10D ISO ratings a lie?

Has anyone tried using "Sunny f16" in manual mode with a 10D on a sunny day?

When I do, the pictures are under-exposed by over a stop. When I check exposure of a flat, grey surface or blue sky in the camera, and compare it to my Sekonic meter, there is about 1.3 stops difference.

This casts a shadow on the integrity of the camera's "low noise at high ISO", when ISO 800 is really ISO 320, no?

--
John

Bookmark (nt)

John Sheehy said:

Has anyone tried using "Sunny f16" in manual mode with a 10D on a
sunny day?

When I do, the pictures are under-exposed by over a stop. When I
check exposure of a flat, grey surface or blue sky in the camera,
and compare it to my Sekonic meter, there is about 1.3 stops
difference.

This casts a shadow on the integrity of the camera's "low noise at
high ISO", when ISO 800 is really ISO 320, no?

--
John

Click to expand...

John,

First off, I am not anywhere remotely close to being an expert or even familar with the specifics of photography. I do remember an earlier thread having something to do with the fact that the 10d uses 12% gray and not 18% gray metering. I am not sure how, if at all, that figures into explaining the difference but in a simplistic way it could explain at least part of the difference.

FWIW

Ernie

John Sheehy said:

Has anyone tried using "Sunny f16" in manual mode with a 10D on a
sunny day?

When I do, the pictures are under-exposed by over a stop. When I
check exposure of a flat, grey surface or blue sky in the camera,
and compare it to my Sekonic meter, there is about 1.3 stops
difference.

This casts a shadow on the integrity of the camera's "low noise at
high ISO", when ISO 800 is really ISO 320, no?

--
John

Click to expand...

ErnieW said:

John,

Click to expand...

ErnieW said:

First off, I am not anywhere remotely close to being an expert or
even familar with the specifics of photography. I do remember an
earlier thread having something to do with the fact that the 10d
uses 12% gray and not 18% gray metering. I am not sure how, if at
all, that figures into explaining the difference but in a
simplistic way it could explain at least part of the difference.

Click to expand...

The issue has nothing directly to do with metering. No metering is involved in taking a "sunny f16" shot in manual mode. If the camera says that it is ISO 100, and you set the aperture to f16, and the shutter speed to 1/100, a neutral object should be somewhere in the middle of the range, like 128 out of 255. Instead, what I get is more like 60 out of 255, and even a white car in direct sunlight is only about 190 to 200.

Also, 12% grey as 128 is actually a higher exposure than 18% grey. The Sunny f16 problem is an underexposure.
--
John

John Sheehy said:

Has anyone tried using "Sunny f16" in manual mode with a 10D on a
sunny day?

When I do, the pictures are under-exposed by over a stop. When I
check exposure of a flat, grey surface or blue sky in the camera,
and compare it to my Sekonic meter, there is about 1.3 stops
difference.

This casts a shadow on the integrity of the camera's "low noise at
high ISO", when ISO 800 is really ISO 320, no?

--
John

Click to expand...

Interesting,

Tried this and got a slight overexposure. This measured from light cement and a light tan siding on a house.

My Luna-Pro gives me an almost exact f16 @ 1/100 @100 ASA in incident mode and in reflected gives me about f16 @ 1/125 at 100 ASA from both sidewalk and siding.

This perfectly supports the slight overexposure (about 1/3-1/2 stop) I got with the 10D just clicking away.... at f16, 1/100. and 100ASA.

I would calibrate either your Sekonic meter or have the meter on the 10D checked. They should be able to do this while you wait at a good camera store...

G.

Holden said:

John Sheehy wrote:
Interesting,

Click to expand...

Holden said:

Tried this and got a slight overexposure. This measured from light
cement and a light tan siding on a house.

Click to expand...

Holden said:

My Luna-Pro gives me an almost exact f16 @ 1/100 @100 ASA in
incident mode and in reflected gives me about f16 @ 1/125 at 100
ASA from both sidewalk and siding.

This perfectly supports the slight overexposure (about 1/3-1/2
stop) I got with the 10D just clicking away.... at f16, 1/100. and
100ASA.

I would calibrate either your Sekonic meter or have the meter on
the 10D checked. They should be able to do this while you wait at a
good camera store...

Click to expand...

My Sekonic meter is properly calibrated. It reports exactly as it should, and agrees with the sunny f16 rule.

Taking a sunny f16 picture in manual mode is not affected by the camera's meter! I'm not sure we're communicating here. Did you actually set the camera to ISO 100, manual mode, 1/100, and f16, and take a picture of

something in direct sunlight? And the white objects were near 255 or clipped? Medium grey came out about 128, in the middle of the histogram?
--
John

I just tried it on my 10D, and the brightest thing in the image (other than specular highlight reflections off cars) came out at 99%, according to Photoshop's eye-dropper. Note that the bright item in this case was white-painted wood trim on a house. (For reference, the RGB value was 252, 248, 247.) Oh, and a gray metal sign-post came out as 54%, and the blue sky came out at 80%. Not exactly scientific, but I don't have any gray cards.

