Pages: [1] 2   Go Down

Author Topic: Camera-to-Print -- exposing to the right question  (Read 21571 times)

Gregory

  • Full Member
  • ***
  • Offline Offline
  • Posts: 191
    • http://www.gregory.hk
Camera-to-Print -- exposing to the right question
« on: August 08, 2007, 12:48:34 am »

I was a little confused by one statement in the tutorial regarding exposing to the right (which I'm familiar with).

Quote
...you'll get a better result by increasing the exposure a little bit to the right, which is the same thing as lowering the ISO...
you shouldn't follow the 'expose to the right' religion if you're shooting at high ISO. there's no point in shooting at ISO 400 and exposing to the right. you may as well drop to ISO 200.
I don't understand this. the purpose of exposing to the right is to get more levels in the middle of the histogram because the first stop provides 4096 levels, the second stop 2048, etc. if you shoot at ISO 800 and expose to the right, you're getting more levels for the middle and low regions. how does this equate to shooting at ISO 400 without exposing to the right in terms of levels and detail? i.e., ISO 800 ev +1.0 == ISO 400 ev 0.0

please explain. details and levels in the lower region of the histogram are important to me because many of the birds I photograph have dark regions which don't contain as much detail as I would prefer.

and... is this level structure carried into the RAW processor and beyond? i.e., after processing, is the lowest stop still restricted to only 128 levels?

and... the Canon Mark III shoots in 14-bit (from Canon's page: "The extra power of dual DIGIC III processors has also allowed analog-to-digital conversion to improve from 12 to 14 bits per channel, meaning better tonal gradation for RAW images."). would this increase the levels of the lower stop in the histogram by 2^2 from 2^7 to 2^9 == 512 levels?

sincerely,
Gregory
« Last Edit: August 08, 2007, 12:55:20 am by Gregory »
Logged
Gregory's Blog: [url=http://www.gregory.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Camera-to-Print -- exposing to the right question
« Reply #1 on: August 08, 2007, 08:57:51 am »

Lets say you find that exposing plus one stop over the meter does indeed give you plenty of highlight detail to pull back in a Raw converter (I've actually found 1 ½ stops on my 5D). OK so you're starting out at ISO 400 at F8 based on the meter but using ETTR which tells you you'd get better data at F5.6. Well isn't that the same as shooting at ISO 200? If you have a scene that requires you to shoot at F8, not 5.6, well ETTR isn't going to fly for you.

Where I see ETTR really making a big difference is when I set the ISO higher and expose to the right which is the equivalent of using a lower ISO setting in practicality.

Extra bit depth in the new Canon doesn't necessarily mean that you have less noise or more dynamic range. You just have more levels in whatever you shot. That could be more levels of noise in theory.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Gregory

  • Full Member
  • ***
  • Offline Offline
  • Posts: 191
    • http://www.gregory.hk
Camera-to-Print -- exposing to the right question
« Reply #2 on: August 08, 2007, 09:39:44 am »

"Lets say you find that exposing plus one stop over the meter does indeed give you plenty of highlight detail to pull back in a Raw converter."

I'm not looking for more detail in the highlights. I want more detail in the lowlights, hence ETTR.


"I've actually found 1 ½ stops on my 5D."

how did you find that extra 1 ½ stops? I'd be interested to know. in my images, any highlights above 255 are blown out and irretrievable.


"OK so you're starting out at ISO 400 at F8 based on the meter but using ETTR which tells you you'd get better data at F5.6. Well isn't that the same as shooting at ISO 200?"

I don't think it is the same. in terms of overall exposure, it's the same. in terms of levels per stop, it's different, at least from my understanding.

let's see if I can try to explain what I'm referring to. assuming 6 stops range in my image:

       0                                      255
stop   |  6  |  5  |  4   |   3  |   2  |   1  |
levels | 128 | 256 | 512  | 1024 | 2046 | 4096 | ISO 400 ev 0.0
levels | 128 | 256 | 512  | 1024 | 2046 | 4096 | ISO 800 ev +1.0
levels | 256 | 512 | 1024 | 2046 | 4096 |   ?  | ISO 800 ev +1.0 adjusted back to ev 0.0 in the RAW processor
so by using ETTR, I get twice as many levels in the lowest 3 stops of data and therefore more detail in the dark areas of my images which for me is relatively important.

