...

>The number of significant defects in the PIC16C65 is an outrage.  I spent a
> great deal
>of time trying to debug some serial-comm code, only to discover that the SCI
> doesn't
>work dependably when the "BRGH" bit is set.  It was an insidious problem, since
> only
>about 10% of my received packets were corrupted.
>
>
>-----------------------------------------
>Frank Latos, Duo Systems Inc.
>flatos@mich.com
>-----------------------------------------
>
I could use the 16c74 in my ap, would you suggest it?

I am not really too concerned with cost. Only about 10 units will be built.
I won't be using serial i/o but I will make heavy use of TMR1.