Upon further investigation, the same thing happens at Fosc/16. A
32-instruction loop fails (sends every other byte), a 33-instruction loop
works (but one bit is one instruction time wider than the others).

:-(

This is really very unfortunate. This could be a nice efficient way to
generate
high-speed serial bit streams (eg: manchester), but....

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

----- Original Message -----
From: "Bob Ammerman" <RAMMERMAN@PRODIGY.NET>
To: <PICLIST@MITVMA.MIT.EDU>
Sent: Monday, June 04, 2001 4:20 PM
Subject: [PIC]: SPI hosing


> I had hoped to be able to generate a continuous stream of bits at Fosc/4
> (ie: 8 instruction times per byte) using SPI mode on the MSSP on an
16F87x.
>
> Unfortunately, my testing seems to indicate this is a non-starter.
>
> I have successfully done this using the USART on an 18C, but I need the
> USART for other purposes in the current application.application.
>
> Specifically, on the 18C USART in sync mode (and presumably the 16F87x's
but
> I haven't tested it), if I set SPBRG=0 and BRGH=1 then bytes can be sent
> every 8 instruction times. I do this without testing the UART ready flag,
> because I 'know' it will be set (and who has time to check it when sending
> bytes this close together).
>
> I hoped (expected) to be able to use the same trick with the SPI mode of
the
> MSSP, but found that I had to allow an extra instruction time between each
> byte (ie: 9 bytes per instruction). This causes a one-cycle 'hitch' in the
> bit stream (and the sync clock).
>
> In other words, this 8-instruction loop doesn't work (specifically I get
> every other byte):
>
> forever:
>  movlw  B'10101010'  ;[7]     ;value to output
>  movwf  SSPBUF         ;[0]     ;send it out
>  bsf  PORTB,1            ;[1]     ;sync output for 'scope
>  clrwdt                           ;[2]
>  bcf   PORTB,1             ;[3]      ;end sync
>  nop                                ;[4]
>  goto     forever            ;[5-6]
>
> But this 9-instruction one does (except for the extra 1-cycle hitch):
>
> forever:
>  movlw  B'10101010'  ;[8]     ;value to output
>  movwf  SSPBUF         ;[0]     ;send it out
>  bsf  PORTB,1            ;[1]     ;sync output for 'scope
>  clrwdt                           ;[2]
>  bcf   PORTB,1             ;[3]      ;end sync
>  nop                               ;[4]
>  nop                              ;[6]
>  goto     forever            ;[6-7]
>
> Anybody have any idea how to get around this?
>
>
> Bob Ammerman
> RAm Systems
> (contract development of high performance, high function, low-level
> software)
>
> --
> http://www.piclist.com#nomail Going offline? Don't AutoReply us!
> email listserv@mitvma.mit.edu with SET PICList DIGEST in the body
>
>

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body