I could be completely wrong. one thing I'm very unsure of is whether the extra levels are preserved in the RAW Processor's output. is the converted file made up of logarithmic data which will always only preserve 128 levels in stop 6 or is the converted file made up of linear data or... ?

if I'm missing the mark, please enlighten me. I need to understand this.

sincerely,
Gregory
Logged
Gregory's Blog: [url=http://www.gregory.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Camera-to-Print -- exposing to the right question
« Reply #3 on: August 08, 2007, 09:53:16 am »

Quote
"Lets say you find that exposing plus one stop over the meter does indeed give you plenty of highlight detail to pull back in a Raw converter."
Quote
I'm not looking for more detail in the highlights. I want more detail in the lowlights, hence ETTR.

In order to use ETTR, you have to find the point of full sensor saturation and back off a tad. So you need to use the adjustments in the Raw processor to see if you can get just shy of 255 in all three rendered channels. I'm not referring to highlight detail, I'm referring to shooting so that specular highlights might be 255 but anything you wish to reproduce is just below that. ETTR isn't about over exposure, its about proper exposure and that means deciding where you want to clip and not clip. The results are more data in the shadows.


Quote
how did you find that extra 1 ½ stops? I'd be interested to know. in my images, any highlights above 255 are blown out and irretrievable.

Use a very white, spectrally neutral target, setup lighting and exposure tests (bracket). Bring images into Raw converter and find image where you can neutralize exposure to be at the ideal ETTR values (not clipping highlight data you want).

Quote
I don't think it is the same. in terms of overall exposure, it's the same. in terms of levels per stop, it's different, at least from my understanding.

OK so my meter tells me that a normal exposure is F8 at 125th. I know I can shoot at F5.6 at 125th and not blow out highlights and the net result is more data in the shadows due to the ETTR. IF I set my external meter to a stop less ISO, it would have indicated I should shoot at F 5.6 at 125th right?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Gregory

  • Full Member
  • ***
  • Offline Offline
  • Posts: 191
    • http://www.gregory.hk
Camera-to-Print -- exposing to the right question
« Reply #4 on: August 08, 2007, 10:06:42 am »

thank you digitaldog for trying to explain this to me.


I'm referring to shooting so that specular highlights might be 255 but anything you wish to reproduce is just below that. ETTR isn't about over exposure, its about proper exposure and that means deciding where you want to clip and not clip.

yep. that's how I understood it and that's how I try to shoot when possible.


OK so my meter tells me that a normal exposure is F8 at 125th. I know I can shoot at F5.6 at 125th and not blow out highlights and the net result is more data in the shadows due to the ETTR. IF I set my external meter to a stop less ISO, it would have indicated I should shoot at F 5.6 at 125th right?

right, but the histogram would be in the middle instead of to the right and you'd have less detail in the shadows. right?
Logged
Gregory's Blog: [url=http://www.gregory.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Camera-to-Print -- exposing to the right question
« Reply #5 on: August 08, 2007, 10:13:41 am »

Quote
right, but the histogram would be in the middle instead of to the right and you'd have less detail in the shadows. right?


The on camera histogram doesn't show you the Raw data but instead what the JPEG would look like if you shot this way and let the camera process the data as a JPEG. So, not real useful!

The idea is to put as many real levels of data in the last stop, that means moving every level over to the right without clipping highlights you wish to reproduce.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Gregory

  • Full Member
  • ***
  • Offline Offline
  • Posts: 191
    • http://www.gregory.hk
Camera-to-Print -- exposing to the right question
« Reply #6 on: August 08, 2007, 10:27:12 am »

The idea is to put as many real levels of data in the last stop, that means moving every level over to the right without clipping highlights you wish to reproduce.

that's my point. with an ISO setting one stop lower and using F5.6, this wouldn't happen.

ISO 800 ev +1.0 --> F5.6 1/125
--> levels moved to the right to get more levels in the shadows.

ISO 400 ev +0.0 --> F5.6 1/125
--> same exposure but the sensor is 'less sensitive' so the histogram is in its normal 'middle' position rather than to the right and there are less details in the shadows.

