Filters on b/w film

[BrianShaw]

There is common misunderstanding of the meaning of Filter Factor

I’m open to learning. Please teach us.

If all you are saying is that the published filter factors are based on specific assumptions that might not always be the same, I think we understand.

 

[AgX]

I did not say the factors I hinted at post #5 give gross errors, though they can. Actually I did not speak of "such extremes" at all.

You experience did not show bad experience with a given filter factor as by coincidence you just used them as the filter manufacturer intended them to be used: on rendering the same of a neutral grey object.
However the use of a filter goes beyond this...

 

[BrianShaw]

We’re either talking past each other or something getting lost in translation.

bye

 

[Sirius Glass]

@AgX is not really talking about the filter factor. He wants to control the effect of a red X with a red filter. While he has a choice of R23, R25, R29, R72, he wants to control the change in the appears of the red X in each filter. That is not directly controllable factor. One only has control of which filter they choose and after that the pieces fall where they may.

 

[AgX]

The problem is that the moment the topic filter comes up, people refer to filter factor. Moreover as a given thing, similar to a lens-extension factor.
But whereas a lens-extension factor always applies, this is not the case for the filter-factor. One has to contemplate about its meaning, and whether this applies in ones very use of that filter.

 

[eddie]

Filter factors aren't sacrosanct. They may be a good starting point for average usage, though. Using a green filter in landscapes as an example, there are various shades of green in the scene- evergreens, grass greens, leaf greens. Using a dark green filter will have a greater effect on evergreens, which are darker than grass green, which has more yellow in it. A light green filter will have more of an effect on grass (or leaves), which may have more yellow in them.
The only way to be accurate in what you've chosen to use your filter for is through careful experimentation and observation of the tonal changes the filters deliver.

 

[Pieter12]

The problem is that the moment the topic filter comes up, people refer to filter factor. Moreover as a given thing, similar to a lens-extension factor.
But whereas a lens-extension factor always applies, this is not the case for the filter-factor. One has to contemplate about its meaning, and whether this applies in ones very use of that filter.

Please explain what you mean by filter factor. My understanding is it is the factor given to the exposure so the balance of the scene is properly exposed, not how the filter affects certain colors in the scene.

 

[Sirius Glass]

The filter factor relates to the overall exposure not the rendition of a particular color relative to the other colors. @AgX is expecting too much.

 

[Doremus Scudder]

Well, to throw a wrench into this conversation...

When one understands in the theory behind how filters work, it's easy to see that there are many variables that will affect the "factor," which is basically there to prevent underexposure. As michael_r points out, erring on the side of overexposure when one recognizes a situation that may deviate from the average and require a different factor is good insurance against underexposure and won't change the effect of the filter (as long as densities don't block up on the film's shoulder).

But, as Sirius points out, applying a filter factor does not give you "control" over how different colors are rendered; it is the filter's characteristics that determine that.

The main problem with just applying filter factors, as I see it, is that one also has no good way to visualize how the different colors in a scene will be rendered relative to each other. Sure, we can expect that objects the same color as the filter will be rendered proportionally lighter than without the filter and vice-versa, but that isn't nearly precise enough to figure out how much separation there will be between different colors, density-wise, in the negative or final print.

One situation I'm confronted with often is that of green trees against a blue sky. I read the trees with my meter and the sky. My meter reading, without any filter, says that the sky is one stop brighter than the trees. But I want the opposite - darker sky and lighter foliage - so I reach for the green filter. But will it do the job? No way to really know if you just apply the factor and make the exposure. It's a crap shoot.

So, I do my best to quantify the filter effect by taking meter readings through the filter. Yes, I know that there are a plethora of variables that can make my readings inaccurate, however, it is the only way I know to get even an approximate idea of how the filter will affect colors in the scene. And, with some experience and a few tests, one can eliminate many of the variables that affect the meter reading through the filter.

I'm pretty confident when I read through the filters I use most that I can extrapolate to within less than a Zone how colors in the scene will be rendered. I know, for instance, that reading through a red #25 filter reduces my meter's response by about a stop, so I add that mentally. With a green #11 filter, my meter reads just fine, same with yellow #8 and #12 filters. The green #58 filter requires a bit of fudging, but it's closer than the #25. Etc.

