> > Most of the patterns Sinepatterns makes are available off the shelf in 2
> > inch squares and are suitable for testing 35 mm film scanners. But the
> > costs start around US$700.
>
> Really? For some reason I thought that most were 0.5"x0.5". However I went
>
> through the site about 18m ago and once I found out costs I admit I didn't
>
> pay much attention.
>
They make the MTF charts in sizes ranging from 15 x 20 mm up to 45 by 70 mm
(area covered by pattern). The single frequency gratings are mostly 100 x
100 mm until you ask for higher frequencies.
>
> > Unfortunately the slide
> > belongs to
> > work. :-(
>
> Just as a matter of interest, what material is this on?
>
The slide was made on a thick polyester base film. Must be standard -
Sinepatterns seems to use the same base. Here is the basic pattern I
created:
<<CompositeX1024x1792.tiff>> (50kB)
I use losses TIFF compression to avoid any JPEG artifacts. This patch is
1024x1792 pixels. I created a master file from this in Photoshop using
repeated pastes of the basic image. By doing this at 100% and carefully
aligning each layer I was able to create a final 8192x10752 pixel image
which was printed at a resolution of 2 microns per pixel. This image could
also be printed at 300 ppi to obtain a 27 x 36 inch target for testing
cameras.
Just in case any body is still reading and interested, the Sinepatterns site
that talks about this test pattern is at
. The pattern is the sum of 7
different sinusoids, all with different frequencies (1f, 2f, 3f, 4f, 5f, 6f,
and 7f), the same amplitude, and a relative phases of either 0 or 180
degrees. I modified the pattern by including different base frequencies (1x,
2x, 4x, 8x, 16x, 32x, and 64x). This allows me to test systems with spatial
frequencies that range from 0.5 cycles/mm up to 220 cycles/mm. All with one
slide.