I feel like I'm missing something, possible something really obvious to everyone but me ;-)
Logged
Gregory's Blog: [url=http://www.gregory.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Camera-to-Print -- exposing to the right question
« Reply #7 on: August 08, 2007, 12:11:51 pm »

In both scenario’s the ISO of the camera is the same. You set it for ISO 100, you get a fixed amount of noise at normal exposure (normal being what the meter tells you, not based on ETTR). You keep the unit at ISO 100 but open one stop (the effective equivalent of shooting at at ISO 50). You don't blow out highlights and move more levels from last stop of shadow to first stop highlight, the result being less noise in that last stop.

I think the point made on the DVD is if you need higher ISO, its because you don't have enough light and using ETTR reduces the effectiveness here so get the shot and live with a bit more noise (more due to the exposure and the ISO).
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Gregory

  • Full Member
  • ***
  • Offline Offline
  • Posts: 191
    • http://www.gregory.hk
Camera-to-Print -- exposing to the right question
« Reply #8 on: August 08, 2007, 12:51:34 pm »

and then it's a matter of balancing detail and levels in the shadows with ETTR against the extra noise that might come with the higher ISO.

thank you again digitaldog. you've been patient and helpful.

regards,
Gregory
Logged
Gregory's Blog: [url=http://www.gregory.

picnic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 574
Camera-to-Print -- exposing to the right question
« Reply #9 on: August 08, 2007, 12:57:48 pm »

Quote
and then it's a matter of balancing detail and levels in the shadows with ETTR against the extra noise that might come with the higher ISO.

thank you again digitaldog. you've been patient and helpful.

regards,
Gregory
[a href=\"index.php?act=findpost&pid=132157\"][{POST_SNAPBACK}][/a]

I'll say thank you too because I was a bit confused by what was said.   This clarifies it for me.  

Diane
Logged

KenL

  • Newbie
  • *
  • Offline Offline
  • Posts: 2
    • http://
Camera-to-Print -- exposing to the right question
« Reply #10 on: January 03, 2008, 10:58:22 pm »

Quote
In both scenario’s the ISO of the camera is the same. You set it for ISO 100, you get a fixed amount of noise at normal exposure (normal being what the meter tells you, not based on ETTR). You keep the unit at ISO 100 but open one stop (the effective equivalent of shooting at at ISO 50). You don't blow out highlights and move more levels from last stop of shadow to first stop highlight, the result being less noise in that last stop.

I think the point made on the DVD is if you need higher ISO, its because you don't have enough light and using ETTR reduces the effectiveness here so get the shot and live with a bit more noise (more due to the exposure and the ISO).
[a href=\"index.php?act=findpost&pid=132153\"][{POST_SNAPBACK}][/a]

Thanks for this discussion.

I just tried a little experiment:

I set the camera on a tripod in front of an evenly lite wall and took shots at ISO 100  @+1ev and ISO 400 @+1ev. Looking at the histogram for each they are very similar.

From the discussion above I was expecting ISO 100 @+1ev to have a similar histogram to ISO 400 @+2ev or +3ev.

Am I missing something?

Ken
Logged

kaelaria

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2223
    • http://www.bgpictures.com
Camera-to-Print -- exposing to the right question
« Reply #11 on: January 03, 2008, 11:29:42 pm »

1 stop is 2x or 1/2 the light - same as one ISO step.
Logged

larsrc

  • Full Member
  • ***
  • Offline Offline
  • Posts: 173
    • http://
Camera-to-Print -- exposing to the right question
« Reply #12 on: January 04, 2008, 09:02:09 am »

Quote
I was a little confused by one statement in the tutorial regarding exposing to the right (which I'm familiar with).

