Post by thread:[PIC]: Looking for improved integer's math routine

On Fri, 1 Jun 2001, Alejandro Lavarello wrote: > Hi all! > > Recently, I have used a pair of routines from Microchip's AN617 > and it works well. But, anybody knows better routines for > 16-bit and 32-bit integer multiplication and division? > I'm impressed ! Does really the 16bit division from AN617 works ? Because I can't use it. I've tested two shortest routines, one is Mike Predko's one and the other one you found in piclist archive. Andrew Warren has the shortest one but unfortunately is not good only if you don't care about remainder. Nikolai Golovnicenko have also some short division routines which I don't understand. Maybe you will, so check piclist archive first. Vasile -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

I also searching for better math routines, Can u send me some of speedy math routines. Kashif ali Alejandro Lavarello wrote: {Quote hidden}> Hi all! > > Recently, I have used a pair of routines from Microchip's AN617 > and it works well. But, anybody knows better routines for > 16-bit and 32-bit integer multiplication and division? > > (Improved for code size or for speed, or better, for two !) > > Years ago, I have used the Microchip's binary-toBCD routine. > > Recently, I was discovered the Jhon Payson's binary-to-BCD, > and it is four times more fast. > > Maybe for division and multiplcation another Pic-Lister > have developed improved versions... > > Thanks to all! > Alejandro. > > -- > http://www.piclist.com hint: To leave the PICList > .....piclist-unsubscribe-requestKILLspam@spam@mitvma.mit.edu -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

www.piclist.com/techref/piclist/index.htm in Archive, click on Apply here, On Fri, 1 Jun 2001, Kashif Ali wrote: {Quote hidden}> I also searching for better math routines, Can u send me some of speedy > math routines. > > Kashif ali > > Alejandro Lavarello wrote: > > > Hi all! > > > > Recently, I have used a pair of routines from Microchip's AN617 > > and it works well. But, anybody knows better routines for > > 16-bit and 32-bit integer multiplication and division? > > > > (Improved for code size or for speed, or better, for two !) > > > > Years ago, I have used the Microchip's binary-toBCD routine. > > > > Recently, I was discovered the Jhon Payson's binary-to-BCD, > > and it is four times more fast. > > > > Maybe for division and multiplcation another Pic-Lister > > have developed improved versions... > > > > Thanks to all! > > Alejandro. > > > > -- > > http://www.piclist.com hint: To leave the PICList > > piclist-unsubscribe-requestKILLspammitvma.mit.edu > > -- > http://www.piclist.com hint: PICList Posts must start with ONE topic: > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

