rook2c4 Loc: Philadelphia, PA USA

Why not simply take a number of test shots at different ISO settings and analyze the difference for yourself? There's nothing like hands-on experience to get a feel of what different settings do, and how far you can push the ISO under various lighting situations before the results become unsatisfactory.

The Villages Loc: The Villages, Florida

I agree with your comment, but regardless my feeling was that there had to be some underlying digital concept to the practice of making use of the H1 & H2 concept (having 2 ranges - the base and the boosted. I think the posted response from "Brusmen" is probably on the mark.

speters Loc: Grangeville/Idaho

One reason is simply cost of production. It is much cheaper to produce a sensor with a lower base, that has to be pushed, than to produce one that has the pushed sensitivity as it's base!

martinfisherphoto Loc: Lake Placid Florida

I'm going out on a limb here. Why not try for yourself. Take a few photos of the same scene and compare.

martinfisherphoto Loc: Lake Placid Florida

You beat me to it.. The Simplest answer.........

The Villages Loc: The Villages, Florida

My response to a previous similar comment -

I agree with your comment, but regardless my feeling was that there had to be some underlying digital concept to the practice of making use of the H1 & H2 concept (having 2 ranges - the base and the boosted. I think the posted response from "Brusmen" is probably on the mark.

Peekayoh Loc: UK

There are two types of gain applied to increase the ISO, analogue gain and digital or software gain. Analog gain is best (originally labelled ISO) and is applied at the hardware level, digital gain (originally labelled H1,h2,H3 or Boost) is simple signal amplification. Where analogue gain stops depends on the camera but is likely to be 800 or 1600 with the latest and greatest, maybe. It seems that the difference between what was labelled as analogue gain and what was labelled as "boost" has gone by the marketing wayside and is therefore meaningless.

singleviking Loc: Lake Sebu Eco Park, Philippines

That's total bullshit. The amount of noise or grain being generated has to do with the size of individual pixels and the quality of the noise reduction algorithm used by the camera's DSP chip. The extension of the ISO range in newer cameras to the astronomical ranges stated by manufacturers is all related to improvements in the camera system's DSP noise reduction software and increases in the overall size and quality of the sensors. The larger the bucket, the more light it captures so, in past camera sensors, the typical max pixel diameter and light capture ability was determined by the number of pixels versus the overall size of the sensor area. This was why full frame and larger sensors made better and less noisy photos than those made by 1/4" cellphone cameras or pocket point and shoot cameras.

Apaflo Loc: Anchorage, Alaska

Essentially valid, at least as far as the history and original derivation of the differences.

It probably isn't totally marketing today though. But it isn't the cut and dried difference that it originally was either. It does have a technical basis, but it's less clearly defined in terms of what a user will see.

Digital multiplication, as it was originally seen, is not really "simple signal amplification". That would be a good description of analog amplification. Digital multiplication was simply bit shifting! Shifting all bits to the next higher level gives 2X "amplification". It not the signal, but it is multiplying the digital value.

However, modern "digital multiplication" does something different that is very useful. The Analog to Digital Converter (ADC) compares the analog signal to a fixed DC voltage. Hence an analog input of 1 Volt is compare to the DC comparitor's 1 volt reference, and the output is 0, and there can be 14 bits of precision. With bit shifting each 2X (1 fstop) loses 1 bit of precision (and 1 fstop of dynamic range). But the new ADC's don't do that, and instead a 2X multiplication is accomplished by making the 1 volt reference a 0.5 volt reference. No loss of bit depth!

It used to be that analog amplification was used up to ISO 6400 or 12,800. Today it is used up to somewhere between ISO 600 and 1200, and then digital multiplication is used.

The Villages Loc: The Villages, Florida

I certainly appreciate the time and effort of those responding to my question.

I guess it was somewhat of a technical question, but one that I wondered about as we see the base and boosted ISO parameters go higher and higher.

John Howard Loc: SW Florida and Blue Ridge Mountains of NC.

I do not fully understand the technical part but on Nikons I believe the distinction between say 1600 to 6400, and H1.0 and H2.0 is more than just arbitrary or marketing. I think it has to do with how the increAsed sensitivity is accomplished. In the normal range the sensor is made more sensitive with more power whereas in the H range the sensor is not made more sensitive. The image is enhanced after it is shot by the software in the camera. Thus while there are increases in noise as the ISO number gets higher there is a larger increase in noise when you go to one of the H settings. Sorry I could not be more technical.

Apaflo Loc: Anchorage, Alaska

There is no way to actually increase the sensitivity of the sensor, by changing the applied power or otherwise. The output is a 1.0 volt maximum analog signal. When the amount of light hitting the sensor only produces 0.5 volts, if ISO is increased that stays exactly the same... except the change to a higher ISO cranks up the gain on an amplifier (much as when you turn up the volume on a radio). What comes out of the amplifier is 1.0 volts, and that goes to the Analog to Digitial Converter and produces a pure white value.

oldtigger Loc: Roanoke Virginia-USA

I've seen enough explanations that i buy into your comments but there is a difference between regular and 'H' range which is large enough to prevent the manufacturers from claiming a single smooth range.
Redefining acceptable noise performance to include that of the 'H' range just wouldn't look good
Customers are more willing to accept stepping into degradation if they are told up front that it exists so there is a strong marketing component involved.

pithydoug Loc: Catskill Mountains, NY

People are going off on tangents. It should be the same or else we we have to come up with 1200* or 1200** to denote the difference

lamiaceae Loc: San Luis Obispo County, CA

I don't think the OP question or experiment makes any sense with my Pentaxs either. My Pentax K-5 has a ISO range from 100 to 12,800 and expandable range from 80 to 51,200. Also I don't think it is actually boosted beyond the normal concept of that as to "gain" of light on my sensor. It is not like optical zoom vs. electronic zoom with a point-n-shoot camera. If he were to me does ISO look any different at say 200 in either modes where I usually shoot, I'd say no, I don't see a difference. But I am not even sure what camera he is talking about.