Odd things going on with filter factors

Bill Burk said:

You make good points every time, Doremus, thanks.
True that's what I used to do...but I was often disappointed with thin density in the shadows on my negatives shot with red filter. And I could see the immediate difference in reading is 2 stops through filter when the published factor requires 3 stops.

That's what put the bug in my mind that tells me I "should" compensate by reading unfiltered, and then apply factor, instead of reading through and shooting as-is. But... now I rate my film speed lower than its "real" speed (I often de-rate 2/3 stops these days)... So, knowing I already have the safety factor... with a lower film speed rating, red filter reading through the filter will result in a reading that is really correct.

Click to expand...

Bill,

It seems to me that you've found your factor for a #25 filter and that meter, so you could read through the filter and just add a stop and arrive at the same place as if you'd applied the filter factor.

That said, it's actually more complex than that. Film responds differently to strong red illumination that to a full spectrum (and to strong green, blue, etc.). There is a change in sensitivity and a change in contrast. These vary from film to film. A #25 filter with TXP tends to increase contrast, so I indicate less development. The opposite seems to be true for TMY (I haven't got it 100% nailed down yet, but it looks like either no contrast adjustment or perhaps a bit more is needed).

These things happen whether you apply the filter factor or read through the filter. Taking things to the next level, one would test at least the most extreme filters they use with the films they use and arrive at development adjustments for them if needed (that's what I've done). This, coupled with factors for the same stronger filters when reading through them yields an exposure compensation and development adjustment for those filters. My example: with TXP and a #25 filter I give 2/3-1 stop more exposure than the meter indicates reading through the filter and then develop N-1 (or one contraction more than the normally indicated development).

Again, there are good reasons to read through a filter instead of applying a factor. For in-camera-meter users, speed is the main advantage (especially when using an auto-exposure mode); dial in the exposure compensation needed for your filter, if any, and just rely on the meter. For me, a ZS user, spot metering through the filter allows me to look for tonal mergers that might otherwise escape me as well as to more precisely place colored objects in the scene (or at least know better where they fall). I'm 100% aware that there are discrepancies in this approach due to the mismatch of film and meter spectral sensitivities, however, it seems to work generally and gives me some little bit more information to use in visualizing. This latter to me justifies the read-through-the-filter approach.

Best,

Doremus

Leigh B said:

Let's consider what a color filter is/does...

It starts its design life as a neutral-density filter, attenuating all colors equally.
Then it's "modified" to attenuate its nominal color less. That has the effect of brightening that color. ...

- Leigh

Click to expand...

Leigh,

Not to beat a dead horse here, but reading through the meter is the only way currently available to us to quantify, even approximately, just how much a color is "brightened" or "darkened" by the use of a filter. Otherwise it's just guesswork: will that red rose be rendered one stop brighter? two stops? more? Will the blue sky go completely black? or will it be brighter than that? and how much? and will it merge with that oxidized copper rooftop? Reading through the meter is the only way to get any idea of this.

Best,

Doremus

I have yellow, orange and red and polarizing filters. I'm shooting medium format 6x7, Tmax 100, landscapes mainly. I use a separate meter and take either incidence or 10% angle reflectance readings depending on the lighting conditions. I don't read the meter through the filters. I don't develop my own film. The pro shop uses Xtol "flat". (I could get + and - development adjustments at extra cost. But I wouldn't know what to tell them frankly or why.) I usually have them provide a contact print. All printing would be from scanning and digital printing in most cases although I'm open to wet printing as I use to do with a lab many years ago.
1. What recommendations would you make for applying filter factors for both reflective and incident readings?
2. I usually bracket +1 and -1 stop. Any thoughts on changing that?
3. What factors when using polarizing with one of the contrast filters?
4. Any suggestions for the developing lab regarding + and - or anything else?
5. When shooting color, usually Velvia 50, would you use different factor for the polarizer filter and bracketing?

Peter Schrager said:

Meter

Paul-the ONLY spot meter you can use is the ZoneVI modified meter. I regularly use mine through the filter to make the reading and it is rarely off. Filter factors?
I leave that for others to figure out. Also saves alot of time. In reality how many different filters is one using? So I think it can be pretty much memorized if you standardize on 1 film.
Best, Peter

Click to expand...

interesting;the modified meter ever worked for me;the unmodified meter,on the other hand, works just fine

Doremus Scudder said:

Leigh,
Not to beat a dead horse here, but reading through the meter is the only way currently available to us to quantify, even approximately, just how much a color is "brightened" or "darkened" by the use of a filter. Otherwise it's just guesswork: will that red rose be rendered one stop brighter? two stops? more? Will the blue sky go completely black? or will it be brighter than that? and how much? and will it merge with that oxidized copper rooftop? Reading through the meter is the only way to get any idea of this.

