I sent a reply on Friday, wonder why you didn't get it? I have experienced a very similar problem. The SSP seems to insist on inserting the delay. At the rate you are performing this operation why don't you just do it via bit-banging? Bob Ammerman RAm Systems ----- Original Message ----- From: "Jennifer L. Gatza" To: Sent: Monday, October 21, 2002 12:22 PM Subject: [PIC]: 18F SSP Latency? > I didn't receive any replies when I posted this Friday, but I forgot my > [PIC] tag, so I thought I might give it another try. If the repost is an > annoyance, just pretend you didn't see it. ;) > > -------------------------------- > > I am trying to acquire a 16-bit word via SSP (18F452). To do so, I start > two successive transmit/receive sequences, as follows: > > movwf sspbuf ;move dummy data to ssp buffer > btfss sspstat,bf ;if buffer not yet full > goto $-2 ;wait until full > movf sspbuf,w ;move received data to wreg > movwf sspbuf ;move dummy data to ssp buffer > btfss sspstat,bf ;if buffer not yet full > goto $-2 ;wait until full > movff sspbuf,tmp_com_h ;store high byte > movwf tmp_com_l ;store low byte > > When I read the SCK line on the scope, it shows a 50us delay between bytes. > I am using a 5 MHz clock, with the SSP clock at Fosc/64 (78.125 kHz, or 12.8 > us/cyc). > > How quickly can the SSP buffer be rewritten/read? Is there any reason for > this delay between bytes? > > Thanks for your help! > > Jen > > -- > http://www.piclist.com hint: The PICList is archived three different > ways. See http://www.piclist.com/#archives for details. > > -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.