Why bother changing base ISO?

Started Jul 18, 2007 | Discussions
cycle61 Senior Member • Posts: 1,550
Re: To the Left to the Right

Stan-o-Stan wrote:

We are taught to expose to the right as far as poss without blowing
highlights.
The right side retains far more information than the left.

In digital form, this is true, and is the primary reason for the need to expose to the right. Miss the top of the histogram by 1 stop, and you've thrown away half your potential tonal range.

This idea says (please correct me if i'm wrong) - actually you can
expose to the left and pull it back after.

Not without some significant loss, I still believe.

I know the OP said it 's not an ideal but rather than loosing a shot
etc. but the thread has developed into a high asa versus lower and
exposed to the left idea - i think ?

I'm with the OP on not losing a shot. However, I think the theory can be carried to an extreme, and that may be what's happening here.

So, what is it? that after 800asa under expose rather than increase
the asa?
I have a D200. I ask this more from curiosity than necesity.

The link somebody posted above is fascinating and very informative with regards to this. http://www.clarkvision.com/imagedetail/digital.sensor.performance.summary/

It indicates, after thorough analysis, that for the D200 at ISO 800 a single electron is equal to one step of 4096 on a 12 bit scale. This means that if you have to go above ISO 800 to expose to the right, you have more tonal levels available in digital than you have electrons, so amplifying the signal any more prior to conversion won't help. I still think that up to 800, you're going to be better off with analog amplification to achieve your full tonal range. I've tried some test shots, but I don't have them all processed yet.
--
-Nick Davis
Please feel free to critique anything I post. I'm here to learn.

My galleries, such as they may be...
http://www.pbase.com/cycle61

rspino Contributing Member • Posts: 729
Re: To the Left to the Right

from my understanding of what julia borg said in the past, basically you are underexposing even if you cranck up the iso.

in other words, a f/5, 1/250, ISO400 shot and a F/5, 1/250, ISO 1600 one are the very same, exposure wise.

moreover, it would appear that using a lower iso, and correcting in post, be better in terms of dinamic range and color rendition.

the technical details are not quite clear to me, but from what i grasped apparently each camera has its now base iso, and changing it will only result in signal amplification, whereas the raw data will remain the same.

in other words, exposure is always given by the combination of aperture, shutter speed and base iso.

there was a lot of controversy on this, and i recall many tested this.

i came to the conclusion that julia is more than probably right. however, the advantages are probably too limited for a hobbyst like myself to justify caring too much.

oh, by the way: the white balance setting is of great importance just as well!

-- hide signature --

Roberto (equipment in my profile)
http://spino.smugmug.com

Stan-o-Stan Senior Member • Posts: 1,104
Re: To the Left to the Right

cycle61 wrote:

The link somebody posted above is fascinating and very informative
with regards to this.
http://www.clarkvision.com/imagedetail/digital.sensor.performance.summary/
It indicates, after thorough analysis, that for the D200 at ISO 800 a
single electron is equal to one step of 4096 on a 12 bit scale. This
means that if you have to go above ISO 800 to expose to the right,
you have more tonal levels available in digital than you have
electrons, so amplifying the signal any more prior to conversion
won't help. I still think that up to 800, you're going to be better
off with analog amplification to achieve your full tonal range. I've
tried some test shots, but I don't have them all processed yet.

Hi Nick,

thanks for your responce. I followed up to 'analog amplification'. Could you please explain? I thought the whole process was digital from camera to computer?

-- hide signature --

Stan-o-Stan

Iusefilm Senior Member • Posts: 1,143
Using your head is always a good thing!

Thank you Jim for your post.

I have similar problems with indoor flash photos. Fortunately I have had the opportunity to experiment with I-ttl, without ruining anybody"s wedding further.

My formula is, as always against the "grain" here, includes a secret step in post to gain one to one and a half stop of exposer without digital penalties, like blown highlights.

My D2 can not be used on Auto anything or Matrix and BL, so manual controls suite me just fine.
Tony K

OP Jim Kaye Senior Member • Posts: 2,794
I'll second Tom's thanks, Marianne (nt)
-- hide signature --

Jim Kaye

 Jim Kaye's gear list:Jim Kaye's gear list
Nikon D7500 Nikon D750 Nikon D500 Nikon D810 Fujifilm X-Pro2 +2 more
OP Jim Kaye Senior Member • Posts: 2,794
Re: To the Left to the Right

I'm with the OP on not losing a shot. However, I think the theory can
be carried to an extreme, and that may be what's happening here.

Yes, perhaps I was being a bit too provocative in my original post. But I was interested in what some of the people who know more about the theory and the hardware/firmware/sortware than I do would say in response.