Quote
...you'll get a better result by increasing the exposure a little bit to the right, which is the same thing as lowering the ISO...
you shouldn't follow the 'expose to the right' religion if you're shooting at high ISO. there's no point in shooting at ISO 400 and exposing to the right. you may as well drop to ISO 200.
I don't understand this. the purpose of exposing to the right is to get more levels in the middle of the histogram because the first stop provides 4096 levels, the second stop 2048, etc. if you shoot at ISO 800 and expose to the right, you're getting more levels for the middle and low regions. how does this equate to shooting at ISO 400 without exposing to the right in terms of levels and detail? i.e., ISO 800 ev +1.0 == ISO 400 ev 0.0

[a href=\"index.php?act=findpost&pid=132070\"][{POST_SNAPBACK}][/a]

My understanding of that bit was that if you ETTR into high enough ISO that the camera itself pushes the exposure, you won't get any more detail than if you kept within the range where the ISO is "real".  Having the camera multiply the readings after the A/D conversion doesn't give you any more shadow detail. However, I don't know what modern camera it would be where ISO 400 is pushed, so it puzzled me a bit, too.

-Lars
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Camera-to-Print -- exposing to the right question
« Reply #13 on: January 04, 2008, 11:12:40 am »

Quote
In order to use ETTR, you have to find the point of full sensor saturation and back off a tad. So you need to use the adjustments in the Raw processor to see if you can get just shy of 255 in all three rendered channels. I'm not referring to highlight detail, I'm referring to shooting so that specular highlights might be 255 but anything you wish to reproduce is just below that. ETTR isn't about over exposure, its about proper exposure and that means deciding where you want to clip and not clip. The results are more data in the shadows.
[a href=\"index.php?act=findpost&pid=132120\"][{POST_SNAPBACK}][/a]

If ETTR requires exposing to just short of sensor saturation, then ETTR applies only at base ISO. When you increase ISO 1 EV over base, then the sensor would only be half saturated and clipping would occur in the ADC rather from saturation. The same considerations apply to even higher ISOs.

If you are shooting at ISO 400 rather than ISO 200, then presumably you need a higher shutter speed, smaller aperture, or both. If ETTR requires a positive exposure adjustment of +1 EV then you are back to where you started with ISO 200 exposure settings. However, if you decide to use ISO 200, then the histogram would indicate that an exposure adjustment of +1 EV is still needed, so the same argument would indicate that you should drop back to ISO 100.

One does not need to invoke religion into ETTR theory. The physical principles are well understood. Under normal photographic conditions, noise in a digital image is composed almost entirely of shot noise (photon noise, related to exposure, and read noise). In the case of the ISO 200 exposure without ETTR and ISO 400 exposure with +1 EV, the exposure is the same and the shot noise will be the same. Read noise varies with the camera and ISO. According to Roger Clark's data for the Canon 1D MII, read noise at ISO 200 is 8.95 electrons and at ISO 400, it is 5.56 electrons. These differences could result in small differences in the deep shadows. However, the ISO 200 exposure without ETTR compensation will have more headroom in the highlights. Overall, an objective analysis confirms Jeff's statement.
Logged

Gregory

  • Full Member
  • ***
  • Offline Offline
  • Posts: 191
    • http://www.gregory.hk
Camera-to-Print -- exposing to the right question
« Reply #14 on: January 04, 2008, 12:09:06 pm »

Quote
If you are shooting at ISO 400 rather than ISO 200, then presumably you need a higher shutter speed, smaller aperture, or both. If ETTR requires a positive exposure adjustment of +1 EV then you are back to where you started with ISO 200 exposure settings. However, if you decide to use ISO 200, then the histogram would indicate that an exposure adjustment of +1 EV is still needed, so the same argument would indicate that you should drop back to ISO 100.
confused!!! let me see if I can understand this.

"exposure settings" means aperture and shutter speed without regard for the ISO speed? so:

with ISO 400, the camera calculates that correct exposure would require f8 @ 1/125.
you add 1 EV by increasing the aperture to f5.6.

with ISO 200, the camera calculates the correct exposure would require f5.6 @ 1/125; i.e., the same aperture and shutter speed as that required at ISO 400 +1 EV.

