Bob Blakely wrote: >>From my point of view... > > Only so many photons are captured by a sensor element (pixel, if you will) > of a given size and that to a certain efficiency. There is an upper limit. > Further, everything that has a temperature generates noise in proportion to > that temperature. There is a lower limit. > > The upper limit can only be expanded by increasing the element size to > capture more photons per element. Maintaining the effective resolution then > means increasing the overall sensor size (to full frame?) The lower limit > can only be pushed further down by operating the sensor at a lower > temperature. Currently, the K10D shows noise beginning on the side where > most of the hotter the electronics is located. Red pixels light up first, > then green, then blue. Noise temperature can be further reduced by active > cooling. I suspect that this is not likely to happen with digital cameras > any time soon, sensors for astrophotography and other scientific purposes > excepted. Everybody knows this, and ultimately the larger sensors will > prevail. When this happens, lenses with APS size image circles will become > as useless, practically speaking, as 8 tracks. > > Have you noticed that the upper ISO limits for digital sensors and film are > about the same, 1600 and sometimes 3200? Tere is a reason for this and > ultimately it is the physics of noise that produce thes limits. > > Noise power, N = k*T*B*Nf, where: > > k = Boltzmann's constant; > T = Absolute temperature; > B = Noise Bandwidth of the sensor or film; > Nf = Noise figure, a measure of sensor efficiency. > > Bottom line... there are rules and nature enforces them. > > So... where's my effecient full frame sensor? > > Regards, > Bob...
In the Nikon D3. Improvements in fill factor (reducing the wasted space between sensor sites) have significantly increased sensor performance by increasing the effective area of the sensor sites by a fair margin. The current crop oof 10/12MP APS-C sensors are capable of ISO6400 with quality superior to the old ones at 1600-3200, and can match a 5D at 1600-3200. The D3, which is unique in being a low-density sensor with the new sensor tech, is capable of natve ISO 6400 (the cropped bodies achieve it in Boost) and boost up to ISO 25,600. From the posted samples 6400 on the D3 looks as good as 1600 on the similar-density 5D did, with similar amounts of detail. -Adam -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
