Lets back up a bit here. You have a sensor that has an analog out put. You send that output to an analog to digital converter. The converter produces an 8, or 10, or 12, or in this case 22 bit digital representation of the analog signal. Then that digital representation is processed digitally. Note that the analog signal does not change. The only difference is that the higher bit rates produce a more accurate representation of the analog signal. There is no magic here, just technology.
Rephrasing that, the analog signal is simply broken up into more digital pieces, nothing is added to it. -- graywolf http://www.graywolfphoto.com http://webpages.charter.net/graywolf "Idiot Proof" <==> "Expert Proof" ----------------------------------- Antti-Pekka Virjonen wrote: >> -----Original Message----- >> It's been mentioned about two point six zillion times by now that you >> probably don't, though. The sensor itself doesn't have a lot more than >> 12-bits worth of latitude, so producing more bits in the A/D doesn't >> help a lot. > > I disagree with that a little. It's still the same sensor reveiving the > photons no matter if you set the ISO to 100 or 1600 in a normal > DSLR. When setting the camera to ISO 1600 you just capture a lot > less photons in one pixel well than you would if you set ISO 100 > (having the same amount of light available on both exposures). This > is because you will have less light entering the sensor to get the > correct exposure. Every single photon generates one electon (not > exactly but to about 70% success rate) and there is a lot of > space for those electrons. A good modern CCD sensor does have 14-16 bits > worth of depth (or latitude). Don't know about these DSLR camera > sensors though, maybe they are just 12 bits :(. That would be a little > disappointing, heh. (Ok, I need to do some CCD data sheet reading I > guess). > > If you shoot at 1600, the amplifier gain is set high because > less photon generated electrons are available to be read from the CCD > wells to get the output voltage values for all the pixels in the picture > to closely fill the full dynamic range of the following a/d converter. > The overexposed cells or wells (still far from flowing over) overflow > the amplifier (at the set gain) and the a/d reads full scale value for > those "pixels" (overexposure). > > If you set the same camera to lower ISO (like 100). You still use the > same ccd, just the wells of the sensor get a lot more photons and > a lot more electrons are generated. Then you use a lower gain amplifier > to get the output matched as close as possible to the input range of the > > a/d conversion. > > When using the 22 bits a/d you can have a lower gain amplifier and then > choose the suitable part of the full scale input of the a/d converter to > get the (for example) 12 bits of the wanted exposure and latitude (which > > is a small part of the whole dynamic range of the conversion). > At least, this is how I believe it could work in theory. In real world > the CCD well depth vs. noise level (dynamic range) is of course less > than 22 bits but you still can have an optimized fixed gain amplifier. > It is a lot more easier to design a good fixed gain amp than a variable > one. > >> Not amplifying the signal (in a variable manner) probably does help *a >> bit* as one source of noise has been removed (and the 12 bits are the >> latitude after noise has been chopped off), but it's probably not too >> significant compared to the noise in the actual sensor and/or the >> interface to it. >> >> - Toralf > > Of course, the real world "gain" from all this technology is not > world breaking but I think we may well notice an easily visible > difference between Pentax and the current competition. I don't think > either that we can set the post prosessing ISO value from the whole > 100-1600 range, but I am pretty convinced there will be an advantage > over the current models. > > It would be nice if they would explain exactly why they decided to > have 22 bits of a/d. > > This got a bit too technical... I even did some amplifier design today > and looked at bit too deep into the future *grin*. > > Can't help it, being an electronics design engineer... > > Let's just wait and see, > Antti-Pekka > ________________________________________ > > Antti-Pekka Virjonen > > Computec Oy > R&D Turku > > www.computec.fi > > > -- PDML Pentax-Discuss Mail List PDML@pdml.net http://pdml.net/mailman/listinfo/pdml_pdml.net
