Converting EV to Lux to millilux-seconds

[MattKing]

The meters do what they do in the way that they do it.
The reason that incident meters come up is that part of what you are talking here involves how much light is hitting (incident on) your subject.
If a reflected light meter is being used in some way where that level of incident light is important, you have to talk about how that reading relates to the level of incident light.

 

[RalphLambrecht]

I'm confused. A millimeter is 1/1000 of a meter (the prefix "milli" meaning 1/1000 of...), but a millilux is 1000x, not 1/1000 of a lux?...

No, 'milli' (as in mm) always means 1/1,000 or 10^-3

 

[Sirius Glass]

No, 'milli' (as in mm) always means 1/1,000 or 10^-3

"kilo" as in kilogram means 1000 or 10+3

 

[Bill Burk]

The Pentax spot meter is a reflectance meter, not an incident meter.

I’m thinking of what the meter is coming up with EV 12 2/3 is an estimate of the light incident on the scene - based on a reflected light measurement for sure… but the estimated Light Value is an incident measure.

 

[Nodda Duma]

I think you’re right, but maybe we’re supposed to be squaring the 1,000 to get to milli since these are area-light

Multiply by 1000 to convert from Lux to millilux.

Same as multiplying by 1000 to convert from meters to millimeters.

 

[Bill Burk]

Multiply by 1000 to convert from Lux to millilux.

Same as multiplying by 1000 to convert from meters to millimeters.

What about meter candle seconds to millimeter candle seconds?

 

[Chuck_P]

based on a reflected light measurement for sure… but the estimated Light Value is an incident measure.

You keep saying this Bill, but no explanation behind it. Why isn't it the estimated light value a reflected measure?

 

[MattKing]

You keep saying this Bill, but no explanation behind it. Why isn't it the estimated light value a reflected measure?

Because Light Value is a measurement of the intensity of light that hits a surface.

 

[Nodda Duma]

What about meter candle seconds to millimeter candle seconds?

Science nerd humor is the best.

 

[Bill Burk]

You keep saying this Bill, but no explanation behind it. Why isn't it the estimated light value a reflected measure?

Not trying to be obtuse or scientific. Just thinking like this... If you used an incident meter it would tell you EV 12 2/3 in the same light (well actually you did two things that affected readings, you used a white target and you placed the reading on Zone X).

But if you used the meters normally (and an 18% gray card with the spotmeter) you might see EV 12 2/3 on each dial. Which you might convert to the same millilux.

So they are probably the same kind of measure, and I think millilux is a measure of incident light.

 

[Stephen Benskin]

Chuck, this is from a post not too long ago. All exposure meters want to produce the same exposure at the film plane depending on the speed setting. The basic equation is 8/ISO which might include a slight rounding. One problem with hand held meters is that have to assume certain variables associated with the camera's optical system. A TTL meter reads the actual illuminance. So a slight rounding really won't hurt.

I believe Schaefer uses a density of 2.70 as an aim for his speed point exposure. This is good as you want some room to the left for the toe and for higher development. Determine the exposure needed to obtain a density of 0.10 for a given film speed. As the film speed equation is 0.80 / Hm = EI, where Hm is the exposure in mcs or lxs that has a corresponding negative density of 0.10 over Fb+f. To find the lxs for each film speed use 0.80 / EI.

100 = 0.0080 lxs
125 = 0.0064 lxs
400 = 0.0020 lxs

The required exposure would be

For a 100 speed film, the exposure would be 4 lxs to obtain a value of 0.0080 lxs at the 2.70 density step.

100 = 4 lxs
125 = 3.2 lxs
400 = 1 lxs

Confirming

where Density= 2.70.

100 = 0.0080 lxs
125 = 0.0064 lxs
400 = 0.0020 lxs

The 64 thousand dollar question always comes down to whether the available equipment is capable to achieving this level of precision. If the answer is no, then there will be a level of inaccuracy introduced into the testing.

A camera and meter might be able to achieve something closer to this level of accuracy, even considering all the variables of an optical system, and questions regarding the calibration of the f/stops, shutter speed, and meter. The best possible work around would be to take the meter reading at the film plane, but most people don't have that type of meter.

