Do You Center The Meter?

[table1349]

Centering the meter is not always the exposure you want. I know how close to 1/3 stop each of the spots on my meter are and can adjust for the scene as necessary from the meter as I want it. There are times to over expose a bit and there are times to under expose a bit. Perhaps this is just a hold over habit from my film days.

 

[Gary A.]

Meters, all meters will calibrate to a medium gray, a gray card, (18%-22% gray). (Internal software adjustments notwithstanding.)

You'd think but I'm beginning to wonder. Centering the meter gives a useable exposure for both JPEG and Raw, it just seems like it consistently stops short of a full sensor exposure.

My ignorance, but what do you mean by "... a full sensor exposure"? The Spot only samples the 'spot', (usually somewhere between 1 to 10 degree spot), and provides the exposure setting based upon medium gray for that individual spot.

 

[smoke665]

a full sensor exposure"

As much data as possible without blowing the highlights. I may not have stated it right in the OP, I know how to get a proper exposure, how to use the in camera meter and my hand held. I routinely use a combination of judgement, average of spot readings, and watching the histogram, to determine the "correct exposure ". However, it's a known fact that different manufacturers apply certain "proprietary adjustments" to both the Raw file and a processed JPEG, so is it possible that they also calibrate the meter to give a proper exposure per "their standards" for a JPEG processed by their internal software, regardless of if you saving Raw or JPEG?

Here's an example of something I was playing with the other day. I had the camera set to Raw+ (save both Raw and JPEG to card). I maximized the exposure per the histogram (which by the meter was not quite a full stop right of center). The first image is the histogram from the Raw File.

The second image below is the histogram from the JPEG created from the same exposure at the same time

The Raw file had maximum data and showed no sign of blown highlights when opened post but the JPEG looked over exposed. Had I centered the meter, the JPEG would have been properly exposed, and the Raw would have needed a bump in post.

Am I missing something?

 

[Overread]

Overexposure is when the line hits the far right hand side rather than when it peeks through the top of the graph; just as full underexposure is when it hits the far left hand side rather than doesn't appear at all.

JPEGs always have some editing applied because they are essentially adjusted RAWs - even on faithful/neutral settings some changes have to be made. It's also important to realise that in the camera the RAW should have the same data because the RAW uses an embedded JPEG edited to the same standards as the settings you've got set for JPEG editing.

When you take the photo out of the camera and use software for the RAW the software has to pick a starting point for all its settings to display the RAW as an image. As a result different software options will display the same RAW slightly differently. It's still the same data underneath; just different default starting points.

 

[smoke665]

Overexposure is when the line hits the far right hand side rather than when it peeks through the top of the graph;

Agreed, this represents the point where I had pulled back as more exposure sent it up the right side, but again the question I'm asking.

"Is it possible that camera manufacturers calibrate the in camera meter to their proprietary standards for a JPEG, or is it a true reading based on gray scale?"

 

[fmw]

a 15% gray subject at proper exposure

True, but is that a proper exposure for a JPEG per the camera manufacturer internal software - meaning it's hedging on a full ETTR?

Easy enough to find out.

 

[Ysarex]

a full sensor exposure"

As much data as possible without blowing the highlights. I may not have stated it right in the OP, I know how to get a proper exposure, how to use the in camera meter and my hand held. I routinely use a combination of judgement, average of spot readings, and watching the histogram, to determine the "correct exposure ". However, it's a known fact that different manufacturers apply certain "proprietary adjustments" to both the Raw file and a processed JPEG, so is it possible that they also calibrate the meter to give a proper exposure per "their standards" for a JPEG processed by their internal software, regardless of if you saving Raw or JPEG?

Here's an example of something I was playing with the other day. I had the camera set to Raw+ (save both Raw and JPEG to card). I maximized the exposure per the histogram (which by the meter was not quite a full stop right of center). The first image is the histogram from the Raw File.
View attachment 139657
The second image below is the histogram from the JPEG created from the same exposure at the same time
View attachment 139658

The Raw file had maximum data and showed no sign of blown highlights when opened post but the JPEG looked over exposed. Had I centered the meter, the JPEG would have been properly exposed, and the Raw would have needed a bump in post.

Am I missing something?

