Post by thread:[PIC]: confused about general delay methods!

Hi!

I4m just starting to learn how to program Microchips microcontrollers so i4m very glad that information sources like the piclist exist.

I found a quite interesting page about delay methods at http://www.piclist.com/techref/microchip/delay/general.htm.

I have a question about method nr. 2 in the article.

Consider the code:

According to the article you will get a total delay of three times the initial value plus three.

DELAY   MOVLW   D'95'           ; 288 cycle delay
                MOVWF   COUNTER
                DECFSZ COUNTER,F
                GOTO    $-1

But when I simulate it in MPLAB IDE (using PIC16F872), I just got a 286 cycles of delay! What have went wrong?

Let us give it the initial value of 2, which give us 2*3+3=9 cycles according to the article.

But according to me the "program flow" will be like

MOVLW D'2' ;1 cycle

MOVWF COUNTER ;1 cycle

DECFSZ COUNTER,1 ;1cycle, COUNTER have now the value 1

GOTO $-1 ; 2 cycle

DECFSZ COUNTER,1 ; 2 cycle, because the result will be 0, this will end the delay loop

A total delay of 7 cycles is what I get. I would like the formula to be 3 * intial_value + 1 instead of 3 * initial_value +3.

I welcome any comments that will bring light to this!

Thanks in advance!

MSN Photos is the easiest way to share and print your photos: Click Here
-- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

On Mon, 15 Apr 2002 18:01:00 +0300, "Haris I. Volos" wrote: >Hi, > >Propaply the two additional cycles are for the call/return instructions so >you know if you call the routine you will have the total delay stated in the >routine. > That could be true except for the fact that call and return are both 2-cycle instructions. Used inline, the routine does give a 3*n + 1 cycle delay. Called as a subroutine, the call and return make it 3*n + 5 cycles total. The cycle count for the delay routine itself works out as: 1 cycle for the movlw 1 cycle for the movwf n-1 cycles for the decfsz (while not zero) 2 * n-1 cycles for the goto 2 cycles for the decfsz (when zero) This totals to 3*(n-1) + 4 cycles = 3 * n + 1 cycles. Regards, Bob > >{Original Message removed}

----- Original Message ----- From: "Bob Barr" <.....bbarrKILLspam@spam@CALIFORNIA.COM> To: <PICLISTKILLspamMITVMA.MIT.EDU> Sent: Monday, April 15, 2002 7:22 PM Subject: Re: [PIC]: confused about general delay methods! .... > >Propaply the two additional cycles are for the call/return instructions so > >you know if you call the routine you will have the total delay stated in the > >routine. > > > > That could be true except for the fact that call and return are both > 2-cycle instructions. > .... Thanks for pointing this to me because I am not so experienced with PICS and I wasn't aware that instructions who change the program counter (such return/call/goto) need two cycles. Haris -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservKILLspam.....mitvma.mit.edu with SET PICList DIGEST in the body

On Mon, 15 Apr 2002 22:42:13 +0300, "Haris I. Volos" wrote: <snip> > >Thanks for pointing this to me because I am not so experienced with PICS and >I wasn't aware that instructions who change the program counter (such >return/call/goto) need two cycles. > Don't forget btfss, btfsc, decfsz, and incfsz. They all take 2 cycles if they do the skip, but only 1 when they fall through. And now this discussion's got me wondering about doing operations directly on PCL. I would think that an operation such as a movwf to PCL should take 2 cycles but I'm not certain of this. Hmmm, I hadn't thought about that before. Regards, Bob -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspam_OUTTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

At 02:45 PM 4/15/02 -0700, Bob Barr wrote: >And now this discussion's got me wondering about doing operations >directly on PCL. I would think that an operation such as a movwf to >PCL should take 2 cycles but I'm not certain of this. Hmmm, I hadn't >thought about that before. Yes - any operation where PCL is the destination takes 2 cycles. dwayne Dwayne Reid <dwaynerspam_OUTplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 18 years of Engineering Innovation (1984 - 2002) .-. .-. .-. .-. .-. .-. .-. .-. .-. .- `-' `-' `-' `-' `-' `-' `-' `-' `-' Do NOT send unsolicited commercial email to this email address. This message neither grants consent to receive unsolicited commercial email nor is intended to solicit commercial email. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

On Mon, 15 Apr 2002 17:44:22 -0600, Dwayne Reid wrote: >At 02:45 PM 4/15/02 -0700, Bob Barr wrote: > >>And now this discussion's got me wondering about doing operations >>directly on PCL. I would think that an operation such as a movwf to >>PCL should take 2 cycles but I'm not certain of this. Hmmm, I hadn't >>thought about that before. > >Yes - any operation where PCL is the destination takes 2 cycles. > Thanks, Dwayne. I thought I had read that somewhere in the Microchip docs but I didn't see it mentioned when I was posting. Regards, Bob -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email KILLspamlistservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

> And now this discussion's got me wondering about doing operations > directly on PCL. I would think that an operation such as a movwf to > PCL should take 2 cycles but I'm not certain of this. Hmmm, I hadn't > thought about that before. PCL is not the high byte of PC, just a buffer for it. Writing to PCL does not affect the PC, so it is one cycle only. Wouter van Ooijen -- Van Ooijen Technische Informatica: http://www.voti.nl Jal compiler, Wisp programmer, WLoader bootloader, PICs kopen -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

> PCL is not the high byte of PC, just a buffer for it. Writing to PCL does > not affect the PC, so it is one cycle only. I think you meant to say PCLATH, not PCL. Neither of your statements are true for PCL, but are for PCLATH. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com hint: To leave the PICList spamBeGonepiclist-unsubscribe-requestspamBeGonemitvma.mit.edu

