steveg48

When you are using the boosted iso, the camera will have less dynamic range. The electronic gain is being increased which shifts the 'instantaneous' dynamic range up. What was previously the upper end of the dynamic range will now be clipped.

The Villages Loc: The Villages, Florida

I have a Nikon D200, but the camera doesn't matter. Its the concept that I was trying to delve into.

There is some general standard out there, and I think that a number of those responding have given informed feedback.

Bear2 Loc: Southeast,, MI

How do you like the D7200?
I am considering up grading from my D7000, which I love.
Not high jacking just a quick question.
Thanks
Duane

burkphoto Loc: High Point, NC

Go to dxomark.com and learn about signal-to-noise ratios, dynamic range, color depth, etc.

H1 and H2 are just indicators that the camera can be SET that high, but that you might not want to use those speeds if you expect high quality. The manual will tell you their equivalent speeds. They've really cranked up the gain on these speeds beyond what most people would consider reasonable!

There may be an L1 and L2 on the other end of the scale. It's not a "true" ISO 80 or 64 or whatever... Highlights are sacrificed to lower the ISO.

In either case, performance is guaranteed to be less than optimal. "We just did this to appease the spies!" as one Nikon rep told me at PMAI in 2010, about the high speeds. "They don't care if it's grainy or noisy — they just want an image."

pecohen Loc: Central Maine

I don't know whether or not this is true (the measurements reported at https://photographylife.com/sony-a7-vs-nikon-d600-iso-performance suggests it is not) but it is an interesting conjecture.

I've long wondered whether sensor technology might eventually allow the light sensitivity of sensors to be adjustable. As your comment suggests, even the availability of just two settings could be very useful for keeping noise under control.

Apaflo Loc: Anchorage, Alaska

That is also true when using higher ISO that is not in the "boosted" range. Any increase in ISO above "base ISO" will necessarily cause a decrease in dynamic range. (ISO's below base ISO do not get higher dynamic range, but result is less "headroom" between middle gray and pure white.)


Increases ISO clips the brighter parts of the data potentially captured by the sensor, and will record them all as pure white. I.e., if the ISO was set to 100 and then changed to 200, any values that would have originally been recorded as less that white in the first fstop will be recorded, at ISO 200, as pure white. They are clipped. That continues to be true for each fstop at the top of the dynamic range as the ISO is increased. It may not be exactly linear, but it will continue right up to the highest ISO setting possible.

bruswen Loc: Eugene OR

I like the D7200 very much, the autofocus is better (more keepers) and the improvements to low light performance are impressive.

Apaflo Loc: Anchorage, Alaska

:thumbup: :thumbup:

big-guy Loc: Peterborough Ontario Canada

If you turn on the extended ISO it merely gives you X stops more to choose from. 100 ISO is still 100 ISO regardless.

zigipha Loc: north nj

\
And all things being equal, a sensor with smaller number of MPs have larger sensor sites.

So its not total BS...but could have used an *.

singleviking Loc: Lake Sebu Eco Park, Philippines

This analogy only holds true if the overall size of the sensor is not taken into play. You seem to think that all sensors have the same pixel diameter and this is simply not true. Compressing more pixels onto a sensor die will only make for more noise, but when the sensor size is made twice as large in overall area and the individual photo sensor cell/pixel gets twice as large, this allows for more light to be captured by each individual photo cell or pixel giving greater dynamic range.

zigipha Loc: north nj

So if you consider the signal to be the analog information that represents the image but not any noise, then I agree. But if you consider signal to be the analog level that is entering the A/D converter, then I disagree...in the latter case, bit shifting is just multiplying by 2



although no loss of bit depth, you have now lowered the accuracy of the lower bits due to internal A/D noise being more prominent wrt the signal level.
Do you have a refernce for this? I did a quick scan and could not find anything (I understand that the Vref can be altered..but what A/D is designed for this to happen?) ty in advance

kymarto Loc: Portland OR and Milan Italy

Camera manufacturers set a certain range as the "normal" range based on what they consider acceptable results. They give you the option of going higher but want you to understand that quality has crossed a line that will show significant IQ compromises.

That being said, it is not that suddenly IQ takes a steep dive when you enter extended range, it gets worse smoothly for the most part.

Apaflo Loc: Anchorage, Alaska

The analog signal, either at the output of the sensor (the input to the analog ISO amplifier) and at the output of the amplifier (the input to the ADC) is still "signal" in both cases.

What comes out of the digital port of the ADC is digital data, which is not "signal". (Technically the "signal" at the digital output is the analog signal that has digital information encoded on it. But we don't even think about that.)

Bit shifting does indeed multiply the digital value by 2. It does not multiply the signal by two (because that would not change the digital value).


That is true, but it is also of very little significance. On the otherhand, the increase in bit depth is very significant in terms of how much "blocking" in the shadows will result from quantization distortion at higher ISO values.


Unfortunately I don't know of anything online that gives a good rundown on it. I first became aware of it when the Nikon D4 was released, but I don't remember exactly where the clues came from. One additional item to be aware of is that they are not using discrete ADC's anymore! They put one ADC per line of sensor wells on the same chip. So while back when the Nikon D2X came out and if opened up the actual part number for an off the shelf ADC could be seen and the spec sheet tracked down, that is not true with anything Nikon makes today.

BobHartung Loc: Bettendorf, IA

Google turned up these:

http://photo.stackexchange.com/questions/4074/what-does-expanded-iso-mean

http://www.cameralabs.com/reviews/Nikon_D610/noise.shtml

HTH