So what values we have if the exposure is based on a metered target and the exposure is adjusted by opening up five stops form the metered reading? Will it be the same or different from what was previously determined? Exposure at the film plane for Hg is 8/ISO. This is what meters are calibrated for.

100 = 0.080 lxs
125 = 0.064 lxs
400 = 0.020 lxs

Open up 5 stops 25 * (8/ISO)

100 = 2.56 lxs
125 = 2.05 lxs
400 = 0.64 lxs

You've probably noticed the illuminance values based on Schaefer's meter and open up 5 stops doesn't match the values required to produce the desired film speeds at the 2.70 density strip. For a 100 speed film, the density producing Hm would be 2.50 . The speeds are 2/3 stop off which would produce Zone System speeds.

 

[Kilgallb]

I'm hearing that my Pentax 1 degree spot meter is actually providing me with an incident reading..

A spot meter reads the light reflecting or emitting from a surface. Definitely not the incident light.

 

[Stephen Benskin]

I didn’t think that. A white target would certainly throw out my story point that an incident meter would agree with the spotmeter. But still the light meter was estimating the incident light according to what it wants to put on film.

I am sure interested in what took you down from 3,620,375.125 millilux seconds (your reading) to 2512 millilux seconds (my estimate).

Figuring it out involves K which we talked about but maybe never finished the class. I know I didn’t learn it well enough to be able to teach. All I can do is facilitate


p.s. The thread was “Is the K factor relevant to me or should I cancel it out?”

Bill, all the attachments from that thread are in the Defining K paper which is available on your site. There are also a couple of papers relevant to this discussion Calibration Levels of Film and Exposure Devices by D. Connelly, An Interpretation of Current Exposure Meter Technology by Allen Stimson, and Re-evaluation of Factors Affecting Manual or Automatic Control of Camera Exposure by James Scudder, C.N. Nelson, and Allen Stimson.

 

[Bill Burk]

Where we’re running into trouble is the conversions.

I think Chuck_P put 2.512 meter candle seconds at the film plane.

The light value reading converts to 14,400 lux. He placed the reading on Zone X. A quarter second exposure at f/7.1 was given.

How does a number that large come down to about two and a half?

 

[Stephen Benskin]

Where we’re running into trouble is the conversions.

I think Chuck_P put 2.512 meter candle seconds at the film plane.

The light value reading converts to 14,400 lux. He placed the reading on Zone X. A quarter second exposure at f/7.1 was given.

How does a number that large come down to about two and a half?

It's a little hard to figure since Chuck was using a spot meter and lux is illuminance, so I had to make a few assumptions. But this is basically how it might be working.

Lux reflected off an 18% target and then plug it in with the exposure equation.

and

It pretty much comes out the same if you go with the Ev and Pentex's chart in the Digital spotmeter manual. Ev 12.5 equals 859 cd/m2 which is close enough. Or it's just a coincidence.

 

[Bill Burk]

Excellent! Thanks Stephen!

Chuck_P … see it comes down from 14,400 to about two and a half.

 

[Chuck_P]

Not trying to be obtuse or scientific. Just thinking like this... If you used an incident meter it would tell you EV 12 2/3 in the same light (well actually you did two things that affected readings, you used a white target and you placed the reading on Zone X).

But if you used the meters normally (and an 18% gray card with the spotmeter) you might see EV 12 2/3 on each dial. Which you might convert to the same millilux.

So they are probably the same kind of measure, and I think millilux is a measure of incident light.

Ok, so, if millilux is a measure of incident light........then the conversion of a reflected light reading....to millilux....has to somehow speak to the intensity of the incident light striking a surface....that's if I understand you correctly.

Where we’re running into trouble is the conversions.

Bill............I typed the above response last night before your comment here regarding the conversions........but did not post it right away as I wanted to sleep on it. So my observation that the conversion of a reflected light meter reading to millilux seems consistent with your comment here. This morning I read all of Stephen's posts, and man! It's all so heavy and intriguing at the same time.......there's a lot to try and soak in, but I'm afraid my sponge is at the point is just about full.