On Tue, 16 Apr 2002 19:48:30 +0200, wouter van ooijen & floortje hanneman wrote: >> And now this discussion's got me wondering about doing operations >> directly on PCL. I would think that an operation such as a movwf to >> PCL should take 2 cycles but I'm not certain of this. Hmmm, I hadn't >> thought about that before. > >PCL is not the high byte of PC, just a buffer for it. Writing to PCL does >not affect the PC, so it is one cycle only. > You're correct that writing to PCL is a 1-cycle operation. At least, according to the MPLAB simulator, it is. As far as it not affecting the PC: Since writing to PCL takes me to a new program location hasn't it, by definition, affected the PC to accomplish this? Regards, Bob -- http://www.piclist.com hint: To leave the PICList TakeThisOuTpiclist-unsubscribe-requestEraseMEspam_OUTmitvma.mit.edu

> You're correct that writing to PCL is a 1-cycle operation. At least, > according to the MPLAB simulator, it is. My mistake: I meant PCLATH. PCL is the lower byte of the PC, and writing to it definetly - takes two cycles - will take you (your program) somewhere else (defined by the value written to PCL and the current value of PCLATH) Wouter van Ooijen -- Van Ooijen Technische Informatica: http://www.voti.nl Jal compiler, Wisp programmer, WLoader bootloader, PICs kopen -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

Best check your definitions, Wouter. Microchip changed things a few years back - PC got changed to PCL. The high byte register is PCLATH. PC and PCL refer to the same thing: the lower byte of the Program Counter located at address 0x02. Any write to PCL (also known as PC) takes 2 clock cycles to execute. If the simulator does not reflect this, the simulator is broken and Microchip needs to be told about it so they can fix it. Bob - the 2 cycle execution time of any instruction where PCL is modified is talked about in the instruction set summary portion of the data sheet. For the 16F87x series, that is page 135 of document DS30292C dwayne At 11:58 AM 4/16/02 -0700, Bob Barr wrote: {Quote hidden}>On Tue, 16 Apr 2002 19:48:30 +0200, wouter van ooijen & floortje >hanneman wrote: > > >> And now this discussion's got me wondering about doing operations > >> directly on PCL. I would think that an operation such as a movwf to > >> PCL should take 2 cycles but I'm not certain of this. Hmmm, I hadn't > >> thought about that before. > > > >PCL is not the high byte of PC, just a buffer for it. Writing to PCL does > >not affect the PC, so it is one cycle only. > > > >You're correct that writing to PCL is a 1-cycle operation. At least, >according to the MPLAB simulator, it is. > > >As far as it not affecting the PC: Since writing to PCL takes me to a >new program location hasn't it, by definition, affected the PC to >accomplish this? > > >Regards, Bob > >-- >http://www.piclist.com hint: To leave the PICList >piclist-unsubscribe-requestEraseME.....mitvma.mit.edu Dwayne Reid <EraseMEdwaynerplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 18 years of Engineering Innovation (1984 - 2002) .-. .-. .-. .-. .-. .-. .-. .-. .-. .- `-' `-' `-' `-' `-' `-' `-' `-' `-' Do NOT send unsolicited commercial email to this email address. This message neither grants consent to receive unsolicited commercial email nor is intended to solicit commercial email. -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestEraseMEEraseMEmitvma.mit.edu

On Tue, 16 Apr 2002 14:03:22 -0600, Dwayne Reid wrote: >Best check your definitions, Wouter. > >Microchip changed things a few years back - PC got changed to PCL. The >high byte register is PCLATH. > >PC and PCL refer to the same thing: the lower byte of the Program Counter >located at address 0x02. Any write to PCL (also known as PC) takes 2 clock >cycles to execute. If the simulator does not reflect this, the simulator >is broken and Microchip needs to be told about it so they can fix it. > Maybe I didn't check it correctly. A 'movlw low($+5)' followed by a 'movwf PCL' results in the jump occurring and the stopwatch window indicating that 1 cycle was used. >Bob - the 2 cycle execution time of any instruction where PCL is modified >is talked about in the instruction set summary portion of the data >sheet. For the 16F87x series, that is page 135 of document DS30292C > Thanks, I thought I had read that somewhere. I had just forgotten where. Regards, Bob -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestspam_OUTKILLspammitvma.mit.edu

Bob Barr wrote: > Maybe I didn't check it correctly. A 'movlw low($+5)' followed by a > 'movwf PCL' results in the jump occurring and the stopwatch window > indicating that 1 cycle was used. movlw 5 movwf PCL Takes 3 cycles on my MPLAB which is correct. -- Best regards Tony mICros http://www.bubblesoftonline.com RemoveMEsalesTakeThisOuTspambubblesoftonline.com -- http://www.piclist.com hint: To leave the PICList EraseMEpiclist-unsubscribe-requestspamspamBeGonemitvma.mit.edu

On Wed, 17 Apr 2002 08:30:50 +1000, Tony Nixon wrote: >Bob Barr wrote: > >> Maybe I didn't check it correctly. A 'movlw low($+5)' followed by a >> 'movwf PCL' results in the jump occurring and the stopwatch window >> indicating that 1 cycle was used. > > movlw 5 > movwf PCL > >Takes 3 cycles on my MPLAB which is correct. > Ooops, I stand corrected. It does on mine too as long as I don't inadvertently have the 17C43 device selected. When simulating with the 16C745 selected, it does show the correct 2 cycles for the 'movwf PCL' instruction. I forgot that I had last been doing 17C43 simulations. Regards, Bob -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestKILLspammitvma.mit.edu