Yahoo Groups archive

Digital BW, The Print

Index last updated: 2026-04-28 22:56 UTC

Message

Re: [Digital BW] Dimage 5400 vs Nikon Coolscan 4000

2003-10-02 by Anthony Atkielski

Daniel writes:

> Then they're quite similar.. The Minolta is a touch
> sharper than the Canon.

You can gauge the quality of the scanner with the lower resolution by
comparing it with a downsample of a good scan at much higher resolution.
The more the two resemble each other the more "real" the 4000 dpi is for the
scanner with the lower resolution.  The Nikon 8000, for example, seems to
come extremely close to the claimed 4000 dpi, thanks to good optics (this is
based on my microscope examinations of slides vs. scans).  Some other
scanners may not, such that they put 4000 dpi in the file, but each pixel is
of such low quality that you aren't really getting anything close to 4000
dpi.

For the same reason, I notice that desktop wallpapers made from MF images
look a lot sharper than desktop wallpapers made from 35mm images, even
though the actual resolution of the desktop wallpaper (1600x1200, in my
case) is far below the resolution provided by either of the film images.  It
seems to be due to having more original pixels to create each final pixel in
the MF case; the more pixels are used to produce the final pixel, the higher
the quality of that final pixel.  The resolution doesn't really change, but
the accuracy with which each pixel represents the detail that isn't there
increases if there were lots of original pixels (i.e., lots of original
samples) to produce it.  The changes in the pixel color are subtle, but the
image looks sharper.  It's an interesting effect.  I wonder if LF would look
even better.  The 35mm images are 13 times smaller on the desktop, and the
MF images are about 32 times smaller, so I'm surprised that the difference
even shows, given how much both have been reduced.

Of course this applies to scanners, too.  Scanning at 16000 dpi wouldn't
theoretically pick up much of anything more than an 8000 dpi scan on most
films, and yet a 16000 dpi scan would probably look better when downsampled
than a 8000 dpi scan.  The same applies to 5400 downsampled to 4000, or 4000
downsampled to 2700.

> It successfully removes those black specks
> which can sometimes appear with Velvia slides ...

They are bubbles in the emulsion, actually.

> I know. When photoshop decides to eat up all your memory the best thing
> to do is usually to restart the whole application. It doesn't seem to
> know how to free up memory again even if you close large files for
> instance. Haven't had any conflicts with scanner software though...

My Photoshop (5.x) behaves even more strangely.  When I load up a MF scan
(468 MB), modify it, and then try to save it as a JPEG, Photoshop releases
all of its memory, until it gets down to 30 MB or so, then it tells me that
it doesn't have enough memory to save the image.  I can watch it releasing
memory on the task monitor.  Adobe denies that there is a problem.  Adobe
doesn't want to rework the memory management because that would cost money
and wouldn't be very visible, and the real idea is to add endless new bells
and whistles to bloat the software and persuade people to buy upgrades.
Pretty much the standard phenomenon that you see in every software company.
Eventually Adobe's products will be blended into one giant mess that will
look eerily like the mess that is Microsoft Office.

Attachments

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.