determining native ISO of sensor

G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

Is there a way to determine the native ISO of the sensor in my Canon
20D? From what I've heard, it isn't necessarily the lowest ISO
setting. Someone said the sensor in the 10D was best at 200 and that
100 was actually a slight interpolation.
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

drs@canby.com wrote:
> Is there a way to determine the native ISO of the sensor in my Canon
> 20D? From what I've heard, it isn't necessarily the lowest ISO
> setting. Someone said the sensor in the 10D was best at 200 and that
> 100 was actually a slight interpolation.

Adjusted by changing the gain in the amplifiers that are between the
CCD and the A/D converter that turns the voltage levels from the CCD
into a digital value. In general it is good to run lower gain because
this means lower noise.

I have no idea how interpolation relates to the ISO setting.

Scott
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

drs@canby.com wrote:

> Is there a way to determine the native ISO of the sensor in my Canon
> 20D? From what I've heard, it isn't necessarily the lowest ISO
> setting.

Yes, although it really is the quantum efficiency,
expressed in electrons captured in the well in a given
time for a given light level.

> Someone said the sensor in the 10D was best at 200 and that
> 100 was actually a slight interpolation.

No.
The 10D as well as all other cameras I've seen data for
are photon noise and read noise limited.
The 10D has a full well capacity of about 44,000
electrons and read noise of about 16 electrons.
The full well range is digitized at ISO 100.

See:

The Signal-to-Noise of Digital Camera images and Comparison to Film:
http://clarkvision.com/imagedetail/digital.signal.to.noise

Digital Cameras: Does Pixel Size Matter?
Factors in Choosing a Digital Camera:
http://clarkvision.com/imagedetail/does.pixel.size.matter

Roger
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

In message <7c1v011erejme03cbplmhcvighd9tkp3fp@4ax.com>,
drs@canby.com wrote:

>Is there a way to determine the native ISO of the sensor in my Canon
>20D? From what I've heard, it isn't necessarily the lowest ISO
>setting. Someone said the sensor in the 10D was best at 200 and that
>100 was actually a slight interpolation.

I would call it the "lowest analog" ISO, not native. I don't think that
there is such a thing as native ISO; the lowest ISO is amplified too,
just like the rest. The gain is just different. What is the native
amplification of your home stereo?

I've seen RAW data from a 10D that looked like it was really ISO 66
digitally pushed to 100. The other direction (200 digitally pulled to
100) is not possible, as ISO 100 on the 10D has full dynamic range in
the RAW data.

20D seems to be straight-up at ISO 100, unless Canon covered their
tracks very well. They leave the data in 20D ISO 3200 shots all even,
showing that it is 1600 pushed to 3200, so I don't know if they'd try to
hide anything (perhaps they were ridiculed for the fake odd numbers in
the 10D's 1600 and 3200 modes, which are both pushed a stop).

--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

On Sun, 13 Feb 2005 20:32:02 GMT, JPS@no.komm wrote:

>20D seems to be straight-up at ISO 100, unless Canon covered their
>tracks very well. They leave the data in 20D ISO 3200 shots all even,
>showing that it is 1600 pushed to 3200, so I don't know if they'd try to
>hide anything (perhaps they were ridiculed for the fake odd numbers in
>the 10D's 1600 and 3200 modes, which are both pushed a stop).

I quickly get more information than I understand. And that's fine.

