Very good treatise on the EXT. Thank you Darrell. This is wonderful as I was trying to work out a convenient way to do structures in PBP. I guess you could consider putting this post in the announcements as a sticky. It is very useful.
Jerson
|
Very good treatise on the EXT. Thank you Darrell. This is wonderful as I was trying to work out a convenient way to do structures in PBP. I guess you could consider putting this post in the announcements as a sticky. It is very useful.
Jerson
|
Thanks Jerson! Here's some more to chew on.
Arrays within Arrays
Along with BYTEs and WORDs, you can have Arrays as part of the "Packet". It's just up to you to make sure that everything fits properly.
Perhaps you need to send a series of A/D samples along with the previous Packet to a VB program on a PC.
By Mapping a WORD Array inside the BYTE array, the Sample data automatically becomes part of the Packet.
This adds a 10 WORD array to the previous Packet. And just for good measure I'll put another BYTE variable after it for a CheckSum.
Note here that you don't need to specify the array size for the Samples variable. But you do have to leave enough room for it in the array.Code:MyArray VAR BYTE[29] BANK0 ASM Arg0 = _MyArray ; word Arg1 = _MyArray + 2 ; byte Arg2 = _MyArray + 3 ; word Arg3 = _MyArray + 5 ; word Arg4 = _MyArray + 7 ; byte Samples = _MyArray + 8 ; 10 WORD array (20 bytes) ChkSUM = _MyArray + 28 ENDASM Arg0 VAR WORD EXT Arg1 VAR BYTE EXT Arg2 VAR WORD EXT Arg3 VAR WORD EXT Arg4 VAR BYTE EXT Samples VAR WORD EXT ; 10 WORD array ChkSUM VAR BYTE EXT
And if you get an "Unable to fit variable ____ in requested bank x" warning, change the BANK0 modifier to BANK1 or another bank that has room.
The entire array must be in a single bank for this technique to work.
Now you can just take the A/D samples and store them in the WORD array...(which is really in the MyArray).
Calculate a CheckSum for the whole Packet...Code:FOR X = 0 to 9 ADCIN 0, Samples(X) NEXT XAnd with very little program space used, you have a Complete Packet with 5 WORD or BYTE variables, 10 word sized A/D samples and a checksum for data integrity, ready to send with ...Code:ChkSUM = 0 FOR X = 0 to 27 ChkSUM = ChkSUM ^ MyArray(X) NEXT XAnd, once again, it's bi-directional. You can receive the Packet of Samples just as easily, with ...Code:HSEROUT ["Header", STR MyArray\29]Code:HSERIN [wait("Header"), STR MyArray\29]
DT
|
Just what the doctor ordered!
Thanks for your great work on this. I was looking at doing just this type of process.
I plan to capture a series of equally timed ADCs. This data would eventually be sent to my computer. It will have a time stamp header to indicate the time of the first capture, followed by the series of ADC values, followed by a parity word. It seems you have given me just the tool I need for the task at hand. I can't wait to try it out.
Have you been shuffling through that pile of scribbled notes on my desk?
Thanks again for your contribution. I'm quite new to PICs and this forum, but it is clear from the reading I've done so far that you stand out as one of the top shelf folks here. I've learned a lot.
Steve
BTW, what about instant interrupts for the 16-bit flavors? My project requires quite a bit of RAM (more than any 14-bit PIC) and I really like what I saw reading through those threads. I even thought about trying to shoehorn them into my 18F4620, but balked. I am still quite the newbie. I know it would teach me a lot to work it out on my own, and I would enjoy the process, but I can't spare the time right now.
|
PIC18F2431 feature?
