Filter over Light Meter?

bvy · Dec 11, 2016

I'm shooting large format with a blue filter, sometimes a green filter, and I'm using Travelite 750 strobes in an indoor studio. I believe the bulbs are UV-coated, but my question is a general one. If I'm placing a filter over the lens, does it make sense to take meter readings with the filter over the meter?

480sparky · Dec 11, 2016

I'm assuming you're referring to a hand-held meter as opposed to an in-camera one.

The deduction for a given filter should already be known. So placing the filter over the meter would merely be an unneeded step.

bsdunek · Dec 11, 2016

One problem is that the meters don't see color the same as the film or we do. The difference may not be a lot, but it is far better to take the light reading and apply the filter factor.

480sparky · Dec 11, 2016

bvy said:
I'm shooting large format with a blue filter, sometimes a green filter, and I'm using Travelite 750 strobes in an indoor studio. I believe the bulbs are UV-coated, but my question is a general one. If I'm placing a filter over the lens, does it make sense to take meter readings with the filter over the meter?

If your meter says to shoot at f/16 and you have a 2-stop filter, then you can figure out that you need to shoot at f/8 a lot quicker than the time it would take to remove the filter off your lens, hold it up to the meter, then reinstall it over the lens.

bvy · Dec 11, 2016

I wasn't very clear. I'm not proposing this as routine practice. Rather, the filter factor is what I'm trying to arrive at. These are cheap filters, and there's nothing inscribed except "Telesar 62mm Blue - Japan." Even so, I'm wondering if the UV-coating reduces blue light to the extent that even the true filter factor would need to be compensated for.

I'm using a Sekonic (hand held) meter.

markbarendt · Dec 11, 2016

Yes bvy, it's a fine idea, even as a routine practice.

The UV coating isn't a big enough deal to worry about IMO. The colored filters can be though, a little experimentation will be needed and you may find that an adjustment to your EI helps.

williaty · Dec 12, 2016

Yes, I do it all the time.

Filter factors are basically useless. They give a vague generalization of how much to increase exposure that varies anywhere from perhaps somewhat close to wildly inaccurate depending on how your scene differs from the one used to determine the official filter factor. I don't care what that filter does to the test engineer's scene back at the factor. I care what it does to my scene in the specific areas I'm metering. Placing the filter over the lens of the light meter achieves what I need to know.

Typically, I meter numerous points around the scene without the filter to get a look at how the exposure values of different areas relate to each other while thinking about how that's going to look in the final print. If the exposure values between two areas are either too far apart or too close together to easily produce the tonal separation I want in my final print, I start thinking about which filter I can use to push the important areas of the photo into a relationship that works better for the way I want my final print to look. Then I re-meter the scene with the filter I think will work held over the front of the light meter. I confirm that the exposure readings have moved into the relationship that I was intended. Then I meter the single most important thing in my scene through the filter and set the exposure that places that important thing in the tonal position I want it in the final print.

The most common scenario I run into is that the sky is too bright compared to what I want to do with the foreground. Typically, I can move the value of the sky darker while not doing anything problematic to the foreground by using a deep red filter. If I need to move the sky a LONG ways and the angles work out favorably, I often stack a polarizer and deep red filter to reduce the sky's value even farther. Typically, I'll then meter most important foreground shadow in which I still want full detail and set the shutter to 2-2.5 stops under that reading.

Gerald C Koch · Dec 12, 2016

What seems to have gotten lost is that filter factors are not only a function of a particular filter BUT also of the type of film. Manufacturers publish filter factors based on the particular sensitivity of their film. Putting a filter over your light meter ignores the role of the film in determining exposure.

http://wwwuk.kodak.com/global/en/professional/support/techPubs/f4017/f4017.pdf

Since the OP does not state what film he is using I include data for the most popular film. In the above example look at page 3 for a table of filter factors.

The OP is also ignoring the latitude of the film he is using. Stick with the film manufacturer's recommendation and you should be fine.

BTW if your images are important you might consider purchasing a better grade of filter. Remember it is being inserted into the image path and cheap ones can causes distortions.

Brook Hill · Dec 12, 2016

When I just relied on ttl reading with the filter attached I tended to under expose. I now adjust the ttl reading to bring it up to the manufacturers filter adjustment and I get much better negs which are easier to print. Metering through the filter gives a smaller difference in exposure than applying the filter factor. Which is correct? Surely the anwser is the one that gives you the negs which are easier to print.

