RussellInCincinnati: Further validation of some of the concepts used by Eric Fossum's Quantum Image Sensor approach. In the QIS case, you read the sensor so often that perhaps each pixel-well only needs to hold a super small number of electrons (i.e. need only record accurately the perception of a super small number of photons).
Once you start thinking about it, who wouldn't want all our cameras to work by storing a zillion small subexposures and then stacking however many you want as needed for a given rendition--if there were no prohibitive costs.
Sounds like a pretty good approach. Even better, if you think about it, the process of choosing how many sub-exposures to stack from each pixel could be automatically done by looking at the value of each pixel produced during the first read, and determining how many samples from an exposure set for that particular pixel would then be required to produce a final image that matches a pre-set tone curve.
DPreview, what would be interesting is a "long term" review on some of these cameras, as well as repair/service options for them. I own one of these Panasonics, the TS3, and the thing lasted just under two years before it died (Update: after 3 months in storage, it suddenly started working again!). Turns out that a few people I've met had the same or similar model - and ran into the same problem. OTOH, I know at least one person with one of the Pentax models and it's still going strong after several years of service.
CanadianCoolpix: Since it goes up to the year 2000, I would have included the Olympus C-2100 UZ. This camera was way ahead of its time. It had everything, including a 10X optical zoom with stabilization. Electronic viewfinder.
Peiasdf: The color on the car is off vs. bayer picture. I guess phone users don't really care for color accuracy.
Admittedly I was looking more at noise rather than colour resolution. Wouldn't equal size/amount of pixels result in equal noise across R G and B for every exposure?
This might open up another subject entirely, but I wonder why hexagonal image arrays didn't take off? This seems to be the best way to get an equal number of R, G, and B sensors into the array, which should produce the best results.
Streicher: As I understand the TS3 already featured both a compass and an altimeter ( see http://shop.panasonic.com/shop/model/DMC-TS3S ), so I have a hard time seeing exactly what it is that has been added hardware wise. Any ideas?
That's correct. I have the TS3 and it has both of those features. Even the images in this report look exactly like the TS3 (except for the litle "TS4" lettering, of course).