it this way), I prefer to shoot all the tests at a set distance, like 8 feet, and tape up copies of many charts to a blank wall, filling the center and the corners. To me, 8 feet is a much more likely shooting distance than the 3 feet to fill the frame with a single chart with a wide angle lens. It also makes the lighting and shooting much simpler and faster to shoot everything from one position. Just personal preference, the other way works fine, too.

We shoot the three 20mm lenses, then the three 24mm lenses, then the three 27mm lenses, etc., about 20 feet of film (35mm) on each lens, at each stop, so that when screened, we will see the three lenses in sequence for long enough to examine the center and the corners, and be able to easily pick out the best of the lot, by serial number, which is on the camera report as well as on the screen, of course. Three of each focal length gives us something to compare with. Hopefully the rental house has that many available. If not, we make do with what is available.

For the zoom lenses, we start at full telephoto, shoot a few seconds, then slowly zoom out while the camera runs, to full wide, hold a few seconds, then back in to full telephoto. The same with the next lens, and the next. This shows the overall quality of the full zoom range, as well as the tracking error.

This lens sharpness is something the Camera Assistants can shoot themselves, without the DP present, but I always phone the DP, first to ask what stop or stops he wants us to shoot the lenses at, and then after, to ask when we can schedule a screening (usually at the lab) so he can come and participate in the selection of the best lenses. Most DP´s want to be present for that, but not for the actual shooting of the test. The Camera Assistants should also be able to shoot the registration test and the racking leader (aka "ground glass markings leader" or "projection leader") without the DP present.

Of course, if there are film emulsion tests to be shot, or filter tests, or makeup, hair or wardrobe testing to be done, the DP will have to be present for all of those.

Once the DP has selected the sharpest lenses, the Camera Assistants go back to the rental houses, separate out the chosen lenses from the lenses being returned to stock, and begins checking the lens focus markings on the chosen lenses, adding "in-between" focus markings where necessary.

I hope this helps.

Doug Hart

> I hope this helps.

Great post, Doug. Thank you, Geoff (et al) for making this place so great!

David Perrault, CSC

>What kind of lens test do ya'll shoot when prepping for a feature?

Besides the excellent reply of Doug Hart, I would like to add some other tips. I found it depends on the camera equipment list, the rental house, the testing facility at one's disposal, the screening options, time allotted and last but not least wether the DoP can be around or not.

It is a bit of a mixture of established testing procedures, and what seems a more going trend. All in all testing evolves as the gear does. I learned testing from the people I assisted combined it with reading books like Doug Hart's "Camera Assistant", David Elkins "The camera assistant manual" or John Fauer's books containing excellent testing tips like the "16 SR Book" very useful at the time. Also when possible I like to compare with testing happening next door to our booth. Sharing information is very useful you can always pick up a trick or two. The archives on cml contain a lot of info on the matter. That is for all the tests involved.

As far as lens testing for a feature is concerned.

If working with primes of the latest generation, (e.g. Cooke S4-Ultra Primes and so on) I would start test only one lens of each. With not so new gear (e.g. Superspeeds), when possible, I would follow Doug Hart's description; as in testing several identical lenses at the same time, but this becomes more complicated for various reasons. (see the first line) With older zooms I become more picky. Testing is then very much to Doug Hart's description.

Also before shooting any test I check for the mechanical condition of the lens : aperture ring focusing ring and so on.

Check for infinity setting, quick distance mark, breathing. With a zoom I will check for tracking. A good visual test might saves time at a later stage. At the very beginning when not so much experienced, I tested a zoom and later realised that it was not mechanically sound. I had to shoot a test again cause the zoom had to be swapped. If in doubt with a particular lens I would check and compare on the projector with special charts first, prior to shooting the test. I find it a quick and reliable way. For instance you can very quickly see how the sharpness evolves or not when stopping down. In doubt I would talk about it with the lens technician during the projection. Unfortunately this is not around as often or one does not have access to these.

I usually test on two distances and two separate stops (full open and given shoot stop or two stops down). Having a large mobile test wall makes this fast and smooth, on a dolly saves time to; if not, oh well it is hopping around like mad for a couple of days (it's a good warm up before the shoot :-P ).

Depending on time, I will also should into the light to see how the lens reacts on contours, flares (but there is no guarantee). I will do this on TV-series when the canon zooms are listed.

Not unusual also to test various types or make of lenses and so compare their qualities. Especially when new lenses are available, sometimes weeks before a job even gets going.

After the lens test is done, then you have to test for the "compatibility test" of the lenses with all accessories. The right mattebox for the right lens, adapter rings, filter holders and so forth. Sounds obvious but I have made that mistake in the past, learned from it. Believe me I witnessed it quite often. No point of having a top lens if you can use it the way it should afterwards. Everything has to be tested and must fit. Becomes all the more important when rental house is thousands of km away.

If you are working with a good reputable rental house on a regular basis, you can skip some basic steps if time is not on your side as the equipment will have been checked thoroughly up to several days prior to the assistants arrival. The lens technician often knows the DoP's demands and the assistants habits. If you end up in a place for the first time, and you are working for a "non regular production house" type customer (or even worse on a tight package) the testing might not be so elaborate or as smooth as one would hope. When possible I go to the rental house in person before, rather than call. It creates a good working relationship with the technicians. Whatever the relationship with the rental house, test. It is not a question of trust-mistrust but of responsibility. Also if the production manager wants to save money on the testing, make him understand his responsibilities.

When shipped gear has arrived and other tests as make-up, emulsion tests are being conducted I will try a quick sharpness test besides standard registration test again.

There is also the slop test. Have only done very few of those, but sometimes the shape in which the boxes came on arrival made us do so, and as no processing facility close at hand there was no other choice.

BTW The first three features I worked on as clapper loader it was with French assistants. Now quite some French have a very academic way of testing their lenses, but I will leave it up to them to explain it.

My two eurocents,

Emmanuel, Munich

Assistant Caméra - Camera Assistant - Kamera Assistent
BVK- European based
Mobil# Munich +491608036889 - London +447910034443 (please use German number)

Please be careful with your exposure when shooting test charts. An overexposed chart will look soft even though the lens is good, the line pairs "bleed" into each other and makes it difficult to assess sharpness. My rule of thumb is to underexpose charts by about 1/2 - 2/3 of a stop.

Saves wear and tear on the Rental lens 'tech too...

Andy Taylor
Camera and Lens Engineer
Arri Media
3 Highbridge
Oxford Road
Middlesex UB8 1LX
01895 457141

Thank you very much, Gents, Doug and Emmanuel especially, for the excellent in depth data on Lens Testing.

I've been under going a stretch where the marks seem to be short (i.e. the index mark wants to be at 3'10 instead of 4') and this always seems to be at low budget houses. I'm starting to consider shooting a ruler of sorts, with a number of siemen stars at two inch distance from each other and then seeing which scales photograph correctly and if tape or eyefocus is more productive.

As for the Cinetape, I actually built my own SonarTape, 'cause I find the Panatape and the Cinetape too slow for my tastes. I use it religiously, finding it useful for rehearsals to "map' the actor's motions out and previewing the take. It's not good to pull from it, as you'll be behind the motion, but it's great to back check yourself.

Just make sure to always double check it with a tape
measure, these things can get bumped. B^)

Michael Bratkowski


© copyright CML all rights reserved