What is the difference in the f-stop measurement and area of measurement and method for the Leica integral/integrative metering system and the traditional Nikon 60/40 centerweighed metering and Cosina/V'lander centerweighed metering systems? I know that Nikon measures 60% in the spot and 40% in the outside part of the spot for the whole viewfinder. The integral/integrative metering in Leica I presume is different? 80/20? Thanks for your help.

-- Alfie Wang (leica_phile@hotmail.com), February 17, 2002

Answers

For what Leica camera?

-- John Collier (jbcollier@powersurfr.com), February 17, 2002.

I meant the Leica R4 and the Leica R8 separately...

-- Alfie Wang (leica_phile@hotmail.com), February 17, 2002.

I assume you are referring to the R4-R7. Those are based on the classic Minolta pattern which is center and bottom-weighted (the theory was to bias for foreground and exclude sky--the fallacy of course is when you turn the camera for a vertical shot it's weighted to one side or the other). The meter "weighting" tapers off from the weighted portion in about a 3:1 ratio from very center to outer fringes, which would make it about 75/25 if my math is correct. However since the weighted section is elliptical rather than circular, and "squashed" down toward the bottom of the frame, it's hard to make a direct comparison to the classic Nikon 60/40. If you really want to understand the integral metering, shoot test shots on slide film of black and white teddy-bears against dark and light backgrounds, with the bears occupying different proportions of the frame, using the meter's suggested exposure. You will then see how the meter sees. Or, you can do what I do: use the selective metering. I haven't used the averaging metering on my R cameras ever.

-- Jay (infinitydt@aol.com), February 17, 2002.

In re the R8: the integral metering is similar to the R7 in that it's somewhat elongated but it doesn't have the bottom-bias. The evaluative metering is a 6-segment very near to the Nikon F4/8008/6006 of the late 1980's. Something like pixels in a digital camera, the more segments in an evaluative meter the more complex the algorithms can be programmed. To the R8's six segments, the Nikon F100 has 10, the EOS 1n 16, the 1V 21, the Elan 7 35 and the Nikon F5 has 1005! All the foregoing also have much smaller-area spotmeters. However the R8's selective metering is much more sensitive to low light levels than any camera I currently know of, and using the AE lock (if you are careful not to trip the shutter--most other well- designed cameras have a separate button for AE lock)in A mode you can get excellent exposures even with chrome. In manual-mode the R8 is once again hindered by its 1/2-stop shutter and aperture increments vs the 1/3-stops available on most if not all AF SLR's today.

-- Jay (infinitydt@aol.com), February 17, 2002.

Alfie:

I suppose you are matching a couple of cameras there.

All right, Nikon 60/40 is a True/False type of measurement

The Leica uses a linear integrated measure with coefficients from middle to extremity. It is far more subtle. In spot it's 80/20 but for integral it's 70/50/30 or a combination close to that. Read the R7 book by Arthur Landt.

In Practice, both works.

I used to have a Nikon 35TI, wonderful system with fuzzy logic. But I decided to be fuzzy myself, bought the Leica and tried.

Both approach work quite nice and there is not so much difference when you know how to use it. Keep in mind that with a SLR, it depends on the lens: a 24mm integrates more light than a 90mm since 60% of the scene is larger with a 24mm than with a 90mm.

Sorry to say but most R4 till R7 have been developped with Minolta, a great company, renonwed for their hand held light measurements (along with sekonic). So be assured that the best compromise has been taken. It's far from perfect. Just learn how to use the tool.

Nikon approach is quite good to be honest. I think it's works in 92% of the cases. Leica is 96%. The photographer using a well known film and a well known lens is 99%.

Sorry to upset.... Xavier Xavier

-- Xavier d'Alfort (hot_billexf@hotmail.com), February 19, 2002.