Is ISO 1067 close enough to 1000 that it doesn't matter?

I'm using Kodak TMZ and want to rate it at 1000. There are two settings between 800 and 1600 on the back of an M6. Divide that difference by 3 and you get about 267, so we have 800, 1067, 1337 and 1601.

I'm guessing that 1067 is close enough to 1000 that it doesn't matter but thought I'd check with folks here. Thoughts?

Regards,
Fergus

-- Fergus Hammond (fhammond@adobe.com), July 17, 2001

Answers

The ISO's between 800 and 1600 are 1000 and 1200 (even if they aren't spaced arithmetically) All ISO's are multiples, so that 800/1000/1200/ 1600 are two stops faster than 200/250/320/400 - in effect it's the '1200' that should be 1280, not the 1000 that should be '1067'

Manufacturers take all this into account - if a developer (film, not Adobe) lists times to process a film at 1000 or 1200, they assume you use the meter marks spaced between 800 and 1600, even if the numbers aren't exactly even.

In the same way, ISO 125 is twice as fast as ISO 64, even though it really should be ISO 128 (or ISO 62.5) to come out mathematically correct. If Kodak made memory chips, you'd be able to get a 250 Mb card for your PC/Mac, 'sted 256, 'cause they round off the ISO's.

-- Andy Piper (apidens@denver.infi.net), July 17, 2001.

A bit of pointless nitpicking: the "standard" is actually 1250 rather than 1200.

ISO ratings are a logarithmic progression rather than a linear progression, so dividing the difference between 1600 and 800 by 3 isn't really an appropriate methodology for determining intermediate speeds.

-- Mike Dixon (mike@mikedixonphotography.com), July 17, 2001.

To add still more to the excess info that Fergus probably didn't want:

1.) On a pocket calculator, or use the calculator function on your PC: Take the cube root of 2. Answer: 1.2599 (approximately). Save this answer in memory.

2.) Multiply the saved ratio by 800. Answer: 1007.9 (approximately 1000).

3.) multiply 1007.9 by the saved ratio of 1.2599. Answer: 1,2699 (equivalent to the standard speed figure of 1,250).

4.) Now try this starting with a different ISO speed, like 100. Result: 100, 125.99, 158.7, 199.99. In other words, we get the standard ISO values of 100, 125, 160, and 200, after rounding off.

So the conclusion is that by taking the cube root of 2, we get a ratio which spaces the ISO speeds into even one-third stop increments. The spacing between the successive speeds is thus seen to be multiplicative, by a constant ratio.

Now if I could just take pictures like Rob Appleby . . .

-- Bob Fleischman (RFXMAIL@prodigy.net), July 17, 2001.

To be even more nit-picky, I think it is more correct to say film speeds and shuter speeds follow a linear progression when compared to themselves and each other, but a geometric progression when compared to f-stops; and f-stops are related to each other in an inverse square realationship. Hence, ISO 1600 film is in fact 8 times faster than ISO 200 film, and a shutter speed of 1/1000 is 8 times faster than 1/125, both of which are direct linear relationships. But these represent a difference of only 3 f-stops which is a (inverse) geometric relationship. So the linear difference turns out to follow the formula 1/(2^n), where n is the actual difference in number of f- stops. I think :-)))

-- Jack Flesher (jbflesher@msn.com), July 17, 2001.

Sorry, I was writing my above post as Bob was putting up his post. By taking the cube root of 2 as he suggests, we are effectively finding the appropriate film speeds for the 1/3 f-stop differences. ie; 2^ (1/3) = the cube root of two, where 1/3 replaces n, the number of stops, in my earlier formula.

-- Jack Flesher (jbflesher@msn.com), July 17, 2001.

Yahh...1250 is correct, not 1200...shows how long it's been since I shot Royal-X Pan in a Yashica twin-lens....

-- Andy Piper (apidens@denver.infi.net), July 18, 2001.