Tony

bvy · Dec 12, 2016

Gerald C Koch said:
Since the OP does not state what film he is using I include data for the most popular film. In the above example look at page 3 for a table of filter factors.

The OP is also ignoring the latitude of the film he is using. Stick with the film manufacturer's recommendation and you should be fine.

BTW if your images are important you might consider purchasing a better grade of filter. Remember it is being inserted into the image path and cheap ones can causes distortions.

I'm using instant film -- specifically, the Impossible Project 8x10 film. They don't publish data sheets (I'd love to see spectral sensitivity), so I'm having to do my own tests -- which I'm okay with, but the stuff ain't cheap. Latitude? Not much, in my experience.

And I'd love to find proper filters to fit over my Ilex Caltar 15" lens. The barrel (~85mm in diameter) has no threads, and all solutions I've read about are improvised ones. That said, I do plan to get some higher quality, larger filters to work with. What I'm doing now is just "proof of concept" to see if I like the look and can work with the increased exposure before investing more money.

DREW WILEY · Dec 12, 2016

I use filter factors instead, which I've tested with specific film and meter combinations. As many probably already know, Zone VI at one time marketed a modified Pentax spotmeter with an internal filter pack allegedly matched to the spectral sensitivity of panchromatic film itself, intended
to be used with contrast filters over the meter. But not even all pan films have the same specific sensitivity. My own standard of comparison has been
to meter TTL in-camera with attached contrast filters, then compare the results with simple handheld spotmetering (unmodified meter). The handheld
version is always more trustworthy. But of course, you gotta do the tests and have some experience on specific situations.

silveror0 · Dec 12, 2016

^^^ What he said.

Drew is right on, as usual.

Mr Bill · Dec 12, 2016

bvy said:
I'm using instant film -- specifically, the Impossible Project 8x10 film. They don't publish data sheets (I'd love to see spectral sensitivity)...

Hi, given the situation, including unknown specs on the filters, I don't see that there's much you can do other than meter through the filter.

I don't know how expensive the 8x10 "Impossible" film is, but I wonder if it might be worth doing a few tests on either roll or 35mm film to nail down filter factors on THAT film. (Basically you shoot a neutral test target with no filter, then do an exposure series to find the best match.) Then you would at least have a baseline factor for the filter.

Regarding the UV (or not) coated flashtubes, I've looked into this sort of thing before, on behalf of a large chain outfit. Offhand, I'd say that the overall exposure effect for a panchro film is near insignificant, on the order of perhaps a couple percent as a wild guess. I can probably look up some spectral curves and put a better number on it if you're worried about it. (Our issue was with white objects taking on a bluish tinge if they contained brighteners, not exposure differences.)

bvy · Dec 12, 2016

Mr Bill said:
Hi, given the situation, including unknown specs on the filters, I don't see that there's much you can do other than meter through the filter.

I don't know how expensive the 8x10 "Impossible" film is, but I wonder if it might be worth doing a few tests on either roll or 35mm film to nail down filter factors on THAT film. (Basically you shoot a neutral test target with no filter, then do an exposure series to find the best match.) Then you would at least have a baseline factor for the filter.

Regarding the UV (or not) coated flashtubes, I've looked into this sort of thing before, on behalf of a large chain outfit. Offhand, I'd say that the overall exposure effect for a panchro film is near insignificant, on the order of perhaps a couple percent as a wild guess. I can probably look up some spectral curves and put a better number on it if you're worried about it. (Our issue was with white objects taking on a bluish tinge if they contained brighteners, not exposure differences.)

Thank you but don't go to any trouble. The IP 8x10 is $20 per sheet; I've been using some smaller format (Spectra) frames to test, and that's working okay (at "only" ~$3 per frame). What I'm finding is that the meter over the filter tells me to use 50, but that's overexposing. I'm shooting it at 100 now, and that's looking better. (FYI, the box speed is 640.)

Poisson Du Jour · Dec 12, 2016

Irrespective of film, format or lens, filter factors as published and as recommended are reliable in use and do not warrant angst. If you have reason to think that the FF are wrong, start with the baseline recommendation and make exposures. Along with this, take detailed notes! In my teaching, people are surprised how well they pick up on small details (and hold them!) like this when attendant notes are taken as referrals!