Click to expand...

Sorry, Doremus, but there's a fundamental error in your analysis.

The filter factor provided by the manufacturer is correct, based on extensive testing and knowledge of the filtration medium used.

The problem you have in applying it to a particular real subject is that your analysis of the colors is not accurate.

- Leigh

For decades I have applied the published filter factors to all of my work (multispot meter/incident and transfer to the camera) and never got my knickers in a knot over the result, which is to say, perfect: the photograph was made, printed, framed and often sold. With my trusty EOS 1N, I leave filter factors to the onboard meter, but I can see that for a polariser it varies between 1.5 and 2.0, depending on the illumination prevalent on the scene (lower FF is for soft, diffuse, light-shadow illumination; upper FF is very flat, dull and shadowless illumination, commonly found in the rainforests where I work).

There is very rarely any cause to deviate from the published filter factor range that manufacturers supply. Their figures arrive from the science of testing, not populist opinion.

There are striking parallels between Alan's MO and my own work...

Alan Klein said:

I have yellow, orange and red and polarizing filters. I'm shooting medium format 6x7, Tmax 100, landscapes mainly. I use a separate meter and take either incidence or 10% angle reflectance readings depending on the lighting conditions. I don't read the meter through the filters. I don't develop my own film. The pro shop uses Xtol "flat". (I could get + and - development adjustments at extra cost. But I wouldn't know what to tell them frankly or why.) I usually have them provide a contact print. All printing would be from scanning and digital printing in most cases although I'm open to wet printing as I use to do with a lab many years ago.
1. What recommendations would you make for applying filter factors for both reflective and incident readings?
• Baseline FF published by the filter manufacturer or established through your own experience (not the experience of somebody else).
2. I usually bracket +1 and -1 stop. Any thoughts on changing that?
• None. This is standard professional procedure; I bracket and I record notes; it ensures I come away with 1, but more often 2 of the best images
3. What factors when using polarizing with one of the contrast filters?
• +1.5 to +2.0; refer to my remarks above. Polarisers are responsible for stuffing up more photographs than any other filter, so it is worth nutting them out!
4. Any suggestions for the developing lab regarding + and - or anything else?
• No. Not if you are shooting RVP50. If you are working B&W, there is a whole range of adjustments you can order, more still with sheet film. But are those adjustments valid?
5. When shooting color, usually Velvia 50, would you use different factor for the polarizer filter and bracketing?
• Answer 3., above. The FF does vary with POL. Whether you want to add additional ( + ) compensation is entirely up to you and your understanding of how RVP50 interprets the scene.

Click to expand...

Leigh B said:

Sorry, Doremus, but there's a fundamental error in your analysis.

The filter factor provided by the manufacturer is correct, based on extensive testing and knowledge of the filtration medium used.

The problem you have in applying it to a particular real subject is that your analysis of the colors is not accurate.

- Leigh

Click to expand...

Leigh,

There is no error in my analysis. Filter factors for black-and-white filters given by manufacturers are an average based on average panchromatic film sensitivity and some standard illumination (usually daylight, 5600K). Since these parameters vary in practice, the actual factors for any given situation that deviates from the "standard" are approximate. You can call this "correct" if you like, but it doesn't change the fact that the actual filter factor can be markedly different from the published "average" if the illumination and/or film's spectral sensitivity vary from the standard that was used to determine the filter factor. Note that B+W in their filter catalog lists all their factors, with the exception of neutral density filters as: "It's filter factor is approximately XXX." Tungsten filter factors are not even given in most cases these days, even though factors for some filters will vary widely from their factors in daylight (e.g., strong red or green filters).

I have no doubt that reputable filter manufacturers base their published factors on extensive testing. The problem is, is that real-life situations often vary from industry standard, so the factor for those situations will vary as well.

Furthermore, simply applying a filter factor makes absolutely no attempt to analyze the colors in the scene and compare their reflectances in relation to their transmission (or lack thereof) through a colored filter. It just guesswork and hoping that colored objects will be rendered how we wish. This is the usual way of using filters.

At least I attempt to quantify things a little. Metering through the filter is the only way I know of coming even close to comparing relative transmission through colored filters and, even with its discrepancies due to mismatches of spectral sensitivities between film and meter, the only way to base exposure on a strongly colored area when using a filter.

Best,

Doremus