Moon Maid's sensor comparison (PART 2)

Started Mar 1, 2012 | Discussions thread
AgentHEX Contributing Member • Posts: 676
Re: D800 read noise climbing more slowly than ISO

Pierre Sottas wrote:

Marius D wrote:

Marianne Oelund wrote:

I haven't included this effect in the posted curves because I need more data before I'll accept it, but it does appear from the samples I have, that the D800 read noise (in ADU) is climbing more slowly than proportionality to ISO. On the surface, one wouldn't expect this, since ISO sensitivities above ISO 1000 (if it follows the D7000 example) should be only by numerical scaling.

I can see two possible explanations: 1) Some gain sharing at high ISO, such that some of the boost is still provided by the analog electronics; 2) A slight amount of NR is being applied to the RAW data.

There is another possible explanation: 3) reducing ADC reference voltage (V.ref) to match the input analog domain as ISO raises. This should appear like a numerical scaling, except for reduced quantization errors. My guess is they are using both analog amplification and V.ref scaling.

My guess as well. On-chip track and hold amplifiers and on-chip voltage reference seem likely given Marianne's (impressive) results.

That said, I remain disappointed that the engineers who design today sensors put so much energy in the optimization of the output while staying in this outdated paradigm of ISO=sensitivity that we inherited from film days. Personally I don't care if the ISO of my camera is set to 100 or 800 or whatever, what is important is the exposure, not the rules that define the gain between the exposure and the final image.


I think the latter stems from the fact that most photographers are not tech/sci people, and like many professions base craft on tradition and experience over experimentation and deduction. You can tell from the cornucopia of misconceptions even on a forum supposedly dedicated to the most knowledgeable category of the bunch. Thus, transforming their primary toolset using a new paradigm would be disruptive and frankly unnecessary.

For example, there's no real reason for aperture to be anything except calculated real distances (eg. sharp from 5-10 ft), and there's no real reason to use anything other than peak focusing display UI's (on EVF's) to preview focus. Integrate those two + touch input together and it's a superior interface to the clunky AF / Dof mechanism currently used. Live histogram + clipping alert should be the standard for metering but like some of the above is still held up by SLR body design mechanics.

The switch if it were ever likely would've been during the digital changeover, but even then there was enough overlap between the two markets that the manufacturers would be taking too much of a risk if they moved first vs their competitors. If any one of them are in a position it's Sony, and you can see for yourself the flack they get in reviews for experimenting with novel UI's (esp the Nex). I think the "hope" now is that the new generation of photographer raised on P&S's and smartphones simply don't care about traditional settings, and that cameras simply become smart enough to figure most things out and only offer manual assistance on a shot when necessary.

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow