Potentially moronic question about metering
Hey folks, I'm new here, but I've always had a question that I could not get an answer to.
Let's say we are in spot metering mode, so the meter is working off a single part of the frame. When your camera (or a meter) meters a subject, it is trying to match it to a neutral 18% grey. Using aperture and shutter priority changes those two elements together, but always with the mind of matching that certain 18% grey value. f 5.6 at 1/125 sec is the same amount of light at f 4 at 1/250 sec. So, the meter in your camera is measuring for the light on the spot, then bringing it to some constant level, x, for when it hits the CCD (the CCD's receptivity does change based in ISO, however, so the meter is looking for another x, in that case). x must be some precise, measurable amount of light that engineers must have built into the meter.
First, is that analysis correct, or have I missed something?
Second, Does anyone know the value of x, and what units it would be in? I'm thinking moles of photons, or candlepower.
Finally, If i were to take a picture of something that had the exact same amount of light as x, meaning that the camera did not have to change the brightness of the subject in any way before it hit the CCD to get a proper exposure, what would be my ISO, shutter speed, and f stop values?
Sorry if that sounded like a math problem. I would be greatly indebted to anyone who knows or can lead me to an answer -- or at least tell me why it's a dumb question.
Let's say we are in spot metering mode, so the meter is working off a single part of the frame. When your camera (or a meter) meters a subject, it is trying to match it to a neutral 18% grey. Using aperture and shutter priority changes those two elements together, but always with the mind of matching that certain 18% grey value. f 5.6 at 1/125 sec is the same amount of light at f 4 at 1/250 sec. So, the meter in your camera is measuring for the light on the spot, then bringing it to some constant level, x, for when it hits the CCD (the CCD's receptivity does change based in ISO, however, so the meter is looking for another x, in that case). x must be some precise, measurable amount of light that engineers must have built into the meter.
First, is that analysis correct, or have I missed something?
Second, Does anyone know the value of x, and what units it would be in? I'm thinking moles of photons, or candlepower.
Finally, If i were to take a picture of something that had the exact same amount of light as x, meaning that the camera did not have to change the brightness of the subject in any way before it hit the CCD to get a proper exposure, what would be my ISO, shutter speed, and f stop values?
Sorry if that sounded like a math problem. I would be greatly indebted to anyone who knows or can lead me to an answer -- or at least tell me why it's a dumb question.
0
Comments
As far as I know, and someone correct me if I'm wrong, the metering part of your camera just looks at the amount of light it receives and then says "this is pretty bright, I need a relatively high shutterspeed/small aperture" or something like that. No matching, changing ISO's etc. The metering part is calibrated at 18% grey. As long as it knows the amount of light in combination with a certain shutterspeed/aperture it can easily calculate the speeds/apertures for other amounts of light. It doesn't know anything about the actual color of the object your are measuring, so correction for white, black or any non-18% grey colors need to be done by you, manually. Is this what you said/meant?
The 'brightness' of an object can never be changed by your camera. The amount of light that hits the sensor can, by changing the aperture or shutterspeed. There is no 'default' or 'correct' setting, as in "for this amount of light, the shutterspeed will be X, the aperture will be Y and the ISO will be Z". Most dSLRs don't auto-change the ISO to begin with, and it depends on how you have set up the camera as well (aperture priority, shutter priority, etc.)
Does this help? at all? We may need someone who is more knowledgeable here...
Btw, you play the cello by any chance?
www.ivarborst.nl & smugmug
Check this out. http://www.kenrockwell.com/tech/ev.htm
Natural selection is responsible for every living thing that exists.
D3s, D500, D5300, and way more glass than the wife knows about.
Setting the ISO declares that X is the amount of light energy needed to record as middle gray.
The 'exposure meter' reads the portion of the frame chosen by the operator and measures the amount of light falling on that portion and chooses a shutter speed/aperture combination that will provide that amount of light to the sensor.
Lew
On this part: I have no clue to the answer as to the units. Ziggy may. Hopefully he jumps into this forum occasionally.
On this:
I'm betting that those algorithms are carefully guarded secrets by the various manufacturers. It seems it would be a function of the actual sensor, the camera's processor, and the metering system. DIGIC II vs DIGIC III, CCD in the 5D vs the XTi, metering in the 40D vs the 1DsMkII. How each of those components works in tandem gives you the answer. So I'm not sure you could (easily) replicate that. You could find out the chemical makeup of Kodachrome or Velvia, but to reproduce the look they give you without using them? Nah.
BTW - I loved the title of your thread!
-Fleetwood Mac
That is why some camera/meter combinations under-expose slightly with some lens combinations.
I used to use a Weston Master V meter that used an 'absolute' measure in lumens of light falling on a sensor and I used that to give me a exposure value. Not only did I have to adjust from that value for my specific camera/film combination, I adjusted from the value to correct for my own film/developer. Additionally, I would have to recalibrate my meter every couple of years.
So, IMO, there is no absolute X, there are only relative Xs for each camera/lens combination and within that, specific Xs for each specific camera/lens. (and that is why there are exposure compensation dials on most SLR cameras)
When taking a picture of a properly metered 18% grey card, what is the amount of light that the camera allows to hit the CCD? Isn't this amount of light always the same regardless of the light conditions around the grey card because the Lens and shutter adjust for it? What units is this amount of light hitting the CCD measured in?
JdRyan: I had never thought of that before, you might be right that it is a closely guarded secret. But, Isn't the ISO rating on a CCD a universal measurement? 100 ISO is a measure of the sensitivity to light that is the same across all CCDs and even film, right? Then, this information should be out there.
Icebear: Very interesting, it told me at least that it's probably a matter of lumens.
LittleLew: Isn't that the point of the f stop/shutter speed/iso rating? to provide an equal measure of light across all lenses, cameras? Maybe I'm totally wrong about this, but a f 5.6 1/250 sec 100 iso shot is the same across all lenses and cameras. That's why you can use a traditional meter that is not even connected to the camera. Perhaps everything isn't always exactly calibrated, but there is still some ideal X that is attempted. The exposure compensation dail's job is essentially to change the amount of lumens hitting the CCD off of the magic X based on setting by the user.
Ivar: I don't play, but I am a fan.
Right.
But the finite adjustments you make in process adjust your combination of lenses to the ideal.
Fair enough -- I agree that knowing the value of x isn't important to know in order to get an ideal picture, and in fact is often overidden by the photographer, I still want to know as a matter of theoretical curiosity. The only way i think it might be important practically is if you were designing a camera, which I am most certainly not doing.