Digital BW, The Print

Make sure you set the highlight and shadow values very low in Vuescan,
otherwise the contrast will be hard. When I had the Minolta I would use
0.0001 for both values and 'White balance' for the color mode. My scans
would come out very flat, retaining all detail in shadows and 
highlights. Then I could increase the contrast in Photoshop with the 
curve tools to my liking.

As for grain I haven't noticed that big a difference in how different
scanner models will scan grain. My two previous scanners were the Canon
FS4000US and the Minolta 5400, and now I have a Microtek Artixscan 
120tf. The grain from all of them looks very similar - assuming I adjust 
the contrast in the images from the different scanners to match as 
closely as possible.

Contrast will always make a big impact in how you see the grain, so it's 
important not to compare a contrasty image from one scanner to a flat 
image from another, as the former will always look more grainy than the 
latter.

Edge to edge sharpness is an issue with the Minolta 5400 though. It can 
be difficult to get sharp scans if the film is even a little bit curved. 
Actually this appears to be a problem with most film scanners and I 
would recommed getting one that supports a glass holder to keep the film 
perfectly flat. This will also allow you to play with fluid mounting 
which can be a good way to smooth out the grain structure in many films.

I've found that the latest versions of Neatimage has become very good at 
reducing grain in high resolution film scans. Of course, if you use too 
much noise reduction the image will look artificial and plasticky, but 
if you're conservative you get to keep the grain structure of the film 
while reducing it to less objectionable amounts. Noise reduction values 
of 20-30% seems good for 4000dpi film scans of 400iso BW films.

This can also be a good way to counter grain enhancing effects when 
sharpening an image. If you reduce noise first and then sharpen you end 
up with a much less grainy image than if you just sharpen directly, 
which can easily increase the grain to unacceptable amounts.

--
Daniel Staver
http://daniel.staver.no

> Thanks for the suggestion, but I have already given it a go, and I 
> still see similar results. I'm pretty much convinced it's the light 
> source. As I mentioned, it's possible to correct matters somewhat 
> using curves, but that can have a deleterious effect on image 
> quality. I think the "problem" is with the raw data coming off the 
> sensor, not in the post-processing. I'm not trying to suggest that 
> B&W shooters should avoid this scanner, just that it's worth checking
>  out some scans before putting your money down, as it may not provide
>  you with the results you want.