Just found some erratic results on a 18F2431, the following work
while the following work weird, and depending the code overhead, results are different...Code:' ' Pic Configuration ' ================= ASM __CONFIG _CONFIG1H, _OSC_XT_1H & _FCMEN_OFF_1H & _IESO_OFF_1H __CONFIG _CONFIG2L, _PWRTEN_ON_2L & _BOREN_ON_2L & _BORV_45_2L __CONFIG _CONFIG2H, _WDTEN_OFF_2H & _WINEN_OFF_2H ;& WDPS_32768_2 __CONFIG _CONFIG3L, _T1OSCMX_OFF_3L & _HPOL_LOW_3L & _LPOL_HIGH_3L & _PWMPIN_OFF_3L __CONFIG _CONFIG3H, _MCLRE_ON_3H __CONFIG _CONFIG4L, _LVP_OFF_4L & _DEBUG_OFF_4L ENDASM DEFINE OSC 4 ' ' Hardware configuration ' ====================== ' ' I/Os ' ---- PORTA = 0 LATA = 0 PORTB=0 LATB = 0 PORTC =0 LATC = 0 TRISA = 0 TRISB = 0 TRISC = 0 ' ' ADCs ' ---- ANSEL0 = 0 ' ' Timers & pouet pouet ' -------------------- T0CON = 0 T1CON = 0 T5CON = 0 CAP1CON=0 CAP2CON = 0 CAP3CON = 0 DFLTCON = %00111010 QEICON = 0 ' ' MSSP ' ---- SSPSTAT = 0 SSPCON VAR BYTE EXT SSPCON = 0 ' ' CCP ' --- PTCON0=0 PTCON1= %10000000 PWMCON0 = %01001111 PWMCON1 = %00000001 PTPERL=$FF PTPERH=$3F DTCON = 0 OVDCOND= 255 OVDCONS=255 FLTCONFIG = %00000000 SEVTCMPL = 0 SEVTCMPH = 0 ' ' Software variables ' ================== DUTY VAR WORD ' ' Software constants ' ================== Duty0 VAR WORD Duty0 = 16383 '------------------------------< Main program >----------------------------------- ' Start: ' ' Those bellow work 100% ' ---------------------- FOR DUTY = 0 TO Duty0 PDC0L=DUTY.LowByte PDC0H=DUTY.HighByte PAUSEUS 50 NEXT FOR DUTY = 0 TO Duty0 PDC1L=DUTY.LowByte PDC1H=DUTY.HighByte PAUSEUS 50 NEXT FOR DUTY = 0 TO Duty0 PDC2L=DUTY.LowByte PDC2H=DUTY.HighByte PAUSEUS 50 NEXT GOTO Start
When i say weird, PDC2 PWM duty cycle seems to be affected.. work or not, blink and so on.Code:' ' Pic Configuration ' ================= ASM __CONFIG _CONFIG1H, _OSC_XT_1H & _FCMEN_OFF_1H & _IESO_OFF_1H __CONFIG _CONFIG2L, _PWRTEN_ON_2L & _BOREN_ON_2L & _BORV_45_2L __CONFIG _CONFIG2H, _WDTEN_OFF_2H & _WINEN_OFF_2H ;& WDPS_32768_2 __CONFIG _CONFIG3L, _T1OSCMX_OFF_3L & _HPOL_LOW_3L & _LPOL_HIGH_3L & _PWMPIN_OFF_3L __CONFIG _CONFIG3H, _MCLRE_ON_3H __CONFIG _CONFIG4L, _LVP_OFF_4L & _DEBUG_OFF_4L ENDASM DEFINE OSC 4 ' ' Hardware configuration ' ====================== ' ' I/Os ' ---- PORTA = 0 LATA = 0 PORTB=0 LATB = 0 PORTC =0 LATC = 0 TRISA = 0 TRISB = 0 TRISC = 0 ' ' ADCs ' ---- ANSEL0 = 0 ' ' Timers & pouet pouet ' -------------------- T0CON = 0 T1CON = 0 T5CON = 0 CAP1CON=0 CAP2CON = 0 CAP3CON = 0 DFLTCON = %00111010 QEICON = 0 ' ' MSSP ' ---- SSPSTAT = 0 SSPCON VAR BYTE EXT SSPCON = 0 ' ' CCP ' --- PTCON0=0 PTCON1= %10000000 PWMCON0 = %01001111 PWMCON1 = %00000001 PTPERL=$FF PTPERH=$3F DTCON = 0 OVDCOND= 255 OVDCONS=255 FLTCONFIG = %00000000 SEVTCMPL = 0 SEVTCMPH = 0 ' ' Software variables ' ================== ASM PDC0REG=PDC0L PDC1REG=PDC1L PDC2REG=PDC2L ENDASM PDC0REG VAR WORD EXT PDC1REG VAR WORD EXT PDC2REG VAR WORD EXT DUTY VAR WORD ' ' Software constants ' ================== Duty0 VAR WORD Duty0 = 16383 '------------------------------< Main program >----------------------------------- ' Start: FOR DUTY = 0 TO Duty0 PDC0REG=DUTY PAUSEUS 50 NEXT FOR DUTY = 0 TO Duty0 PDC1REG=DUTY PAUSEUS 50 NEXT FOR DUTY = 0 TO Duty0 PDC2REG=DUTY ' wierd stuff :o( PAUSEUS 50 NEXT GOTO Start
Any idea?