So, back to my scenario: I get out my #11 filter, take a reading of that blue sky and the green foliage and find that, whoops, they both now read the same. I've darkened the blue sky and lightened the foliage with the filter just enough to create a tonal merger, which I don't want at all. So, I try other filters and combinations. Often, I'll end up adding a polarizer to the mix to darken the sky a bit more to get the effect I wanted. Or, I'll use a stronger filter, or I'll re-imagine the scene, or I'll give up and not make the shot at all.

If I had just applied the filter factor, I would have wasted a sheet of film...

Best,

Doremus

 

[Sirius Glass]

Well, to throw a wrench into this conversation...

When one understands in the theory behind how filters work, it's easy to see that there are many variables that will affect the "factor," which is basically there to prevent underexposure. As michael_r points out, erring on the side of overexposure when one recognizes a situation that may deviate from the average and require a different factor is good insurance against underexposure and won't change the effect of the filter (as long as densities don't block up on the film's shoulder).

But, as Sirius points out, applying a filter factor does not give you "control" over how different colors are rendered; it is the filter's characteristics that determine that.

The main problem with just applying filter factors, as I see it, is that one also has no good way to visualize how the different colors in a scene will be rendered relative to each other. Sure, we can expect that objects the same color as the filter will be rendered proportionally lighter than without the filter and vice-versa, but that isn't nearly precise enough to figure out how much separation there will be between different colors, density-wise, in the negative or final print.

One situation I'm confronted with often is that of green trees against a blue sky. I read the trees with my meter and the sky. My meter reading, without any filter, says that the sky is one stop brighter than the trees. But I want the opposite - darker sky and lighter foliage - so I reach for the green filter. But will it do the job? No way to really know if you just apply the factor and make the exposure. It's a crap shoot.

So, I do my best to quantify the filter effect by taking meter readings through the filter. Yes, I know that there are a plethora of variables that can make my readings inaccurate, however, it is the only way I know to get even an approximate idea of how the filter will affect colors in the scene. And, with some experience and a few tests, one can eliminate many of the variables that affect the meter reading through the filter.

I'm pretty confident when I read through the filters I use most that I can extrapolate to within less than a Zone how colors in the scene will be rendered. I know, for instance, that reading through a red #25 filter reduces my meter's response by about a stop, so I add that mentally. With a green #11 filter, my meter reads just fine, same with yellow #8 and #12 filters. The green #58 filter requires a bit of fudging, but it's closer than the #25. Etc.

So, back to my scenario: I get out my #11 filter, take a reading of that blue sky and the green foliage and find that, whoops, they both now read the same. I've darkened the blue sky and lightened the foliage with the filter just enough to create a tonal merger, which I don't want at all. So, I try other filters and combinations. Often, I'll end up adding a polarizer to the mix to darken the sky a bit more to get the effect I wanted. Or, I'll use a stronger filter, or I'll re-imagine the scene, or I'll give up and not make the shot at all.

If I had just applied the filter factor, I would have wasted a sheet of film...

Best,

Doremus

Good summary, Siriusly

 

[Alan Edward Klein]

Tmax data sheet from Kodak says to take meter reading through camera without filter then apply the factor Kodak provides for each filter type.

 

[Doremus Scudder]

Tmax data sheet from Kodak says to take meter reading through camera without filter then apply the factor Kodak provides for each filter type.

Kodak's recommendation is the basic, "standard" way to use filter factors. It will yield good results in most cases. Use this method first; if you're happy with the results, then there's no reason to change to a different way of applying factors.

All the other methods in this thread are refinements of the above with the goal of either modifying the factors for deviations from average in illumination, overall subject color, etc., or to better visualize the effect of the filter (e.g., the metering-through-the-filter technique I describe above). All these different methods require a better knowledge of color theory and the spectral response of your film and meter as well as quite a bit of testing to be useful at all.

Best,

Doremus