SO... to understand this, I need to know how ISO speed affects the light received by the sensor and how it affects the number of levels per stop. to my uneducated mind, your explanation seems to infer that the sensors pick up the same amount of light and levels per stop regardless of the ISO speed, and that the ISO speed affects the ADC's 'multiplier' effect rather than the sensors; i.e., how it multiplies the original analog signal to produce the final digital data.

further, if it is correct that the sensors pick up the same amount of light regardless of the ISO speed, then it follows that (as bjanes has stated) ETTR can only be used with benefit at the base (true) ISO speed, because only then can more light be directed to the lower stops of the image. higher ISO speeds increase the apparent exposure by multiplying the original signal, and exposure is calculated incorporating that multiplication factor. consequently, using a higher ISO speed may in fact reduce the amount of light and detail in the lower stops of the image rather than increase it. i.e., if you need to shoot with a higher ISO speed because of a minimum aperture or shutter speed requirement, then forget ETTR, "decide[ing] where you want to clip and not clip" and shoot accordingly.

whew!

(corrections definitely welcome)
« Last Edit: January 04, 2008, 12:11:07 pm by Gregory »
Logged
Gregory's Blog: [url=http://www.gregory.

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Camera-to-Print -- exposing to the right question
« Reply #15 on: January 04, 2008, 01:41:28 pm »

Quote
confused!!! let me see if I can understand this.

"exposure settings" means aperture and shutter speed without regard for the ISO speed? so:

with ISO 400, the camera calculates that correct exposure would require f8 @ 1/125.
you add 1 EV by increasing the aperture to f5.6.

with ISO 200, the camera calculates the correct exposure would require f5.6 @ 1/125; i.e., the same aperture and shutter speed as that required at ISO 400 +1 EV.

SO... to understand this, I need to know how ISO speed affects the light received by the sensor and how it affects the number of levels per stop. to my uneducated mind, your explanation seems to infer that the sensors pick up the same amount of light and levels per stop regardless of the ISO speed, and that the ISO speed affects the ADC's 'multiplier' effect rather than the sensors; i.e., how it multiplies the original analog signal to produce the final digital data.
[a href=\"index.php?act=findpost&pid=165014\"][{POST_SNAPBACK}][/a]

I think your assumptions are true here. If the shutter speed and f/stop are the same (and other factors are the same), then the sensor will receive the same amount of light regardless of the ISO setting. Increasing the ISO does not change the amount of light received by the sensor, but merely increased the analog amplification so that the ADC is operating at full scale.

In general, you lose DR when you increase ISO. This can be seen from the  Excel plots derived from Roger Clark's data for the 1D MII. The base ISO for that camera is not 100, but is used here because of peculiarities of that camera. The DR is determined by the noise floor. If you require a Signal to Noise (SN) of 5 in the deepest shadows, then the DR at ISO 100 would be slightly better than 9 f/stops and that at ISO 400 would be slightly less than 8 stops.

The number of levels is not taken into account. DR is usually limited by noise and not posterization in the shadows.

ShotN = shot noise, ReadN = read noise, TN = total noise, S/N = signal / noise. The amount of light received by the sensor is directly proportional to the electron count.



« Last Edit: January 04, 2008, 01:48:29 pm by bjanes »
Logged

Schewe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6229
    • http:www.schewephoto.com
Camera-to-Print -- exposing to the right question
« Reply #16 on: January 04, 2008, 02:54:25 pm »

Quote
I need to know how ISO speed affects the light received by the sensor and how it affects the number of levels per stop.
[a href=\"index.php?act=findpost&pid=165014\"][{POST_SNAPBACK}][/a]


Just to be clear...changing the ISO setting will _NOT_ alter the amount of light falling on the sensor...if you think about, that would be SciFi because the camera can't alter the reality of a scene. So, setting F stop & shutter speed is the ONLY way the camera has to alter or control the amount of light hitting the sensor. If you change the ISO and keep the F stop & shutter the same, the same amount of light will be falling...only the ADC will alter the resulting capture.

The other thing you gotta understand is that ETTR is a technique to maximize the quality potential of a capture. ETTR isn't really something that is useful for an ISO other than the camera's optimum ISO. So, there would be no reason to increase the ISO and then also ETTR. You would be better off (I'm pretty sure) simply using a lower ISO.

Also, you must comprehend with ETTR is that it's really only applicable if the scene dynamic range fits within the dynamic range of the sensor. If you are shooting a really contrasty scene, ETTR won't really help the issue of noise in the shadow because you run the risk of blowing highlights.

