slide film iso determination

[weasel_bar]

Does anyone know how ISO is defined for slide film? For negative film, ISO is defined by the speed point, where density is 0.1 above baseline. According to Wikipedia, for slide film, ISO is "determined from the middle rather than the threshold of the curve." I tried looking up the definition on ISO website, but it costs 61 swiss francs to find out. Does anyone know what is the technical definition of ISO for reversal film?

 

[Dustin McAmera]

https://archive.org/details/Manual_of_Photography/page/n319/mode/2up?q=ISO

[Alan Marcus]

For reversal films, the net density (density over base plus fog) is specified as 0.20. The formula is 8/Hm where Hm produces the stated density. From page 304 The Focal Encyclopedia of Photography. 

[AlanKlein]

10 hours ago, weasel_bar said:

Does anyone know how ISO is defined for slide film? For negative film, ISO is defined by the speed point, where density is 0.1 above baseline. According to Wikipedia, for slide film, ISO is "determined from the middle rather than the threshold of the curve." I tried looking up the definition on ISO website, but it costs 61 swiss francs to find out. Does anyone know what is the technical definition of ISO for reversal film?

 

What does the differing calculations mean to the average photographer?  For example, box speed ISO 100 on reversal vs positive color film means what?

[Dustin McAmera]

I remember finding with dismay how much it would cost to get the standard (ISO 732) for the design of 120 and 127 film (I hoped I could either quote it or link to it in Camera-wiki), though I knew I'd been able to download older versions free. Then I remembered that was because I worked at a university before, and we had a subscription. I know the ISO must cost a lot to run, and a lot of their customers are commercial and should pay for the information: but it feels unjust as a little citizen, that standards cost so much: 'This is the law, and it'll cost you sixty bucks to know what it is'.

Edited April 22, 2023 by Dustin McAmera

[weasel_bar]

11 hours ago, Dustin McAmera said:

https://archive.org/details/Manual_of_Photography/page/n319/mode/2up?q=ISO

This is extremely helpful. Thank you! 

[AlanKlein]

21 hours ago, Dustin McAmera said:

I remember finding with dismay how much it would cost to get the standard (ISO 732) for the design of 120 and 127 film (I hoped I could either quote it or link to it in Camera-wiki), though I knew I'd been able to download older versions free. Then I remembered that was because I worked at a university before, and we had a subscription. I know the ISO must cost a lot to run, and a lot of their customers are commercial and should pay for the information: but it feels unjust as a little citizen, that standards cost so much: 'This is the law, and it'll cost you sixty bucks to know what it is'.

Most standards are created by private organizations not the government, hence the cost.  The laws usually refers to standards by name and number and don't necessarily describe all the details which can be very complicated.  This way, they don't have to change the law every time the standard is upgraded.  If the government creates the standard, then it will be described in the law at no cost to the citizens. 

[glen_h]

For black and white negative film, it is, more or less, independent of the developer.

Development time is selected for an appropriate contrast.

For color film, the processing, and development times, are predefined.

As you note, for negative film the point is defined above base plus fog.

Negative films have plenty of exposure latitude, and the point is chosen to use the more sensitive part.

 

Reversal films have minimal exposure latitude. Often less on the overexposure side than underexposure.

Setting the point in the middle seems good to me.

 

That leaves out the possibility of push processing. 

As noted, for black and white films, the developing time is optimal for each film.

For color film, developing time is fixed, and not necessarily optimal.

It is normal for higher speed films to have longer development times, but not for color.

So, higher speed color films can be pushed. Some are designed to be pushed.

[weasel_bar]

On 4/22/2023 at 10:32 AM, AlanKlein said:

What does the differing calculations mean to the average photographer?  For example, box speed ISO 100 on reversal vs positive color film means what?

It may not make any practical difference. However, it provides insight. After reviewing the excellent resource provided by @Dustin McAmera I surmise the following.

For color slide film, the reference point for calculating ISO is the minimum density, which determines WHITE point. Thus, if the characteristic curve is reasonably linear in the middle, I think the light meter will hopefully place mid gray at 18% transmittance of the minimum density. My guess is that this means the midtones will look right to the eye compared to the highlights. How much blacker the image will go in the shadows will vary by film stock, since the Dmax is different. It also illustrates explicitly how, for the same box speed, contrast and tolerance to exposure variation are influenced. 