Does it accurate to think of the changes in ISO settings to be linear
as far as potential noise generation is concerned?
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

"Roger N. Clark (change username to rnclark)" <username@qwest.net>
wrote in message news:420F8930.2020506@qwest.net...
> drs@canby.com wrote:
SNIP
>> Someone said the sensor in the 10D was best at 200 and that
>> 100 was actually a slight interpolation.
>
> No.
> The 10D as well as all other cameras I've seen data for
> are photon noise and read noise limited.
> The 10D has a full well capacity of about 44,000
> electrons and read noise of about 16 electrons.
> The full well range is digitized at ISO 100.
>
> See:
>
> The Signal-to-Noise of Digital Camera images and Comparison to Film:
> http://clarkvision.com/imagedetail/digital.signal.to.noise
>
> Digital Cameras: Does Pixel Size Matter?
> Factors in Choosing a Digital Camera:
> http://clarkvision.com/imagedetail/does.pixel.size.matter

Roger, did you by any chance also try the (optional) ISO 50 setting?
I assume it is just a binary shift operation on the linear gamma data,
which would decrease signal to noise.

Bart
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

Bart van der Wolf wrote:

>
> "Roger N. Clark (change username to rnclark)" <username@qwest.net> wrote
> in message news:420F8930.2020506@qwest.net...
>
>> drs@canby.com wrote:
>
> SNIP
>
>>> Someone said the sensor in the 10D was best at 200 and that
>>> 100 was actually a slight interpolation.
>>
>>
>> No.
>> The 10D as well as all other cameras I've seen data for
>> are photon noise and read noise limited.
>> The 10D has a full well capacity of about 44,000
>> electrons and read noise of about 16 electrons.
>> The full well range is digitized at ISO 100.
>>
>> See:
>>
>> The Signal-to-Noise of Digital Camera images and Comparison to Film:
>> http://clarkvision.com/imagedetail/digital.signal.to.noise
>>
>> Digital Cameras: Does Pixel Size Matter?
>> Factors in Choosing a Digital Camera:
>> http://clarkvision.com/imagedetail/does.pixel.size.matter
>
>
> Roger, did you by any chance also try the (optional) ISO 50 setting? I
> assume it is just a binary shift operation on the linear gamma data,
> which would decrease signal to noise.

Bart,
Yes, I did test ISO 50 on the S60, and ISO 50 is when you get
full well, so it does improve the signal-to-noise, and you gain
root 2 because of double the photons for the bright areas,
and the gain in electrons/DN changes by a factor of 2,
so the read noise is lower by 2x relative to full scale,
so shadows are improved as well.

On the above page, I was comparing the same ISO. I
suppose I should add a note about ISO 50 on the S60.
I have not tested it on the 10D or 1DII, and on these cameras,
something additional is going on, as the sensors reach full well
at ISO 100.

Roger
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

In message <0slv01hnjbk30rv8p8vsu8lq4npcifedgt@4ax.com>,
drs@canby.com wrote:

>On Sun, 13 Feb 2005 20:32:02 GMT, JPS@no.komm wrote:
>
>>20D seems to be straight-up at ISO 100, unless Canon covered their
>>tracks very well. They leave the data in 20D ISO 3200 shots all even,
>>showing that it is 1600 pushed to 3200, so I don't know if they'd try to
>>hide anything (perhaps they were ridiculed for the fake odd numbers in
>>the 10D's 1600 and 3200 modes, which are both pushed a stop).
>
>I quickly get more information than I understand. And that's fine.
>
>Does it accurate to think of the changes in ISO settings to be linear
>as far as potential noise generation is concerned?

If the quantization of the data were not an issue, this would be true.
Imagine, however, that if by some magic, all of a sudden the camera had
ISO 50, 25, 12.5, 6.25, 3.125, 1.5625, etc. All of the super-low ISOs
have the same limitation; the bit depth of the image, which breaks up
the shadows, by posterizing them. It's not noise, per se, but it might
as well be. Once you visualize this, then it is easy to see how the
"noise" in the shadows at ISO 200 is not twice as bad as the noise at
ISO 100, even though it is about twice as strong, in the sensor itself.

It's like, do two drops of rum in a glass of coke taste twice as strong
as one drop? No, but 4 ounces will certainly taste stronger than 2
ounces of rum.