At 06:26 PM 6/1/01 -0500, Mike Mansheim wrote: > > Recently, I have used a pair of routines from Microchip's AN617 > > and it works well. But, anybody knows better routines for > > 16-bit and 32-bit integer multiplication and division? > >piclist.com has a bunch of math routines. I've used 16x16 >multiply and 32x16 divide and they work great. I have made much use of a couple of routines that were posted on the piclist many years back and keep using them because they are so easy to customize. I'll post both the multiply and divide routines from my a previous project. They aren't particularly fast or small but notice how easy it is to extend or shrink them! dwayne ;************************************************************************* ; Multiply 24 bits by 16 bits: 40 bit result, exits with MH_h,m,l destroyed ; MH_h,m,l * t_1,2 -> q_5,4,3,2,1 ; 27 ROM, 358 cycles max ; Based on Scott Dattalo's 16x16 multiply. This version by Dwayne Reid. ;************************************************************************* mul24_16 clrf q_1 ;Clear 40-bit result clrf q_2 clrf q_3 clrf q_4 clrf q_5 bsf q_2,7 ;16 loops mul24_loop rrf t_2,F rrf t_1,F ;If the LSB of t_2:1 is set skpc goto mul24_skip movfw MH_low ;Then add MH l:m:h to the result addwf q_3,F movfw MH_mid skpnc incfsz MH_mid,W addwf q_4,F movfw MH_high skpnc incfsz MH_high,W addwf q_5,F mul24_skip rrf q_5,F ;Shift result, pick up carry bit rrf q_4,F ; rrf q_3,F rrf q_2,F rrf q_1,F skpc ;loop counter bit rotate into carry? goto mul24_loop ;no - loop again ;done! ;************************************************************************* ; Divide 40 bits by 24 bits: 40 bit quotient, 24 bit remainder. ; q_5,4,3,2,1 / ML_h,m,l -> q_5,4,3,2,1 remainder in t_3,2,1 ; Uses LOOPER ; 43 ROM, 1365 cycles max ; based on Bob Fehrenbach's 32_32 div routine. This version by Dwayne Reid ;************************************************************************* div40_24 clrf t_1 clrf t_2 clrf t_3 movlw .40 movwf LOOPER loop rlf q_1, f rlf q_2, f rlf q_3, f rlf q_4, f rlf q_5, f rlf t_1, f rlf t_2, f rlf t_3, f movfw ML_high subwf t_3,W skpz goto check_sign movfw ML_mid subwf t_2,W skpz goto check_sign movfw ML_low subwf t_1,W check_sign skpc goto check_count movfw ML_low ;subtract least significant byte subwf t_1,F movfw ML_mid ;2nd byte skpc ;subtract with borrow incfsz ML_mid,W subwf t_2,F movfw ML_high ;3rd byte skpc incfsz ML_high,W subwf t_3,F setc check_count decfsz LOOPER,F goto loop rlf q_1,F rlf q_2,F rlf q_3,F rlf q_4,F rlf q_5,F ;done! Hope someone finds these useful! dwayne Dwayne Reid <.....dwaynerKILLspam.....planet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 17 years of Engineering Innovation (1984 - 2001) * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * 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: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Did you have 64 bit Bin2dec routine. I have but it taks much time to execute. Kashif ali Dwayne Reid wrote: {Quote hidden}> At 06:26 PM 6/1/01 -0500, Mike Mansheim wrote: > > > Recently, I have used a pair of routines from Microchip's AN617 > > > and it works well. But, anybody knows better routines for > > > 16-bit and 32-bit integer multiplication and division? > > > >piclist.com has a bunch of math routines. I've used 16x16 > >multiply and 32x16 divide and they work great. > > I have made much use of a couple of routines that were posted on the > piclist many years back and keep using them because they are so easy to > customize. I'll post both the multiply and divide routines from my a > previous project. They aren't particularly fast or small but notice how > easy it is to extend or shrink them! > > dwayne > > ;************************************************************************* > ; Multiply 24 bits by 16 bits: 40 bit result, exits with MH_h,m,l destroyed > ; MH_h,m,l * t_1,2 -> q_5,4,3,2,1 > ; 27 ROM, 358 cycles max > ; Based on Scott Dattalo's 16x16 multiply. This version by Dwayne Reid. > ;************************************************************************* > mul24_16 > clrf q_1 ;Clear 40-bit result > clrf q_2 > clrf q_3 > clrf q_4 > clrf q_5 > > bsf q_2,7 ;16 loops > > mul24_loop > rrf t_2,F > rrf t_1,F ;If the LSB of t_2:1 is set > skpc > goto mul24_skip > > movfw MH_low ;Then add MH l:m:h to the result > addwf q_3,F > > movfw MH_mid > skpnc > incfsz MH_mid,W > addwf q_4,F > > movfw MH_high > skpnc > incfsz MH_high,W > addwf q_5,F > > mul24_skip > rrf q_5,F ;Shift result, pick up carry bit > rrf q_4,F ; > rrf q_3,F > rrf q_2,F > rrf q_1,F > > skpc ;loop counter bit rotate into carry? > goto mul24_loop ;no - loop again > > ;done! > > ;************************************************************************* > ; Divide 40 bits by 24 bits: 40 bit quotient, 24 bit remainder. > ; q_5,4,3,2,1 / ML_h,m,l -> q_5,4,3,2,1 remainder in t_3,2,1 > ; Uses LOOPER > ; 43 ROM, 1365 cycles max > ; based on Bob Fehrenbach's 32_32 div routine. This version by Dwayne Reid > ;************************************************************************* > > div40_24 > > clrf t_1 > clrf t_2 > clrf t_3 > > movlw .40 > movwf LOOPER > > loop > rlf q_1, f > rlf q_2, f > rlf q_3, f > rlf q_4, f > rlf q_5, f > rlf t_1, f > rlf t_2, f > rlf t_3, f > > movfw ML_high > subwf t_3,W > skpz > goto check_sign > movfw ML_mid > subwf t_2,W > skpz > goto check_sign > movfw ML_low > subwf t_1,W > > check_sign > skpc > goto check_count > > movfw ML_low ;subtract least significant byte > subwf t_1,F > > movfw ML_mid ;2nd byte > skpc ;subtract with borrow > incfsz ML_mid,W > subwf t_2,F > > movfw ML_high ;3rd byte > skpc > incfsz ML_high,W > subwf t_3,F > > setc > > check_count > decfsz LOOPER,F > goto loop > rlf q_1,F > rlf q_2,F > rlf q_3,F > rlf q_4,F > rlf q_5,F > ;done! > > Hope someone finds these useful! > > dwayne > > Dwayne Reid <EraseMEdwaynerspam_OUTTakeThisOuTplanet.eon.net> > Trinity Electronics Systems Ltd Edmonton, AB, CANADA > (780) 489-3199 voice (780) 487-6397 fax > > Celebrating 17 years of Engineering Innovation (1984 - 2001) > > * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * > 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: PICList Posts must start with ONE topic: > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

