To the following comments, let me say that there is almost no cost reduction associated with makiing a sensor of the same size but lower pixel count; higher frame rates and lower storage needs are the only demonstrated advantage of having fewer, bigger pixels on the same sized sensor. If higher pixel counts in the same format command higher prices, this is evidence of market demand: the greater desirability of the higher resolution for at least some customers.
In fact if anything, designing and tooling up to make two sensors of the same size instead of a single one will increase costs.
Combining this with the fact that the feared IQ disadvantage of having more, smaller pixels on the same size of sensor can be mostly eliminated by binning, downsampling, or even just the dithering effect of printing at the same size and so at higher PPI, and there is very little downside to increasing pixel counts except memory needs and frame rates.
Which is why every sensor maker keeps reducing pixel size in every format, even if the higher pixel counts offer little gain in IQ for many customers.
I have only seen one demonstrated image related advantage of fewer, bigger photosites: higher frame rates within given limits on processing power. The Nikon D1H, D2H, D3 and the Canon 1D models and 5D all fit this model. In the case of the 5D, achieving even its modest 3fps limited it to about 13MP as it shares a processor with the 20D and 30D, and that processor is capable of only about 40 MP per second.
A cheaper and lower resolution 645 size back ...
[a href=\"index.php?act=findpost&pid=208303\"][{POST_SNAPBACK}][/a]
How about a 30MP 645 sensor with 1 frame every 4 seconds, super-clean ISO 800 and exposures of up to 1 hour, specially for us architecture guys? I don't even want an LCD screen! Should be much cheaper, no?
[a href=\"index.php?act=findpost&pid=208304\"][{POST_SNAPBACK}][/a]