The link somebody posted above is fascinating and very informative
with regards to this.
http://www.clarkvision.com/imagedetail/digital.sensor.performance.summary/
It indicates, after thorough analysis, that for the D200 at ISO 800 a
single electron is equal to one step of 4096 on a 12 bit scale. This
means that if you have to go above ISO 800 to expose to the right,
you have more tonal levels available in digital than you have
electrons, so amplifying the signal any more prior to conversion
won't help. I still think that up to 800, you're going to be better
off with analog amplification to achieve your full tonal range. I've
tried some test shots, but I don't have them all processed yet.

Thanks for that clear explanation of what I think is the concept of "unity gain," Nick.
--
Jim Kaye

 Jim Kaye's gear list:Jim Kaye's gear list
Nikon D7500 Nikon D750 Nikon D500 Nikon D810 Fujifilm X-Pro2 +2 more
Julia Borg Veteran Member • Posts: 7,280
Re: Why bother changing base ISO?

my findings were that raw files taken with my D2X are actually better
when I use a lower ISO, and underexpose by one stop intentionally,
then "push" one stop in Capture.

I think Julia has reported this to hold true for anything past ISO 320
on the D2X,

yes. with other cameras results vary. results also vary depending on the raw converter is use (depending on the precision of calculations used for eV "compensation" in raw converter).

I think the proper way to determine the ISO limit is to do a set of simple tests under daylight and incandescent lights using the raw converter of the choice.

one of the biggest disadvantages of bumping ISO in camera is that with higher ISO settings we are getting more highlight clipping. independent of ISO setting, sensor counts photons. for linear analog-to-digital converters if some well counts to the half of its capacity plus 1 and ISO is bumped 2x from "native" ISO, this well count is going to be recorded as overflown and truncated; that is, highlights are clipped.

however there are indications that linear conversion will be substituted with a more suitable "gamma-based" conversion in near future; thus highlights will be compressed instead of clipping.

-- hide signature --

Julia

OP Jim Kaye Senior Member • Posts: 2,794
Re: Why bother changing base ISO?

Is this with or without boosting the impoverished blue
channel signal found in that environment? That is, by using
something like an 80A blue filter to correct the colors back
closer to daylight (as with film)?

I hear what you're saying, Tom, but the thing I've never understood about using filters with digital like this is that the blue filter cuts down the transmission of other (i.e., red) wavelengths, it doesn't "boost" the transmission of the ones you want (blue). So unless you increase the actual exposure (slower shutter speed and/or wider aperture) when you use the filter, I don't see how you would end up with any more blue photons hitting the sensor than without the filter. And if there's enough light to increase the exposure, then why not just do that?

Maybe because increasing exposure without using the blue filter would blow out the red channel in some areas of the scene. Then the blue filter would make sense, I guess, but only to prevent blowing out the red channels, not to "boost" the sensitivity to blue light. (It's increasing the exposure that "boosted" the blue light coming through.)

But if the light overall is very low (for the shutter speed and aperture you need), I don't see the advantage. If you don't also increase the exposure, you won't end up with any less noise in the blue channel, it seems to me.
--
Jim Kaye

'I believe that the electronic image will be the next major advance. Such systems will have their own inherent and inescapable structural characteristics, and the artist and functional practitioner will again strive to comprehend and control them.' -- Ansel Adams, 1981

 Jim Kaye's gear list:Jim Kaye's gear list
Nikon D7500 Nikon D750 Nikon D500 Nikon D810 Fujifilm X-Pro2 +2 more
snorri Regular Member • Posts: 325
Re: To the Left to the Right

from my understanding of what julia borg said in the past, basically you are
underexposing even if you cranck up the iso.

That's the way I understand it, too.

Exposing for X seconds at aperture Y means collecting exactly those Z photons that manage to pass through that aperture during that time. Cranking up the ISO value simply tells the RAW converter to interpret the photon count Z as "bright" that would otherwise be interpreted as "dark". Its like pouring water from a large jug into a small glass: The same amount of water that fills the glass barely covered the bottom of the jug. The ISO value makes no difference during capture, and I don't see a reason why there should be any difference in the RAW data. AFAIK, it is simply a parameter of interpreting the data after the shot.

-- hide signature --

snorri

--
'... haben sind gewesen gehabt haben geworden sein.'

 snorri's gear list:snorri's gear list
Fujifilm X20 Nikon D800
Julia Borg Veteran Member • Posts: 7,280
Re: You achieve better amplitude resolution

Marianne Oelund wrote:

Jim Kaye wrote:

Yes, and to reply to Peter's comment directly, the question boils
down to this: if the image is not going to be "properly exposed" one
way or the other (because I "need" a certain shutter speed and
aperture) how much difference does it really make if I amplify the
signal (and the noise) in the camera or in post?
--

Ideally, we would have converters which are matched to the sensor's
amplitude resolution, which is typically 30,000 to 80,000 levels (the
electron count at full well capacity). This would require 15- to
16-bit converters. If we had that, your point would be completely
valid and there would be little value in using ISO settings above
base.
However, our converters (Nikon) are limited to 12 bits (4096 levels),
so they fall short of the resolution contained in the sensor output
by about 8:1 up to about 12:1.

two factors make the lowest bits useless - noise added to the signal before raw data is recorded and noise added during conversion of raw data to RGB "gamma-corrected" stage. even in MkIII they can't get full direct benefit from 14-bit converter.

dynamic range of 8 stops in the green channel is still fairly typical.