--Travis Cobbs

John Sheehy said:

Has anyone tried using "Sunny f16" in manual mode with a 10D on a
sunny day?

When I do, the pictures are under-exposed by over a stop. When I
check exposure of a flat, grey surface or blue sky in the camera,
and compare it to my Sekonic meter, there is about 1.3 stops
difference.

This casts a shadow on the integrity of the camera's "low noise at
high ISO", when ISO 800 is really ISO 320, no?

--
John

Click to expand...

Travis Cobbs said:

I just tried it on my 10D, and the brightest thing in the image
(other than specular highlight reflections off cars) came out at
99%, according to Photoshop's eye-dropper. Note that the bright
item in this case was white-painted wood trim on a house. (For
reference, the RGB value was 252, 248, 247.) Oh, and a gray metal
sign-post came out as 54%, and the blue sky came out at 80%. Not
exactly scientific, but I don't have any gray cards.

Click to expand...

Sorry if this sounds like an "is the computer plugged in?" type of question, but I don't know you, so I don't know if you understood what I was talking about. What ISO, aperture, and shutter speed did you use? Any non-default contrast/brightness settings in camera (if JPEG)? What did the histogram on the back of the camera look like?

--
John

Fair enough. Shot taken in manual at ISO 100 F/16 1/100th second. Shot was taken as JPEG Large/Fine, but parameters set to Standard; WB set to sunlight; CF 1, 2, & 6 set to 1, CF 13 set to 2, all other CF set to 0. (Note that none of these CF settings will effect the result, other than that 6 must be set to 1 to get the 1/100sec exposure.)

For what it's worth the lens was a 70-200 F/4L (that's what was already on my camera). I'm not sure if the lens' reported F-stop can be inaccurate or not; if so, it would obviously effect the result.

Histogram looked great. Nothing visible at either edge, but very little of the image was actually white, so that's not too surprising. There was a small local maximum at about 92%-93% that dropped to nothing around 95%-96%. (Obviously this is guessing somewhat, but with the vertical separations every 20% the guesses should be reasonably accurate.)

John Sheehy said:

Sorry if this sounds like an "is the computer plugged in?" type of
question, but I don't know you, so I don't know if you understood
what I was talking about. What ISO, aperture, and shutter speed
did you use? Any non-default contrast/brightness settings in
camera (if JPEG)? What did the histogram on the back of the camera
look like?

Click to expand...

It's alpenglow time here but I tried it anyway ... 100 asa 1/100 f/16

Shooting toward the sun (which had set below the hill west of me)
the sky registered 70-80% on the histogram. Lens is Canon 75-100 IS.

--
Chuck Forsberg WA7KGX N2469R http://www.omen.com
Author of Industrial ZMODEM for Embedded Systems

That rule of thumb is mainly for negative film. Digital requires some adjustment to the rule.

John Sheehy said:

Has anyone tried using "Sunny f16" in manual mode with a 10D on a
sunny day?

When I do, the pictures are under-exposed by over a stop. When I
check exposure of a flat, grey surface or blue sky in the camera,
and compare it to my Sekonic meter, there is about 1.3 stops
difference.

This casts a shadow on the integrity of the camera's "low noise at
high ISO", when ISO 800 is really ISO 320, no?

--
John

Click to expand...

John Sheehy said:

My Sekonic meter is properly calibrated. It reports exactly as it
should, and agrees with the sunny f16 rule.

Taking a sunny f16 picture in manual mode is not affected by the
camera's meter! I'm not sure we're communicating here. Did you
actually set the camera to ISO 100, manual mode, 1/100, and f16,
and take a picture of
something in direct sunlight? And the white objects were near 255
or clipped? Medium grey came out about 128, in the middle of the
histogram?
--
John

Click to expand...

There is a possibility that your ISO sensitivity could be miscalibrated.. I didnt shoot for white I shot for medium gray as this is what sunny 16 was for after all... the slight differences I had from my Gossen meter would be due to the fact that the scene was mixed light/full sunlight and I had the chance to use both incident and reflected light.

If I had the time I would have set one of my gray cards down as well... BTW I had the WB set to auto as well... not that it makes a difference... but I was really quite amazed at how precise the 10D was...

G.

Al Pacheco said:

That rule of thumb is mainly for negative film. Digital requires
some adjustment to the rule.

Click to expand...

Sunny 16 is EV15... whether digital, analog, motion or still.

It is the value to which most test benches are calibrated. The sensor of a digital camera should get pretty close... at least to medium gray. Trying to hold all the dynamic range of such a bright scene may be beyond the specification of the given chip.. and yes would require some adjustment..

this is what Fred Mirandas DRI module is all about...

G.

These questions come up a lot among consumers of smaller sensor cameras, where there is even less consistency.

If you check you DSLR reviews, you'll find that most are pretty consistent with each othe (both within and across brands)r, which suggests that either:

• There's something strange with your gear
• All digital SLRs from all brands are off in the same way.

Now, it is true that film and electronic sensors have different methods for establishing sensitivity. Electronic sensors are supposed to be measured in accordance with ISO 12232:1998. This is standard is supposed to produce results close to film, but the measurement methods are fundamentally different.

