However it is people thinking thoughts like Larry's who provide the equipment for our blissful ignorance =) Ecke
2010/12/21 P. J. Alling <[email protected]>: > Sorry, I try not to think about these things when I'm taking pictures, just > like I tried not to think about silver valance states when shooting film. > > On 12/14/2010 7:51 PM, Larry Colen wrote: >> >> On Dec 14, 2010, at 4:09 PM, steve harley wrote: >> >>> On 2010-12-13 11:53 , Larry Colen wrote: >>>> >>>> In general, we want to boost the dynamic range of our cameras. However, >>>> sometimes we're shooting a very flat scene, or a lens with low contrast. >>>> If >>>> we wanted to, for example, expand six stops of dynamic range into 12, or >>>> 14, >>>> bits of data it seems that one way of doing this *might* be to increase the >>>> ISO, but that would also increase the noise. >>>> >>>> Are there be any electronic ways of doing this that wouldn't change the >>>> sensitivity, or increase the noise? I don't expect that there are ways for >>>> an end user to do this, but could a camera manufacturer allow you to dial >>>> in >>>> the dynamic range of your camera for a particular scene? >>> >>> well any tampering of the analog signal would probably increase noise (or >>> add distortion of some sort), but there may be some sort of preset (or maybe >>> variable) gamma in the analog/digital converter that potentially could be >>> tweaked to do the same thing (ISO is essentially variable gamma, but it >>> shifts the dynamic range window rather than changing its width) >> >> This is what I guess is happening. Please correct my errors. >> >> If we express light in terms of Volts (which is what most a/d sensors >> read), a 12 bit A/D would read 1mV as 0001, and 4,095mV (4V) as full scale. >> At base ISO (100), the analog circuitry will convert the raw output of the >> sensor at these voltages to the values 1 to 4095. >> At ISO 200, I assume that they have a gain of two on the signal off of the >> sensor so that the ADC sees 0.5mV (500uV) as 1mV and 2.048V as 4.096V. >> Likewise up to 6400, where 16uV ->1mV and 64mV->4.096V. >> This is adjusting the gain, without adjusting the offset. >> >> Let's say that we're shooting through a crappy lens, or a dirty window, or >> we're taking a picture of a greycard on a beige sedan on a foggy day. There >> is plenty of light, but there is only 5 stops of dynamic range in the scene >> that we're shooting. I.e. all of the values fall between 2048 and 2080 mV >> at base ISO, or ETTR 4063 to 4095mV which get represented by the values >> 2048-2080 (or 4063-4095). What if, instead, we could change the gain and >> offset so that 4.063V off the sensor read as 0V on the A/D, and 4.095V off >> the sensor read as 4095 at the A/D. >> >> My point is that there is always going to be an analog stage between the >> sensor and the A/D, and the reasons that make it the way to get higher ISO >> rather than shooting at lower ISO and doing the math, would be just as >> relevant to low contrast situations. >> >>> in most practical situations there's probably enough dynamic range in the >>> RAW that you might as well do it in post (just expose to the right and bring >>> down the shadows) >> >> In most practical situations, you could do just fine shooting with a point >> and shoot. It's the impractical situations where you need fancy gear like a >> DSLR shooting with a fast prime. >> >>> -- >>> 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. >> >> -- >> Larry Colen [email protected] sent from i4est >> >> >> >> >> > > > -- > Where's the Kaboom? There was supposed to be an Earth-shattering Kaboom! > > --Marvin the Martian. > > > -- > 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. > -- 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.