-- hide signature --

Julia

Julia Borg Veteran Member • Posts: 7,280
Re: Why bother changing base ISO?

the thing I've never understood
about using filters with digital like this is that the blue filter
cuts down the transmission of other (i.e., red) wavelengths,

look at it from a different perspective. shooting under incandescent light you often have attenuation of red channel to achieve the white balance. so using cyan filter which affects only red channel and allows for better exposure of green and blue channels makes certain sense.

-- hide signature --

Julia

Stan-o-Stan Senior Member • Posts: 1,104
Re: To the Left to the Right

Sounds good, but why don't we have an 'iso slider' with our RAW converters?
seems like we should, or is that what the exposure slider is doing to the data?

cheers
--
Stan-o-Stan

Raul Veteran Member • Posts: 8,438
Amazing

Julia Borg wrote:

the thing I've never understood
about using filters with digital like this is that the blue filter
cuts down the transmission of other (i.e., red) wavelengths,

look at it from a different perspective. shooting under incandescent
light you often have attenuation of red channel to achieve the white
balance. so using cyan filter which affects only red channel and
allows for better exposure of green and blue channels makes certain
sense.

Anyone buying blue filters (and a color temp meter) ?

regards
raul

 Raul's gear list:Raul's gear list
Nikon Df Nikon D850 Nikon D5
Raul Veteran Member • Posts: 8,438
I have an arca and one other

hmmm

raul

 Raul's gear list:Raul's gear list
Nikon Df Nikon D850 Nikon D5
jean bernier Veteran Member • Posts: 3,181
Re: Amazing

If we're talking handheld shooting under low intensity tungsten, such as stage shooting, a blue or cyan filter is a very good case of diminishing returns. Just a one stop loss may well make the job impossible. Having to drop from 1/60 to 1/30 of a second has very serious consequences. In order to get some less noise in the blue channel? Gimme a break....

But I can see an advantage when shooting interiors under tungsten light, tripoded. BTW if someone knows why the dark frame substraction only starts at 1/2 sec shutter (if memory serves me well) on the D2X, please let me know...
--
Jean Bernier

All photographs are only more or less credible illusions

OP Jim Kaye Senior Member • Posts: 2,794
Re: To the Left to the Right

Sounds good, but why don't we have an 'iso slider' with our RAW
converters?
seems like we should, or is that what the exposure slider is doing to
the data?

I think that is exactly what it is doing -- emulating what in increase in ISO would have done on film. Nothing you can do in post can possibly affect how many photons really hit the sensor -- so the term "exposure" adjustment in postprocessing is really a misnomer.

But we think of ISO and "exposure" as varying inversely -- decrease the ISO one stop and increase the exposure one stop (slower shutter speed and/or wider aperture) to make a film negative have the same density as you started out with. So since neither term ("exposure" or "ISO" adjustment) would be entirely correct (we're simulating film with digital capture, anyway), I suppose one's as good as the other.
--
Jim Kaye

'I believe that the electronic image will be the next major advance. Such systems will have their own inherent and inescapable structural characteristics, and the artist and functional practitioner will again strive to comprehend and control them.' -- Ansel Adams, 1981

 Jim Kaye's gear list:Jim Kaye's gear list
Nikon D7500 Nikon D750 Nikon D500 Nikon D810 Fujifilm X-Pro2 +2 more
OP Jim Kaye Senior Member • Posts: 2,794
Re: Using your head is always a good thing!

Thank you Jim for your post.

Oh, you're welcome, Tony.

I have similar problems with indoor flash photos. Fortunately I have
had the opportunity to experiment with I-ttl, without ruining
anybody"s wedding further.

My formula is, as always against the "grain" here, includes a secret
step in post to gain one to one and a half stop of exposer without
digital penalties, like blown highlights.

My D2 can not be used on Auto anything or Matrix and BL, so manual
controls suite me just fine.

