[EMAIL PROTECTED] wrote:
> I didn't test it, but I suspect a divisor of 1 will work fine.
> However, I'm not convinced that this would be at all useful.
Here's what I have in mind.
Assuming the resolving power is only 4800 dpi, so even if we
offset by 1 step, we won't have two discrete samples.
But, assuming the pixel structure at 9600 dpi looks like this:
abcdefghij
First pass, we read: ab cd ef gh ij
Second pass, we read: a bc de fg hi j
I assume that the CCD will pick up the average, so for pixels a
and b, we actually read a value of a + b.
Since we can control where to start to read, we start off beyond
the edge, where a = 0 (or some fixed known value). Once we know a,
we can know b, and so on.
Will this work? :-)
(void *) &NHY;
* That's all, folks! *
--------------------------------------------------------------------------------------------------------------------
Unsubscribe by mail to [EMAIL PROTECTED], with 'unsubscribe' in the title or
body
