On 26 Aug 2006 at 10:46, dhbailey wrote:

> I've never had a problem with the appearance of PDFs from Finale,
> since I simply print to a PDF application.  All the data that gets
> sent to a printer is sent to the PDF creator application, so any
> ugliness or problems is the fault of the pdf creator, not Finale.
> 
> I just printed to pdf from FIN2007 and it came out just fine.

It depends on what you use to view the PDFs. Finale creates PDFs that 
with Acrobat Reader's default settings have staff lines that are way 
too thick (and uneven). The 6.x version of the Adobe Reader had a 
setting for line smoothing that if turned on made this display 
problem with Finale-generated PDFs go away. But the same setting in 
the 7.x Adobe Reader does not fix the problem.

And I create all my PDFs with PDF995, as you do.

The problem is with the way Finale is describing the staff lines. 
It's done in a way that stores a description in the PDF file that 
doesn't render well. I would think each staff line would be a line of 
a certain width, but if I'm remembering correctly, Finale actually 
paints thicker lines not as a single line with greater thickness but 
as two or more lines placed together to make the thicker line. That 
may not be the exact situation, but if PDFs and Finale are using 
different assumptions about what is represented (and thus how it 
should be scaled at different resolutions), then the results will not 
always be identical.

Remember that thin lines have to be scaled differently at high and 
low display sizes, otherwise, thin lines at 100% will disappear 
entirely at, say, 50% of so. Thus, the scaling factor has to change 
according to an interpretation of how thin lines should appear. 
Different drawing methods could then produce different results. 
Consider what would happen if the lines were actually defined as two 
close lines instead of as a single line with increased thickness. If 
Acrobat Reader tries to render both lines using it's normal variable 
scaling, the result is going to be two lines rendered very close 
together, whereas the correct result would be to simply ignore one of 
the lines entirely.

Again, I don't know that this is what actually happens, but it is an 
example of how the results could be unreliable based on how the data 
are defined.

-- 
David W. Fenton                    http://dfenton.com
David Fenton Associates       http://dfenton.com/DFA/

_______________________________________________
Finale mailing list
[email protected]
http://lists.shsu.edu/mailman/listinfo/finale

Reply via email to