>-----Original Message----- >From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of John Lovda >Every technology has its own set of obstacles to >overcome for increased performance. Some of the >previous comments had to do with the noise level >current sensors and the optical precision of the >lenses needed to provide a good image on a small >sensor. If there is a market, the technology will >follow. In 1990, a 486 processor had (I believe) just >under one million transistors and ran at 66mhz. >Twelve years later with the P-4, we are at 20-30 >million transistors at speeds of 3+ ghz on almost the >same size die.
Craming more pixels on a CCD/CMOS (or in your example a processor) is not the real problem. Canon and others are already doing that. The biggest problem when doing this is noise. Decreasing pixel size decreases the number of detected and converted photons you can store. The less photons you can store the lower the dynamic range. For very small well sizes (at 'big' well sizes dark noise is more a problem) this is made even worse as the percentage of photon noise increases which further reduces the dynamic range. And while dark noise and read out noise can be reduced by better technology photon noise can not by the laws of quantum physics. This is one of several reasons why to stay with the 35mm format for DSLRs. It also becomes questionable what the new standard will be. Is Canon going to have lenses for full-frame (35mm) sensors, another line of lenses with a smaller image circle for sensor sizes with a 1.3x factor (which cannot be used for lower mag factors), another line of lenses for a 1.6x factor, etc? Robert * **** ******* *********************************************************** * For list instructions, including unsubscribe, see: * http://www.a1.nl/phomepag/markerink/eos_list.htm ***********************************************************