--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

On Mon, 14 Feb 2005 01:31:38 GMT, JPS@no.komm wrote:

>If the quantization of the data were not an issue, this would be true.
>Imagine, however, that if by some magic, all of a sudden the camera had
>ISO 50, 25, 12.5, 6.25, 3.125, 1.5625, etc. All of the super-low ISOs
>have the same limitation; the bit depth of the image, which breaks up
>the shadows, by posterizing them. It's not noise, per se, but it might
>as well be. Once you visualize this, then it is easy to see how the
>"noise" in the shadows at ISO 200 is not twice as bad as the noise at
>ISO 100, even though it is about twice as strong, in the sensor itself.
>
>It's like, do two drops of rum in a glass of coke taste twice as strong
>as one drop? No, but 4 ounces will certainly taste stronger than 2
>ounces of rum.

That's a good, clear illustration. Thanks. I see I have a lot of
homework to do to understand my camera. Rum and coke probably won't be
the answer.
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

> "Roger N. Clark (change username to rnclark)" <username@qwest.net>
> wrote in message news:420F8930.2020506@qwest.net...
> > drs@canby.com wrote:
> SNIP
> >> Someone said the sensor in the 10D was best at 200 and that
> >> 100 was actually a slight interpolation.
> >
> > No.
> > The 10D as well as all other cameras I've seen data for
> > are photon noise and read noise limited.
> > The 10D has a full well capacity of about 44,000
> > electrons and read noise of about 16 electrons.
> > The full well range is digitized at ISO 100.
> >
> > See:
> >
> > The Signal-to-Noise of Digital Camera images and Comparison to Film:
> > http://clarkvision.com/imagedetail/digital.signal.to.noise
> >
> > Digital Cameras: Does Pixel Size Matter?
> > Factors in Choosing a Digital Camera:
> > http://clarkvision.com/imagedetail/does.pixel.size.matter

Actually following up to Roger... Very cool references. I learned a
lot. Thanks.
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

Roger N. Clark (change username to rnclark) wrote:

> drs@canby.com wrote:
>
>> Is there a way to determine the native ISO of the sensor in my Canon
>> 20D? From what I've heard, it isn't necessarily the lowest ISO
>> setting.
>
>
> Yes, although it really is the quantum efficiency,
> expressed in electrons captured in the well in a given
> time for a given light level.
>
>> Someone said the sensor in the 10D was best at 200 and that
>> 100 was actually a slight interpolation.
>
>
> No.
> The 10D as well as all other cameras I've seen data for
> are photon noise and read noise limited.
> The 10D has a full well capacity of about 44,000
> electrons and read noise of about 16 electrons.
> The full well range is digitized at ISO 100.
>
> See:
>
> The Signal-to-Noise of Digital Camera images and Comparison to Film:
> http://clarkvision.com/imagedetail/digital.signal.to.noise
>
> Digital Cameras: Does Pixel Size Matter?
> Factors in Choosing a Digital Camera:
> http://clarkvision.com/imagedetail/does.pixel.size.matter
>
> Roger
>
I know there IS an ISO standard (which specifies a measurement
technqiue) on ISO of camera itself. Anyone know if there IS a standard
on the chip itself- a specific ISO standard with number?

When I worked with semiconductor detectors we always used NEP and
detectivity to write specs for chips and individual detectors.

Also, NEP would be a much better value than quantum efficiency, because
the details of the chip well size and readout electronics on the chip
will affect the noise over and above the quantum efficiency itself.
There is also the charge transfer efficiency in a CCD device.
 
G

Guest

Guest
Archived from groups: rec.photo.digital (More info?)

"Roger N. Clark (change username to rnclark)" <username@qwest.net>
wrote in message news:421016C8.4040108@qwest.net...
SNIP
> I have not tested it on the 10D or 1DII, and on these cameras,
> something additional is going on, as the sensors reach full well at
> ISO 100.

Right, that's why it might be interesting to see what ISO 50 on your
1DII does with the S/N. I'm just curious by nature, that's why I
asked.

Bart