A couple of points worth noting. There are some filters where the FF is variable (from a known start point) e.g. polarisers, and again, the FF varies considerably among brands. An example is Cokin's Pure Harmonie polarisers which have a surprisingly low effective FF of 1.2 compared to 1.5 to 2.0 for many others, depending on prevailing light, not film or film format. Some of B+W's KSM C-POLs have a heavy FF. Solid colour filters also do vary among manufacturers; the FF they publish should be taken as the guide.

Filters used directly on lightmeters are more common in cine photography. The tiny sizes for e.g. spot meters can be hard to find (and easily lost!). I have a red (FF=2.5) and UV(0) -- everything else is full-sized filters on lenses with established FFs entered into an L758D for spot/incident metering. In routine photographic practice the FF is nothing to worry about, other than taking it into account (with a spot meter, enter this in the FF compensation or additional exposure area) and if it is a polariser, indexing the FF to match prevailing illumination.

Gerald C Koch · Dec 12, 2016

bvy said:
I'm using instant film -- specifically, the Impossible Project 8x10 film. They don't publish data sheets (I'd love to see spectral sensitivity), so I'm having to do my own tests -- which I'm okay with, but the stuff ain't cheap. Latitude? Not much, in my experience.

Thanks for your reply which explains a lot.

Sirius Glass · Dec 12, 2016

Because of the different wavelength bandwidths of light meters, I have learned that while I can meter through the lens with an slr camera's light meter, that I am better off if I am using a handheld light meter to take the light reading and then adjust for the filter factor rather than meter through the filter.

Maris · Dec 12, 2016

I test for filter factors by actually exposing film in a series of stepped exposures. My "standard daylight scene" (houses, grass, sky, clouds, mid-grey road, etc) is only steps from my darkroom door. For example a #25 red filter turns out by experiment to have a filter factor of 3 stops if my red filtered mid-grey road is going to have a negative density matching an unfiltered mid-grey road.

But when I point my Pentax Analog Spotmeter at that road and put a red filter over its lens the reading drops by only two stops. Why? The meter is overly sensitive to red. It "thinks" there is more red light than there actually is. So when metering through that filter using that meter and exposing that film in daylight I just give a 1 stop "fudge factor" more than indicated. It's not hard (ok, a bit boring) to record "fudge factors" for all one's filters, films, and meters but the elimination of uncertainty is worth it.

silveror0 · Dec 12, 2016

Filter manufacturers sometimes publish two factors for a filter, one for daylight illumination and a smaller one for tungsten light. If the scene is illuminated by warmer light such as sunrise or sunset, the smaller factor is more appropriate.

DREW WILEY · Dec 13, 2016

The spectral sensitivity of Pentax and Minolta spotmeters peaks in the green, just like normal human vision. Therefore under most daylight situations,
a deep red filter generally needs three full stops of compensation as the filter factor. But it all depends on the specific film, specific filter, and actual
lighting color temperature. But for "typical" pan films in daylight, the popular 25 red with a 3-stop factor is "good enough for government work".
Go deeper into a 29 or paler into a deep orange, and that factor probably won't be accurate.

M Carter · Dec 15, 2016

Gerald C Koch said:
BTW if your images are important you might consider purchasing a better grade of filter. Remember it is being inserted into the image path and cheap ones can causes distortions.

I did some extensive testing with 4K video and raw digital images last year. What I found about filters and image sharpness/distortion - it take a real shitty filter to mess up an image. I ran through some cheap eBay, not-really-glass 52mm ND and pola, glass 4x4 ND, and chinese resin 4x4 ND. The cheaper filters gave me some color shifts, especially when you got into the .9 territory. But image sharpness wasn't an issue, at all, and I pixel-peeped, shot test charts and so on.

If you have a good lens and know the optimal f-stop? My testing really showed that things like B+W or Schneider filters for hundreds of bucks a pop are highly overrated as far as sharpness goes.

DREW WILEY · Dec 15, 2016

Filter quality can actually make a huge difference. What needs to be understood in this context is that film and digital devices sometimes differ in the
way they respond to multicoatings. Nobody spends hundreds a bucks a pop for top-end filters anyway, unless they're custom and utterly huge.
It more like a thirty to sixty dollar item for typical camera lenses. So basically, you own idea of a test was relatively worthless.

Filter over Light Meter?

Privacy & Transparency

Privacy & Transparency