www.piclist.com/techref/microchip/math James Newton (PICList Admin #3) KILLspamjamesnewtonKILLspampiclist.com 1-619-652-0593 Alejandro Lavarello wrote on 01-5-31 20:58: Hi all! Recently, I have used a pair of routines from Microchip's AN617 and it works well. But, anybody knows better routines for 16-bit and 32-bit integer multiplication and division? (Improved for code size or for speed, or better, for two !) Years ago, I have used the Microchip's binary-toBCD routine. Recently, I was discovered the Jhon Payson's binary-to-BCD, and it is four times more fast. Maybe for division and multiplcation another Pic-Lister have developed improved versions... Thanks to all! Alejandro. -- http://www.piclist.com hint: To leave the PICList RemoveMEpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu -- http://www.piclist.com hint: To leave the PICList spamBeGonepiclist-unsubscribe-requestspamBeGonemitvma.mit.edu

All the math routines from the archive have been collected at http://www.piclist.com/techref/microchip/math If you find something missing, please add it. James Newton (PICList Admin #3) TakeThisOuTjamesnewtonEraseMEspam_OUTpiclist.com 1-619-652-0593 Vasile Surducan wrote on 01-6-1 1:35: http://www.piclist.com/techref/piclist/index.htm in Archive, click on Apply here, On Fri, 1 Jun 2001, Kashif Ali wrote: {Quote hidden}> I also searching for better math routines, Can u send me some of speedy math routines. > > Kashif ali > > Alejandro Lavarello wrote: > > > Hi all! > > > > Recently, I have used a pair of routines from Microchip's AN617 > > and it works well. But, anybody knows better routines for > > 16-bit and 32-bit integer multiplication and division? > > > > (Improved for code size or for speed, or better, for two !) > > > > Years ago, I have used the Microchip's binary-toBCD routine. > > > > Recently, I was discovered the Jhon Payson's binary-to-BCD, > > and it is four times more fast. > > > > Maybe for division and multiplcation another Pic-Lister > > have developed improved versions... > > > > Thanks to all! > > Alejandro. > > > > -- > > http://www.piclist.com hint: To leave the PICList > > RemoveMEpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu > > -- > http://www.piclist.com hint: PICList Posts must start with ONE topic: > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestEraseME.....mitvma.mit.edu

