Pixel density - can the playing field be leveled???

Started Jun 6, 2009 | Discussions thread
Malcolm Practice Contributing Member • Posts: 593
Re: [6/6] Myth busted: small pixels bad, 4 legs good - part 6

bushi wrote:

Hi Daniel,

That was BIG job of yours, that series of posts, thanks!

Good, wasn't it. Congrats to Daniel.

BUT, let me look at this from this angle (I am referring to the same size sensors)

  • sensors/software are technologically evolving (compare the 10MP outputs from Panasonic FZ-50 & LX3, both are considered to have excellent optics (well, in their class - they are no match at pixel level - well, here at last we can compare at pixel level :))

I'll take your word for it.

  • for any given generation of sensors, the ones developed with high ISO sensitivity/higher dynamic response in mind are having less physical resolution than peers.

There could be many reasons for this. For instance, the cameras with high ISO capability tend to be those targeted at sporst/PJ kind of use. As well as high ISO, those cameras need high FPS, and too many pixels would slow them down. Moreover, sports/PJ use tends to be targeted at newsprint and web, which aren't the most demanding applications. So, all in all, the best design compromise to get high ISO is a large sensor and to get high FPS, one with not too many pixels. The fact tha a manufacturer puts on a very high ISO setting doesn't mean that it's necessarily much more capable than one with a lower maximum ISO. The 1DsIII only goes to 3200, but is actually pretty much as capable as the D3 all the way up to 25600.

  • there is not really that much difference in resulting output resolution between the "very dense ones" and "reasonably sensitive ones" (examples to check in the DPReview database over last couple of years, e.g. recently 10 and 14 MP sensors, raw shots comparison between Powershot G10 vs. LX3, and yes, even theoretically, this 40% more pixels on the sensor gives about 118% linear resolution over the 10MP one, right?)

This is possibly true for P&S, which are more lens limited. However, tests have shown the 14MP G10 clearly outresolving the G9 and G7.

  • there is a big difference of high ISO performance and dynamic range of the two types (high res vs. high sensitivity/DR), in each "generation" of sensors.

Evidence please, I have not observed this. For instance DxO mark rates the 1DIII (high sensitivity/DR by your definition) 11.7EV DR and 1078 ISO capability, whereas the 1DsIII (high res) gets 12 and 1663.

So theory & resolution aside, if sensor development went the other direction rather than MP race, I would be very happy, because it is:

a) not often that you are shooting resolution charts, and this is by far the only way you can clearly spot the difference in resolution,
b) not that often that you are shooting in perfect light,

c) not that often that you are shooting in strong light, but without high dynamic range in the scene

There is very little (as in none) evidence

and in both b) and c), the less dense sensors tends to perform much better than the high resolution ones.

So, produce the evidence that this is so, because the theory, as I see it, says otherwise.

-- hide signature --


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