For a lot of shooting, "correct exposure" for the scene is better than ETTR–it doesn't ALWAYS work. But if you are in a low contrast situation either outside or in a studio (easy to control contrast range by lighting and fill light), then ETTR can produce technically better images as it relates to noise. Note however that with the increased exposure you lose the potential for depth of field and faster shutter speed. If the shot demands either or both, ETTR will NOT result in a technically better "overall" shot.
« Last Edit: January 04, 2008, 02:56:24 pm by Schewe »
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Camera-to-Print -- exposing to the right question
« Reply #17 on: January 04, 2008, 03:06:42 pm »

Quote
The other thing you gotta understand is that ETTR is a technique to maximize the quality potential of a capture. ETTR isn't really something that is useful for an ISO other than the camera's optimum ISO. So, there would be no reason to increase the ISO and then also ETTR. You would be better off (I'm pretty sure) simply using a lower ISO.

That's what my tests showed. IF you need higher ISO (to capture the scene), you need a higher ISO.

But before we can set minimal or optimal ISO, we need to know the exposure sensitivity of the chip.

Doing that, then exposing for ETTR may not work however, the point about ETTR is about getting the best data possible using the ISO/Exposure/Aperture (its not about over exposing). We are lucky to be able to use various ISO's and exposure's within reason and get a Raw that we can massage into a good looking image. ETTR would simply place the most data within the image, doing so requires the proper setting of ISO. So I don't see ETTR and differing ISO as mutual exclusive, what I do see is this: If you need a higher ISO to capture the image, and you're basing this exposure on highlights (expose for highlights), you're setting said ISO based on the initial premise of ETTR. The proper exposure based on ISO requires you know the actual sensitivity of your chip based upon exposing for highlights such you produce as much useable data in the shadows as possible.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Camera-to-Print -- exposing to the right question
« Reply #18 on: January 04, 2008, 03:41:13 pm »

Quote
The other thing you gotta understand is that ETTR is a technique to maximize the quality potential of a capture. ETTR isn't really something that is useful for an ISO other than the camera's optimum ISO. So, there would be no reason to increase the ISO and then also ETTR. You would be better off (I'm pretty sure) simply using a lower ISO.

[a href=\"index.php?act=findpost&pid=165049\"][{POST_SNAPBACK}][/a]

And what is the optimum ISO of the camera--base ISO? In most cases base ISO will give the best results if one can deal with shutter speed/aperture constraints. However, astronomers will get better results at higher ISOs where read noise is lower.
« Last Edit: January 04, 2008, 04:14:02 pm by bjanes »
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Camera-to-Print -- exposing to the right question
« Reply #19 on: January 04, 2008, 05:22:34 pm »

Quote
That's what my tests showed. IF you need higher ISO (to capture the scene), you need a higher ISO.

The proper exposure based on ISO requires you know the actual sensitivity of your chip based upon exposing for highlights such you produce as much useable data in the shadows as possible.
[a href=\"index.php?act=findpost&pid=165054\"][{POST_SNAPBACK}][/a]

That is where ISO 12232:2006, the ISO standard for digital camera ISO, comes in. According to this specification, a standard light meter reading from an 18% target should give 12.7% sensor saturation.  With a 12 bit ADC the corresponding DN would be would be 522, corresponding to 101 in an 8 bit gamma 2.2 space. This standard leaves a headroom of 0.5 EV for specular highlights. If you want to take a highlight reading and place the highlights at 95% saturation (249 in a gamma 2.2 space), you would increase the exposure by 2.9 stops. Since in camera JPEGs and raw converters apply a saturation and contrast boost prior to encoding into a 2.2 gamma space, one should set the contrast and saturation to low to get a better idea of the data in the raw file.

Of course, some actual testing of your light meter and camera is indicated, but the above calculations could be a starting point. I would take a meter reading from a uniform reflecting surface (it can be gray or white), increase the indicated exposure by 3 stops, bracket up and down by 0.3 EV and use the compensation factor that gives a reading of 249 in a gamma 2.2 space.

You can leave the data on the CF card and then compare the histogram and blinking highlight data to know where you stand when using these tools on your camera. Taking of highlight readings required a spot reading, preferably at 1 degree.

I would venture to say that >90% of the members of this forum who use ETTR do so with ISOs above base and with the in camera light meter.
Logged
Pages: [1] 2   Go Up