Could this be part of the reason why many experience better results scanning dia than negative film? Mounted dias is 1-2mm above the glass, depending on what kind of dia frames used.
Tim Mostly harmless (just plain Norwegian) Never underestimate the power of stupidity in large crowds (Very freely after Arthur C. Clarke, or some other clever guy) > -----Original Message----- > From: Godfrey DiGiorgi [mailto:[EMAIL PROTECTED] > Sent: 3. desember 2005 17:23 > To: PDML > Subject: Re: "Photo scanner" vs real film scanner? > > On Dec 2, 2005, at 9:46 PM, Boris Liberman wrote: > > > Here is a hint... Take a plastic slide frame, polish the inner side > > of the frame so as to allow smooth not scratching motion of the > > film if you want to scan a strip frame by frame... Applying such a > > frame solved all my flatness problems with my Epson 2450. > > I did a lot of tests on the 2450 scanner sharpness. It's a fixed- > focus unit, the optimum position of the scanned negative is critical, > and slightly variable from unit to unit. Pressed directly against the > glass is generally *not* the ideal position. Presuming flat > negatives, I found that mine achieves maximum sharpness with the > negative positioned at 1mm (.040") above the glass, which is very > very close to what 35mm strips in the standard negative carrier > achieves (1.2mm by my measurements). The capture elements' effective > aperture is quite small so it has a good deal of flexibility. > > For Minox negatives, I find that the negative curl is such that I get > best sharpness from laying them emulsion size down on the scanner > glass, allowing the curl to lift them that scant mm above the glass. > > > I think there is one issue here that hasn't been addressed. My > > Epson 2450 is oldish of course. And my main problem is that the > > colors come out differently depending on how much time the scanner > > has been switched on... > > Any scanner should be turned on and allowed to temperature stabilize > for 10-15 minutes, minimum, before use. Just like any monitor should > be turned on and allowed to stabilize (without 'sleeping'!) for 20-30 > minutes before doing any critical calibration or color adjustment work. > > Godfrey > >
