Archived posting to the Leica Users Group, 2002/03/09
[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]At 8:51 PM -0500 3/9/02, Rolfe Tessem wrote: >> >>Grain aliasing is also an issue; Certain films will appear to have >>significantly larger grain when scanned due to interference between >>the discreet nature of the silver grains and the scanner sensor. >>Chromogenic film doesn't have sharp edged film grains to cause this >>problem. For scanning, the lack of acutance of the chromogenic >>films is an asset. > > >Henning, > >I think the grain aliasing issue is more about the software than it >is an issue with the film or even with the scanner hardware. > >I use a Nikon LS-4000 and have yet to see this phenomenon, although >I've certainly seen a number of references to it on the web and >elsewhere. Having said that, I certainly haven't exhautively tested >every B&W emulsion to see how it scans (I'll leave that to Erwin); >I've scanned Tri-X, Delta 400 (new) and Efke 25 and have never seen >anything I would call grain aliasing. > >My guess (and it is only that) is that what we are really dealing >with here is immature scanning software. The masses scan primarily >color reversal and negative so that is what the software is >optimized for. As time goes on, I'd bet that B&W negative will be >better supported. > >It is already better for me with Vuescan than Nikon Scan, although >I'm aware that lots of people dislike Vuescan for its workflow >issues. Nonetheless, you can't argue with results and Vuescan >consistently delivers better scans than Nikon Scan for me, >especially when scanning traditional B&W negative material. > >Rolfe Unfortunately, the aliasing is really a hardware problem that the software can't solve (yet). It's part of the digitizing process, and is related to the issue with the CCD sensors in digital cameras and the 'dumbing' filter that is necessary to make sure that the lens can't resolve too much. Film scanners can resolve somewhat too much for some grain structures. 2700 dpi doesn't work too well with Tri-X, and the 4000dpi scanners have more problems with T-Max 400. Rodinal developed film will have more problems than if you have a lot of solvent in your developer. If the scanners had a _lot_ more resolution, to actually scan the grains' shapes,it would be OK, or if they had somewhat less resolution. Unfortunately, you can't just tell the scanner to scan at lower resolution, because the scanner doesn't actually scan at lower resolution, it just throws the information from every second pixel away, while still scanning at the same resolution. You would need a 'dumbing' filter between the scanner and the film to get the true resolution down. - -- * Henning J. Wulff /|\ Wulff Photography & Design /###\ mailto:henningw@archiphoto.com |[ ]| http://www.archiphoto.com - -- To unsubscribe, see http://mejac.palo-alto.ca.us/leica-users/unsub.html