Re: Film scanner question

Happy Traveler wrote:
I hate to reopen that old can of worms, but it's so hard to resist...



Your throwing '1.5 bits to analog noise' may be a good guess for the quality
of the A/D, though some new designs are doing better these days. In general,
without knowing the specific component and how it's used, one cannot really
tell. Bits are chip, good performance is not, goes the saying... But you are
taking this too far, assuming that the A/D is the limiting component. I
don't know enough about scanner design to tell that. Seems to me that CCD
noise is a major, if not the dominant factor. That's why fancy drum scanners
still use photomultiplier sensors. And there are many other noise sources to
consider, especially if one is looking for high performance.

Regardless, take 1.5 bits as a starting point. Make it 2, make it 3. Regardless of what you throw away it is always on the same end so a 16 bit scanner has those two bits in its favour for otherwise similar machines (5000 v. V).

The point, which flew over your head in your eagerness to top post is that for slide film you need as much dyncamic range as possible to get information out of the shaddow areas (or detail out of the light areas in a negative).



A simple way of reducing noise, useful for slide film, is to average
multiple passes. The LS-5000 can do it directly, and programs like Vuescan
can pull this trick out of scanners that don't provide the feature, with
success depending on the mechanical repeatability of the instrument design.
Ideally, dynamic range will improve by the square root of the number of
passes, so you get a Dmax improvement of 0.3 for four passes, 0.45 for eight
passes, etc. Ultimately other factors (like pattern noise, nonlinearity,
stability over time, etc) take over, so more than 16 is not practical.

Anything beyond 4, maybe 8, is long into the diminishing returns.

Don't top post. It's insufferably rude.




----- Original Message ----- From: "Alan Browne" <alan.browne@xxxxxxxxxxxxxxxxxxxxx>

Typically, throw 1.5 bits to analog noise, so the dmax are:

16 bit 14 bit

log(2^14.5) v. log(2^12.5)

= =

Dmax 4.36 3.76





--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.
.

Relevant Pages

  • Re: Film scanner question
    ... It's insufferably ruder. ... noise is a major, ... Regardless of what you throw away it is always on the same end so a 16 bit scanner has those two bits in its favour for otherwise similar machines. ... success depending on the mechanical repeatability of the instrument design. ...
    (comp.periphs.scanners)
  • Re: Mic transmission line connection
    ... The designer should understand his op-amp's two noise sources - ... > microphones to find some sort of common mean source impedance. ... > than a typical microphone impedance. ... that your proposed theoretically correct design is buying for a lot ...
    (rec.audio.pro)
  • Re: TRNG with 18 Mbit/s using Thermal Noise
    ... And it is a complete design, not just the noise source, or a theory ... in zenner mode as a noise source, just like most of the circuits here. ... Circuit technique wise, I used a small amount of feedback to broaden the range of acceptable transistors while staying clear of instability and I pointed out that because there was capacitor coupling between stages, there would be some level of autocorrelation - which I minimised by making the time constants involved quite large. ... Assuming this is feed into a sound card then: The low frequency cut off will make autocorrelation effects more obvious. ...
    (sci.crypt)
  • Re: What is the jelly bean DAC du jour?
    ... to design something that requires lots of low noise DC levels to be set. ... What is the common jelly-bean 8-bit multi-DAC with a serial bus these days? ...
    (sci.electronics.design)
  • Re: Amplifier design pre-consultation consultation
    ... that design and punch out the number of boards that we would need. ... Last time I had to deal with noise down to about 5Hz and that is no fun at all because of not well defined 1/f noise-knees. ... Given the capacitance of the detectors in question, ... are not generally vacuum compatible. ...
    (sci.electronics.design)