> Post to the site and the list... > On the site, just be sure to select code listing and put the HTML tags > <PRE> > before the code and > </PRE> > after it. OK, here 'tis. I don't know anything about HTML tags - are the <> part of what is typed in? I will post to the site as soon as I hear back. Thanks for the help. /* - based on the 16 bit to 5 digit conversion written by John Payson & explained by Scott Dattalo - this version adds the extra bit so that the 5 digit display can read to 99,999 - want to express the b_i's in terms of the a_i's for the equations: N = a4*16^4 + a3*16^3 + a2*16^2 + a1*16 + a0 N = b4*10^4 + b3*10^3 + b2*10^2 + b1*10 + b0 the solution coded below: b0 = a0 - (4 * (a4 + a3 + a2 + a1)) - 20 b1 = (4 * a4) + (6 * a2) + (2 * a1) - 198 b2 = (5 * a4) + a3 + (2 * a2) - 140 b3 = (5 * a4) + (4 * a3) - 144 b4 = (6 * a4) + 16 - written in CCS C - since ram was plentiful, I used a 3 byte number - only one bit of the extra byte is actually required. I also used three bytes of scratchpad ram to increase the efficiency of the compiled code. The equations can be coded in C exactly as shown above, but that doesn't take advantage of common terms & progressive multiplies. - even with just the one extra bit, the incoming number can exceed 99,999, so it needs to be checked. This code doesn't show that. - CCS C is limited to 16 bit as the largest integer data type. To use a 24 bit number, I use an array to make sure the bytes are consecutive. This function operates on the array as a global, because I think that is more efficient than passing arrays to the function. - the code could be quite a bit smaller - in John's original, the portion of the code that isolated the nibbles and calculated the digits took 32 instructions. If I code that solution in C, using two scratchpad bytes to improve things, it compiles to 64 instructions. This version compiles to 85 instructions, so I'm sure an asm whiz could show quite an improvement. The C version is still much faster than the the original way I did this, and does what I need. */ unsigned int TenK, Thou, Hund, Tens, Ones; unsigned int num[3]; void BIN2DEC(void) { // operates on global variables: // set num[0]-[3] (lo-hi) before calling // returns with TenK...Ones set from 0-9 unsigned int a0, a1, a2, a3, a4; unsigned int t1, t2, t3; // scratchpad variables // isolate the 4 bit nibbles // for the purposes of this routine, a4 can be only 0 or 1 a4 = num[2]; a3 = num[1] / 16; a2 = num[1] & 0b00001111; a1 = num[0] / 16; a0 = num[0] & 0b00001111; // calculate the decimal digits (the b_i's are expressed here // as TenK...Ones, as in the original) // all are negative, except TenK t1 = 4 * a4; Ones = a0 - t1 - 20; t2 = 2 * a1; Tens = t1 + t2 - 198; t3 = 2 * a2; Hund = a3 + t3 - 140; t3 = t3 * 2; // now = 4 * a2 t2 = t2 * 2; // now = 4 * a1 Ones = Ones - t3 - t2; t3 = t3 + a2 + a2; // now = 6 * a2 Tens = Tens + t3; // Tens done t1 = t1 + a4; // now = 5 * a4 Hund = Hund + t1; t3 = 4 * a3; Thou = t1 + t3 - 144; // Thou done Ones = Ones - t3; // Ones done TenK = t1 + a4 + 16; // Tenk done // "normalize" the digits - this asm code was // copied directly from the original #asm movlw 0x0A Lb1: addwf Ones,f decf Tens,f btfss 3,0 goto Lb1 Lb2: addwf Tens,f decf Hund,f btfss 3,0 goto Lb2 Lb3: addwf Hund,f decf Thou,f btfss 3,0 goto Lb3 Lb4: addwf Thou,f decf TenK,f btfss 3,0 goto Lb4 #endasm } -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Yep, the "<" and ">" are necessary. Also, the "&"'s in the code need to be changed to "&" and any "<"'s needs to be "<" One of these days, I'll make that automatic.... ...never mind, some day is today... I've made it automatic. If you just select "program listing" (rather than "question") in the pull down above the text area and paste your code into the text area and press the button, it will do it for you. No <PRE> or </PRE> or & to & or < to < or nothing... just paste it as is and select program listing. Thanks for sharing. I wonder what it would take to extend the code to use more of the bits in the top byte? --- James Newton (PICList Admin #3) RemoveMEjamesnewtonspam_OUTKILLspampiclist.com 1-619-652-0593 PIC/PICList FAQ: http://www.piclist.com or .org {Original Message removed}

Well I'll be damned... (just ask my wife.) I do believe that you are the first person who has ever said that. Or it has been a while. People have donated cash but it is also nice to just hear a "thank you" for all the authors who were willing to share. James Newton (PICList Admin #3) RemoveMEjamesnewtonTakeThisOuTspampiclist.com 1-619-652-0593 Mike Mansheim wrote on 01-6-1 16:43: > Recently, I have used a pair of routines from Microchip's AN617 and it works well. But, anybody knows better routines for > 16-bit and 32-bit integer multiplication and division? piclist.com has a bunch of math routines. I've used 16x16 multiply and 32x16 divide and they work great. Which reminds me, thanks to the writers for posting their code, and thanks to James for making it available. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads -- http://www.piclist.com hint: To leave the PICList EraseMEpiclist-unsubscribe-requestspamspamBeGonemitvma.mit.edu