Steve
It's not a bug, it's a random feature.
There's no problem, only learning opportunities.
|
a LITTLE results report
as this happen on PDC2, i use the following to produce the above report==============================================
DUTY =402
DUMMY_H =01 DUMMY_L = 91
PDC2_H =00 PDC2_L = 92
==============================================
DUTY =403
DUMMY_H =01 DUMMY_L = 92
PDC2_H =00 PDC2_L = 93
==============================================
DUTY =404
DUMMY_H =01 DUMMY_L = 93
PDC2_H =00 PDC2_L = 94
==============================================
DUTY =405
DUMMY_H =01 DUMMY_L = 94
PDC2_H =00 PDC2_L = 95
==============================================
DUTY =406
DUMMY_H =01 DUMMY_L = 95
PDC2_H =00 PDC2_L = 96
==============================================
DUTY =407
DUMMY_H =01 DUMMY_L = 96
PDC2_H =00 PDC2_L = 97
==============================================
DUTY =408
DUMMY_H =01 DUMMY_L = 97
PDC2_H =00 PDC2_L = 98
==============================================
DUTY =409
DUMMY_H =01 DUMMY_L = 98
PDC2_H =00 PDC2_L = 99
==============================================
DUTY =410
DUMMY_H =01 DUMMY_L = 99
PDC2_H =00 PDC2_L = 9A
==============================================
DUTY =411
DUMMY_H =01 DUMMY_L = 9A
PDC2_H =00 PDC2_L = 9B
==============================================
DUTY =412
DUMMY_H =01 DUMMY_L = 9B
PDC2_H =00 PDC2_L = 9C
==============================================
seems to don't pass the Highbyte eh!Code:Start: FOR DUTY = 0 TO duty0 PDC0reg=duty pauseus 50 NEXT FOR DUTY = 0 TO duty0 pdc1reg=duty pauseus 50 NEXT FOR DUTY = 0 TO duty0 pdc2reg=duty ' wierd stuff :o( pauseus 50 DUMMY=PDC2REG HSEROUT ["DUTY =", DEC DUTY,13,10,_ "DUMMY_H =", HEX2 DUMMY.HIGHBYTE," DUMMY_L = ",HEX2 DUMMY.LOWBYTE, 13,10,_ "PDC2_H =", HEX2 PDC2H ," PDC2_L = ",HEX2 PDC2L,13,10,_ "==============================================",13,10] NEXT GoTo Start
Steve
It's not a bug, it's a random feature.
There's no problem, only learning opportunities.
|
Yes, weirdness indeed.
But the weirdness is why Microchip can't seem to keep things consistant between their products.
In the case of PDCxH:L, the HighByte comes first.
But in most other registers like TMRxL:H, the LowByte comes first.
So unfortunately, PDCxH:L, CAPxBUF and probably others won't work with the EXT type variables.
HTH,
DT
|
aaaaahhhh.. cr... and the RTFM datasheet section 5 of the day goes to me...
sorry
Steve
It's not a bug, it's a random feature.
There's no problem, only learning opportunities.
|
No, that's good info to have in this thread.
Until now, I didn't know it either.
Gotta ask questions.
DT
Members who have read this thread : 2You do not have permission to view the list of names.
Posting Permissions
Bookmarks