A little fly in the ointment -- that's not a raw file histogram you're looking at there. That's the histogram of an RGB file processed from the raw file and already substantially manipulated by the processing software. If you exam the actual raw histogram I suspect you'll find the discrepancy you've noted is in fact greater than this would indicate.

Joe

 

[Ysarex]

Overexposure is when the line hits the far right hand side rather than when it peeks through the top of the graph;

Agreed, this represents the point where I had pulled back as more exposure sent it up the right side, but again the question I'm asking.

"Is it possible that camera manufacturers calibrate the in camera meter to their proprietary standards for a JPEG, or is it a true reading based on gray scale?"

The camera manufacturers calibrate their camera meters to a standard brightness (gray scale if you like) for the ISO value they've assigned to the camera. Now think of that and your original question in light of this simple fact. The sensors in your and all of our cameras do not have rated or assigned ISO values.

Joe

 

[smoke665]

That's the histogram of an RGB file processed from the raw file and already substantially manipulated by the processing software

I forget a lot of things, but most of what you've told me I remember, this is one of those things I remember.

Now think of that and your original question in light of this simple fact. The sensors in your and all of our cameras do not have rated or assigned ISO values.

So if I extrapolate this statement, combined with other information I know then the answer to my question is yes? Camera manufacturers can/do adjust the in camera meter to a proprietary setting, other than adhering to a standardized gray scale reading.

 

[Gary A.]

Yes. Sorta like film. Different color films have different characteristics even though they may share similar ASA. Difference cameras will put an individual manufacturer's fingerprint on your capture.

 

[Ysarex]

That's the histogram of an RGB file processed from the raw file and already substantially manipulated by the processing software

I forget a lot of things, but most of what you've told me I remember, this is one of those things I remember.

Now think of that and your original question in light of this simple fact. The sensors in your and all of our cameras do not have rated or assigned ISO values.

So if I extrapolate this statement, combined with other information I know then the answer to my question is yes? Camera manufacturers can/do adjust the in camera meter to a proprietary setting, other than adhering to a standardized gray scale reading.

Not exactly. They calibrate the meter to achieve a standard brightness level in the output JPEG. If you look at the ISO standard for a digital camera, the ISO value is set based on the output of the camera's JPEG processing software. That's why I mentioned that the sensors don't in fact have assigned ISO values.

So the meter is well calibrated to do its job and the brightness level standard (good exposure) is shared between the different manufacturers. The point where they put a spin on it all is in the processing software that creates the JPEG. Here they make a conservative choice between JPEG exposure versus raw file exposure and that's the discrepancy you're seeing. They design the JPEG processing software to render a standard brightness JPEG from a conservatively (under) exposed raw file and the meter is adjusted to that software's JPEG output. So the calibration for the meter isn't what's proprietary here so much as the processing software that creates the JPEG.

Joe

 

[smoke665]

The point where they put a spin on it all is in the processing software that creates the JPEG

True to a point, though we've had the conversation before as to what they actually apply to a Raw file. IE pretty much accepted that all manufacturers apply some noise reduction regardless of the menu settings.

 

[Ysarex]

The point where they put a spin on it all is in the processing software that creates the JPEG

True to a point, though we've had the conversation before as to what they actually apply to a Raw file. IE pretty much accepted that all manufacturers apply some noise reduction regardless of the menu settings.

Oh sure -- they do a whole basket full of stuff as they convert and process the raw file into a JPEG, including stuff they keep locked in a safe. The raw file is demosaiced, white balanced, tone curve applied with any user supplied additions like highlight/shadow settings, distortion adjusted, CA removed, noise filtered, and at least sharpened. All of that is spin and all of that is the end product they're selling. They sell a camera that creates a finished photo when you click the shutter.

If you save a raw file as well you'll find that they designed all of the above to make a good JPEG from a raw file that's conservatively exposed.

Joe

P.S. The free software RawTherapee includes an option to display a raw file histogram.

 

[DanOstergren]

I use it as a starting guide, then adjust the exposure according to how I want it the look.

 

[smoke665]

I use it as a starting guide, then adjust the exposure according to how I want it the look.

Agree, pretty much SOP for most. My post was more to the question of understanding how camera manufacturer's might "alter" the meter reading. To find a common starting point to adjust from.