You might find this interesting:

http://www.kinesisgear.com/opinion.html

You might try a search of dpreview for the text, "sekonic 10D" You'll find some interesting and inconsistent responses if you dig back.

--
Ron Parr
FAQ: http://www.cs.duke.edu/~parr/photography/faq.html
Gallery: http://www.pbase.com/parr/

Travis Cobbs said:

Fair enough. Shot taken in manual at ISO 100 F/16 1/100th second.
Shot was taken as JPEG Large/Fine, but parameters set to Standard;
WB set to sunlight; CF 1, 2, & 6 set to 1, CF 13 set to 2, all
other CF set to 0. (Note that none of these CF settings will
effect the result, other than that 6 must be set to 1 to get the
1/100sec exposure.)

Click to expand...

Since you didn't report it, and more so because I've made the mistake myself more times than I'd like to admit, you don't have any exposure compensation set, do you? The first time I ever checked my body against sunny f/16, I made that mistake and didn't figure it out for a while. Sigh...

And it does not apply as a hard and fast rule for digital. I never calibrated a bench but I make my living calibrating photographic equipment and digital is different enough to warrant a new rule of thumb.

Holden said:

Al Pacheco said:

That rule of thumb is mainly for negative film. Digital requires
some adjustment to the rule.

Click to expand...

Sunny 16 is EV15... whether digital, analog, motion or still.

It is the value to which most test benches are calibrated. The
sensor of a digital camera should get pretty close... at least to
medium gray. Trying to hold all the dynamic range of such a bright
scene may be beyond the specification of the given chip.. and yes
would require some adjustment..

this is what Fred Mirandas DRI module is all about...

G.

Click to expand...

MrFoopy said:

Since you didn't report it, and more so because I've made the
mistake myself more times than I'd like to admit, you don't have
any exposure compensation set, do you? The first time I ever
checked my body against sunny f/16, I made that mistake and didn't
figure it out for a while. Sigh...

Click to expand...

Well, no I didn't have it set, but I fail to see how it makes any difference. Exposure compensation isn't active in manual mode, so if it caused you problems, then you weren't in manual mode. As mentioned, I did the test in manual mode. As mentioned by John Sheehy (the original poster), the test requires you to put the camera in manual mode. That is, after all, the only way to control the F-stop, ISO, and exposure length all at once.

Its a base 2 logrithmic scale, 128 is 1ev from 255
8,16,32,64,128,255

middle of the histogram would be 32~64

Holden said:

There is a possibility that your ISO sensitivity could be
miscalibrated..

Click to expand...

Yes, that's my point, but I'm trying to see how universal it is. The way I figure, my "ISO 100" is actually ISO 40, so my camera has ISO 40, 80, 160, 320, 640, and 1280. I have posted a twin article in rec.photo.digital as well, and some people are reporting true ISO 100 sensitivity, and some are reporting something like 40, like me. The odd thing is that in auto mode, and with the metering in manual, the exposure histograms are what you would expect, so Canon has obviously tuned the uto-exposure of the camera to the true ISO sensitivity. This may explain why I get a lot less noise at "high ISOs" than some people do (using the camera's metering); they are really only 40% of what they say. I figure that either the amplifier gain is set low in my camera, or something is acting as a neutral density filter. If it were the latter, I don't think I'd have the noise "benefit".

Holden said:

I didnt shoot for white I shot for medium gray as
this is what sunny 16 was for after all...

Click to expand...

Sunny f16 should result in grey items in direct sunlight in the middle of the histogram, white objects near the top, white objects in the shade near the middle, etc. I only get bright white cars registering about a stop from the top of the histogram on the back of the camera, and they are the only things registering in the top half. The blue sky registers about a stop below the middle.

Holden said:

the slight differences I
had from my Gossen meter would be due to the fact that the scene
was mixed light/full sunlight and I had the chance to use both
incident and reflected light.

Click to expand...

I'm not sure why you're mentioning this. I'm talking about taking a picture with the camera set to ISO 100, Manual mode, 1/100, and f16, and taking a shot of mixed sun and shade (at mid-day with distinct shadows) with things of various tones in the sunlight. I checked the Sekonic against the the camera's metering in a separate test of a blank white wall with even illumination, and clear blue sky. The camera called for at least 2x the exposure in both cases.

Holden said:

If I had the time I would have set one of my gray cards down as
well... BTW I had the WB set to auto as well... not that it makes a
difference... but I was really quite amazed at how precise the 10D
was...

Click to expand...

--
John

E Scott Cragg said:

Its a base 2 logrithmic scale, 128 is 1ev from 255
8,16,32,64,128,255

middle of the histogram would be 32~64

Click to expand...

No, the middle of the histogram is 128 in some cases, and 117 in others. It is never as low as 32 to 64. 32 is shadowland.

Histograms are gamma 2.2 corrected, and are natural to our subjective experience of light and range of tones.

Besides, normally exposed images center around the middle of the histogram.

--
John