A question about the graph paper you provided..........I have marked on the x-axis the calibrated step densities of my Stouffer step tablet. I can see that you have your calibrated density steps identified nicely. I have marked mine off in pencil, it appears yours were printed with the graph paper.......is there a way I can get mine indicated nicely like that as well.....how did you do that?

 

[RalphLambrecht]

Chuck, this is from a post not too long ago. All exposure meters want to produce the same exposure at the film plane depending on the speed setting. The basic equation is 8/ISO which might include a slight rounding. One problem with hand held meters is that have to assume certain variables associated with the camera's optical system. A TTL meter reads the actual illuminance. So a slight rounding really won't hurt.

I believe Schaefer uses a density of 2.70 as an aim for his speed point exposure. This is good as you want some room to the left for the toe and for higher development. Determine the exposure needed to obtain a density of 0.10 for a given film speed. As the film speed equation is 0.80 / Hm = EI, where Hm is the exposure in mcs or lxs that has a corresponding negative density of 0.10 over Fb+f. To find the lxs for each film speed use 0.80 / EI.

100 = 0.0080 lxs
125 = 0.0064 lxs
400 = 0.0020 lxs

The required exposure would be

For a 100 speed film, the exposure would be 4 lxs to obtain a value of 0.0080 lxs at the 2.70 density step.

100 = 4 lxs
125 = 3.2 lxs
400 = 1 lxs

Confirming

where Density= 2.70.

100 = 0.0080 lxs
125 = 0.0064 lxs
400 = 0.0020 lxs

The 64 thousand dollar question always comes down to whether the available equipment is capable to achieving this level of precision. If the answer is no, then there will be a level of inaccuracy introduced into the testing.

A camera and meter might be able to achieve something closer to this level of accuracy, even considering all the variables of an optical system, and questions regarding the calibration of the f/stops, shutter speed, and meter. The best possible work around would be to take the meter reading at the film plane, but most people don't have that type of meter.

So what values we have if the exposure is based on a metered target and the exposure is adjusted by opening up five stops form the metered reading? Will it be the same or different from what was previously determined? Exposure at the film plane for Hg is 8/ISO. This is what meters are calibrated for.

100 = 0.080 lxs
125 = 0.064 lxs
400 = 0.020 lxs

Open up 5 stops 25 * (8/ISO)

100 = 2.56 lxs
125 = 2.05 lxs
400 = 0.64 lxs

You've probably noticed the illuminance values based on Schaefer's meter and open up 5 stops doesn't match the values required to produce the desired film speeds at the 2.70 density strip. For a 100 speed film, the density producing Hm would be 2.50 . The speeds are 2/3 stop off which would produce Zone System speeds.

View attachment 330266

as said above: lux=2.5*2^EV and ego EV= log2 (lux/2.5) and remember to get the log to base'2' take the log to base '10' and divide that by log2to base'10' or in other words:
log2(x)=log10(x)/log10(2).

 

[Stephen Benskin]

Chuck, here is a graphic with the equations.

 

[Chuck_P]

Chuck, here is a graphic with the equations.

View attachment 330281

Thank you.....it does help.

 

[MattKing]

Chuck, here is a graphic with the equations.

View attachment 330281

This on a T-shirt would make the perfect present for Stephen

 

[Bill Burk]

A question about the graph paper you provided..........I have marked on the x-axis the calibrated step densities of my Stouffer step tablet. I can see that you have your calibrated density steps identified nicely. I have marked mine off in pencil, it appears yours were printed with the graph paper.......is there a way I can get mine indicated nicely like that as well.....how did you do that?

It’s done in InDesign, I could give you the file, or I could make a version for you with lines on your Stouffer scale readings.

 

[Chuck_P]

It’s done in InDesign, I could give you the file, or I could make a version for you with lines on your Stouffer scale readings.

Thank you.....I think I'd like to try the file.

 

[ic-racer]

Chuck, here is a graphic with the equations.

View attachment 330281

That is a really nice diagram.

 

[Stephen Benskin]

That is a really nice diagram.

Thanks. I post it all the time. I guess that will continue as long as people find it useful or totally ignore it because then it will always seem new.