For example, I ballpark the following from characteristic curves

Velvia 100: Dmin=0.2. D=0.4-->logH= -0.6. D=2.2-->logH= -1.5. LogHm= -1.05-->iso ~100

Provia 100: Dmin=0.2. D=0.4-->logH = -0.4. D=2.2-->logH= -1.7. LogHm=--1.05-->Iso ~100

For the same box iso, velvia goes to a density of 2.2 (1/150th transmittance) at almost a full stop before provia does. Even though velvia has a higher Dmax than provia, at a density of 2.2, your eyes will probably not discern much detail, and you're going to perceive black. So velvia goes "darkish" quickly. This might explain why people tend to err on the side of overexposure with velvia, compared to other slide films.

For color negative film, the reference point for iso is also the min density. But here, the min density determines where the SHADOWS lie. Also, we use a single point rather than a midpoint. This may be helpful, since the response stays very linear for many stops of overexposure. Different color neg films have different densities for "mid-gray" exposure, depending on the slope of the curve after Hm.

[glen_h]

On 4/22/2023 at 8:32 AM, AlanKlein said:

What does the differing calculations mean to the average photographer?  For example, box speed ISO 100 on reversal vs positive color film means what?

 

For the average, or not so average, photographer, negative films have a large enough dynamic range, or as it is more often described, exposure latitude.

C41 films, even more than usual black and white films.  You normally expose for the bottom, more sensitive part, of the curve.

 

Reversal films have much smaller exposure latitude, so you want to be right on. Some say within a 1/4 stop.

The three LED light meter on the Nikon FM tells you when you are within 1/5 of a stop.

 

ASA was originally done when we had mostly averaging light meters, that average over the whole scene.

It turns out that most scenes have about average reflectance, so that works pretty well.

 

 

[glen_h]

The data sheet for current Ektachrome is here:

https://imaging.kodakalaris.com/sites/default/files/files/products/e4000_ektachrome_100.pdf

 

As usual, no part of the characteristic curve is completely straight, but the main part is close enough.

It covers about 2 log10 units on the x-axis (exposure axis), so a factor of 100, or about 6.5 stops.

For the same part of the curve, it covers about -3 log10 units on the y-axis (density), so a gamma of about -1.5.

(The slope is negative for reversal films, positive for negative films.)

[glen_h]

For comparison, the data sheet for Portra 400 is here:

https://imaging.kodakalaris.com/sites/default/files/files/resources/e4050_portra_400.pdf

 

The main part of the curve, well all three of them, is amazingly straight!

It covers about 3 log10 units on the exposure axis, and has not started to curve over at the top.

It might have another log10 unit left before it curves too much.

The 3 log10 units, gives about 10 stops, but maybe 12 or 14 if they kept the curve going.

Match up to the density units on the left, and the slope is about 0.5, usual for C41 films.

(That is also usual for C41 black and white films like XP2.)

For a scene that fits in the 6.5 stops of Ektachrome, the additional maybe 6 means that

you can overexpose 6 stops, and still be within the straight part of the curve!

 

The low 0.5 for the slope means that you have to be pretty close when printing.

Color negatives are always printed with an enlarging light meter. 

 

The slope of about 0.7 for more usual black and white films, means you can usually

get away with using test strips, and get close enough.  And maybe only 3 or 4 stops

of overexposure.

[AlanKlein]

1 hour ago, glen_h said:

For comparison, the data sheet for Portra 400 is here:

https://imaging.kodakalaris.com/sites/default/files/files/resources/e4050_portra_400.pdf

 

The main part of the curve, well all three of them, is amazingly straight!

It covers about 3 log10 units on the exposure axis, and has not started to curve over at the top.

It might have another log10 unit left before it curves too much.

The 3 log10 units, gives about 10 stops, but maybe 12 or 14 if they kept the curve going.

Match up to the density units on the left, and the slope is about 0.5, usual for C41 films.

(That is also usual for C41 black and white films like XP2.)

For a scene that fits in the 6.5 stops of Ektachrome, the additional maybe 6 means that

you can overexpose 6 stops, and still be within the straight part of the curve!

 

The low 0.5 for the slope means that you have to be pretty close when printing.

Color negatives are always printed with an enlarging light meter. 

 

The slope of about 0.7 for more usual black and white films, means you can usually

get away with using test strips, and get close enough.  And maybe only 3 or 4 stops

of overexposure.

I found when I bracketed +1 and -1 stops, there is a color shift in the greens especially, even though all three shots can be adjusted for proper luminance exposure after scanning.  

Edited May 2, 2023 by AlanKlein