I'm not sure I get what you mean in saying that your camera "can not be used" on "Auto anything" -- do you mean that you aren't happy with the results you get that way?
--
Jim Kaye

 Jim Kaye's gear list:Jim Kaye's gear list
Nikon D7500 Nikon D750 Nikon D500 Nikon D810 Fujifilm X-Pro2 +2 more
Julia Borg Veteran Member • Posts: 7,280
Re: Amazing

jean bernier wrote:

If we're talking handheld shooting under low intensity tungsten, such
as stage shooting, a blue or cyan filter is a very good case of
diminishing returns. Just a one stop loss may well make the job
impossible.

with cyan filter the loss occurs only in the red channel which is overloaded anyway.

-- hide signature --

Julia

cycle61 Senior Member • Posts: 1,550
Re: To the Left to the Right

Stan-o-Stan wrote:

Hi Nick,

thanks for your responce. I followed up to 'analog amplification'.
Could you please explain? I thought the whole process was digital
from camera to computer?

Stan,

The very first step, where the signal from the sensor is amplified as it offloads into the converter, is the only analog step in the process. This is where you benefit from amplifying this signal to use the whole 4096 steps of tonal range available from 12 bit digital. The analysis linked to previously rates the D200 sensor "well" capacity at around 32,000 electrons per photosite, so working with that signal gives you potentially that many tonal steps. Once the A/D converter does its thing, you're down to 4096, and if the analog signal is too weak to use the whole digital range, you end up with potentially far fewer levels than that. I think the important concept is "unity gain", which seems to be at ISO800 for the D200. My test shots I mentioned earlier were in bright daylight, at f/8 and 1/1250th second, for a proper exposure at ISO 800. I then turned the ISO down to 400, 200, and 100 without changing anything else. Results: ISO 800 is visibly noisy on the D200 (we knew that already) ISO 400, 200, and 100 look similar once boosted up to match, but ISO 100 actually looked the worst, being the most underexposed. I'm going to have to try again including 1600 and 3200. By what I understand, you benefit up to ISO 800 by the analog amplification, but beyond that there are more levels than electrons, and there's no benefit to boosting ISO further.
Test shots coming up in an hour or so...
--
-Nick Davis
Please feel free to critique anything I post. I'm here to learn.

My galleries, such as they may be...
http://www.pbase.com/cycle61

Bill Janes Senior Member • Posts: 1,848
Re: You achieve better amplitude resolution

Marianne Oelund wrote:

Ideally, we would have converters which are matched to the sensor's
amplitude resolution, which is typically 30,000 to 80,000 levels (the
electron count at full well capacity). This would require 15- to
16-bit converters. If we had that, your point would be completely
valid and there would be little value in using ISO settings above
base.
However, our converters (Nikon) are limited to 12 bits (4096 levels),
so they fall short of the resolution contained in the sensor output
by about 8:1 up to about 12:1. Thus when you are not using the full
well capacity of the sensor, it makes sense to increase the analog
gain upstream of the converter. Failing to do so simply deprives you
of the amplitude (tonal) resolution you can obtain with the higher
ISO settings. Depending on the subject nature, you may or may not
notice this loss of tonal resolution.

Marianne,

Your analysis is lucid, but I must take issue on one point and I will present data that expand on your analysis. The full well of a camera may be 30,000 electrons, but the camera can not resolve 30,000 discrete levels because of noise. The following table shows the noise characteristics of a Nikon D200 camera according to the data on Roger Clark's web site.

The top line of the chart represents the data for full well exposure at base ISO and the remaining lines show the values for exposure at ISO 1600. The main components are the shot noise (photon counting statistics) and read noise and these are shown along with the total noise and the signal to noise ratio.

At base ISO the camera can use the full well capacity of 33,000 electrons. However, since the standard deviation of the total noise is 181 electrons, the camera can not resolve 33,000 discrete levels and it is not necessary for the camera to count every last electron.

At ISO 1600, the full well of the sensor is not used, and the camera gain is such that 2043 electrons represents full scale on the ADC. The number of electrons for f/stops below full scale are shown along with the noise characteristics. Six stops down, we collect only 32 electrons and the standard deviation is 9.3 electrons. The camera gain is about 0.5 electrons/ADU (analog to digital units or raw pixel values), and noise rather than ADU resolution will limit effective resolution.

If we left the camera ISO setting at 100, the gain would be 7.98 electrons per ADU and this is the effective resolution; to quantify 32 electrons in the shadow area, we obviously need better resolution. However, the noise of 9.3 electrons will limit effective resolution in the shadows. As you noted, there is little benefit of using an ISO above unity gain, which is about 800 for the D200.

Noise limits the dynamic range at high ISO, and the effective photographic dynamic range depends on how much noise you are able to tolerate. According to Norman Koren's Imatest, the D200 has only 3 stops dynamic range at ISO 1600 for high quality results, but nearly 7 stops of DR for low quality results. This limitation is shown on the table for the limiting shadow value.

-- hide signature --

Bill Janes

Keyboard shortcuts:
FForum MMy threads