Further study of R-D1 metering area

[jim_buchanan]

This discussion should have its own thread.

I read one post that described the metering as bottom weighted and did not understand until I found this:

http://www.cameraquest.com/voigtbmeter.htm

The R-D1 is exactly like the Bessa, good or bad.

My thought is how I liked the Leica way of the round white circle painted on the shutter curtain, and gave a partial spot/center-weighted reading. I wonder how the shape and size of the R-D1 white meter area would effect metering? Would a 10mm white circle in the middle of the shutter curtain give a correct reading of that 10mm image height on sensor?

I find the whole assembly very odd. The horizontal shape of the meter area. The aiming of the meter cell down and to the left.

In fact, I have just mapped the meter sensitivity area within the 28mm frame lines. My light source is a desk lamp from 8 feet away. The brightest reading was set at 1/1000 sec. The frame was divided into 1/3s for reference.

Then, I wondered what would the meter cell sensitivity area look like without any white or grey areas. To do this I covered the shutter with black flocking paper as used in telescopes. The shutter speeds are illustrated in the chart.

Note the "hot spot" on the black chart that corresponds to the meter's position and angle. Remember that the image is reversed and upside down at the focal plane. The black chart is similiar to the R-D1 grey shutter blades chart in that the meter sensitivity is greatest at about 7 o'clock.

If this all is not very logical and a surprise as how the R-D1 meter works, think about this.

The grey and white shutter curtain is the size of a regular 35mm camera covering an image circle of about 43mm. The R-D1 sensor has an image circle of 28mm. ALL M lenses for this camera will deliver light to the entire area of the R-D1 shutter curtain, but the sensor records only 1/1.53 of that area. So, given the hot spot at 7 o'clock, the R-D1 meter will record an image brightness that would not even be in the file, and an underexposed image is highly likely.

 

[pfogle]

Here's my effort, which gives roughly the same response.

the source was a small, very bright LED torch, the lens was the Rokkor 40mm at f5.6, and the camera was scanned over the 28mm frame lines. The positions of the crosses is accurate - though the tripod wasn't very level!

The numbers are the reciprocals of the shutter speed - ie 1/30 sec -> 30 on the picture.

 

[jlw]

Thanks for pinning that down.

The asymmetrical sensitivity pattern is odd -- but I recall some months ago reading an explanation (seems as if it was in Pop Photo) that C-V had designed it that way on purpose, to provide a center-bottom-weighted pattern whether the camera was held vertically or horizontally (vertically assuming the shutter release end up.)

 

[Bob Parsons]

Thanks for pinning that down.

The asymmetrical sensitivity pattern is odd -- but I recall some months ago reading an explanation (seems as if it was in Pop Photo) that C-V had designed it that way on purpose, to provide a center-bottom-weighted pattern whether the camera was held vertically or horizontally (vertically assuming the shutter release end up.)

Thanks for that observation That could explain why I sometimes have to compensate the exposure more than usual when taking vertical shots with sky. I hold the camera with the shutter button down to prevent having to bend my right wrist round the top of the camera. It also helps to keep your elbows tucked in.

Bob.

 

[Erl]

I think the logic expressed above is very good. It does, however, suggest that Epson have designed the camera to be used with the "dumb lever" activated! ie. providing for photographers who don't want to be selective about the area they are metering beyond just P&S technique.

It also may explain some of my exposure mistakes as I was assuming (fool) that the metering pattern was more centre specific.

Cheers,
Erl

 

[jim_buchanan]

In reading the Cameraquest page about the Bessa R and L meter, the point was made that an M lens took more room inside the R camera and positioning the photo cell had as much to do with space concerns as anything. After reading about the L with the photo cell in the bottom of the box at 6 o'clock reading light up into the center of the focal plane, I'd rather have that arrangement on the R-D1.

 

[iml]

Really interesting stuff, and explains a lot of my experience. I'll be keeping this in mind next time I'm shooting, I was beginning to question my metering technique on occasion, assuming I was using something centre-weighted, now I have a much better idea what's going on.

Ian

 

[RichC]

I was beginning to question my metering technique on occasion

Quite - I think I'll start by holding my camera viewfinder-end down when taking portrait shots, though it feels awkward!

 

[jim_buchanan]

Simulation of a circle meter pattern

Simulation of a circle meter pattern

I've taken the black mask I used for the original meter mappings and added a white 10mm diameter circle to the middle of the mask. This was mounted in front of the R-D1 shutter and the readings are indicated in the graph below.

The hot spot in the lower left of the graph corresponds to the photo cell mounting in the upper right of the shutter box (looking from behind the camera).

The white circle does give a separation of meter readings compared to the original mapping. However, the highest reading, 125, appears to be below center frame.

The EV range from left to right of the original white horizontal band is 3 stops. The white circle mask produces 1 stop.

If the hot spot, closest to the photo cell, illustrated in the lower left of the graph, could be isolated or blocked, the white circle concept could provide a more effective metering pattern. For example, reposition of the photo cell, or a micro hood to block the lower left sensitivity.