Print card -- inconsistent results -- clarification

[Timothy Miller]

On Jun 19, 2008, at 1:57 PM, Jan Schenkel wrote:


> You should probably open an entry in the Quality
> Center <http://quality.runrev.com> and attach an
> example stack.

I started to do that by making a minimal version of the problem stack.

After reducing the number of cards (of the same bg) from 450 to 2,  
the number of bg fields from 185 to 5, the number of bg btns from 50  
to 3 (some of these buttons and fields contained substantial scripts,  
and at least one had a script of 40,000 characters), removing 30,000  
characters from the bg script and another 30,000 from the stack  
script, resizing the stack and background (much, much smaller -- the  
cards were four screens tall before), the remaining bg objects  
started printing clearly, with the exception of the text in two bg  
buttons. I compared the printed output side by side. It's not my  
imagination.

Meanwhile, I've discovered I cannot reliably replicate all phenomena  
I previously reported. I've found some inexplicable inconsistencies.

Meanwhile, the original large and cluttered stack continues to print  
badly.

I suppose, at this point, something is painfully obvious to everyone  
but me.

What is it?

FWIW -- I have almost a GB of physical RAM. If no other large  
applications are open, RR does not page out to virtual memory.  
Closing all other applications does not improve the fuzzy printing.  
This machine is an old, slow G4. Even with print jobs stopped, "Print  
card" maxes the CPU for several seconds. But "print card" on a small,  
simple stack that prints without problems also maxes the CPU. I have  
not noticed printing problems with any other application.

Thanks again.


Tim Miller