Shop Learn
P2 PASM SIN 16Bit — Parallax Forums

P2 PASM SIN 16Bit

pic18f2550pic18f2550 Posts: 169
edited 2021-04-15 11:55 in Propeller 2

Hello,
is there somewhere a code snippet how I can generate a sinusoidal signal with a 16 bit (smart pin) resolution?
Frequency range 0.001 to 20000.000 Hz.
Thanks

Edit

Here is my version of 30 x16Bit sine wave generators.
Have fun trying it out.

Untested. :*

«1345

Comments

  • Wuerfel_21Wuerfel_21 Posts: 1,673
    edited 2021-03-02 15:48

    Here's something I quickly threw together (and haven't actually tested...) that explains the concept:

                  fltl dacpins
                  wrpin ##P_DAC_75R_2V|P_OE|P_DAC_DITHER_RND,dacpins ' dac mode
                  wxpin ##1000,dacpins ' dac period
                  wypin dac_center,dacpins
                  outh dacpins
    .loop
                  testp dacpins wc
            if_nc jmp #.loop
                  qrotate dac_center,phase
                  add phase,frequency
                  getqy tmp
                  add tmp,dac_center
                  wypin tmp,dacpins
                  jmp #.loop
    
    dac_center    long $FF00 / 2
    phase         long 0
    frequency     long 1<<21
    dacpins       long 6 addpins 1 ' pins to output to
    tmp           res 1
    

    For more details, read the documentation

  • pik33pik33 Posts: 1,016
    edited 2021-03-02 20:39

    There is also "headphones.spin2" in Propeller Tools examples. It generates 2 sine waves to 2 pins:

    pinbase         =       8
    
    left            =       pinbase+6
    right           =       pinbase+7
    
    freql           =       432.0           'left frequency
    freqr           =       440.0           'right frequency
    
    volume          =       $1000           '$0000..$7FFF volume
    
    _clkfreq        =       256_000_000
    
    
    dat     org
    
            asmclk
    
            wrpin   dac,#left
            wxpin   #256,#left
    
            wrpin   dac,#right
            wxpin   #256,#right
    
            dirh    #left addpins 1
    
            setse1  #%001<<6 + left
    
    .loop   add     p1,f1           'calculate right sample
            qrotate amp,p1
            getqx   x
            bitnot  x,#15
            wypin   x,#left
    
            add     p2,f2           'calculate left sample
            qrotate amp,p2
            getqx   x
            bitnot  x,#15
            wypin   x,#right
    
            waitse1                 'wait for new period
    
            jmp     #.loop          'loop
    
    
    
    x       long    0
    
    p1      long    0
    p2      long    0
    
    f1      long    round(freql * 65536.0 * 65536.0 * 256.0 / float(clkfreq_))
    f2      long    round(freqr * 65536.0 * 65536.0 * 256.0 / float(clkfreq_))
    
    amp     long    volume
    
    'dac    long    %10111_00000000_01_00010_0      'random dither (noisier, needs no period)
    dac     long    %10111_00000000_01_00011_0      'pwm (quieter, needs 256N period)
    
  • pic18f2550pic18f2550 Posts: 169
    edited 2021-03-05 12:40

    I wonder if I have understood this correctly.

    2020_03_04

    • The control of the signal output was incomplete.
    • Loop adjusted for the sampelrate.
      2020_03_05

    • Integrated the frequency conversion of the channels.

    • Loop for sample rate passed to Smartpin and set to 192khz
  • Note that you don't need (and indeed don't really want) to use WAITCTx to time DAC sampling - Just set the DAC period (Pin's X register) to the desired value and IN will be raised when it is ready for another sample. If you're using multiple DACs, just use the IN from the first one.

  • cgraceycgracey Posts: 13,580

    @Wuerfel_21 said:
    Note that you don't need (and indeed don't really want) to use WAITCTx to time DAC sampling - Just set the DAC period (Pin's X register) to the desired value and IN will be raised when it is ready for another sample. If you're using multiple DACs, just use the IN from the first one.

    Yes, if you start them all up at the same time using ADDPINS, they will forever be synchronized.

  • @cgracey said:

    @Wuerfel_21 said:
    Note that you don't need (and indeed don't really want) to use WAITCTx to time DAC sampling - Just set the DAC period (Pin's X register) to the desired value and IN will be raised when it is ready for another sample. If you're using multiple DACs, just use the IN from the first one.

    Yes, if you start them all up at the same time using ADDPINS, they will forever be synchronized.

    Even if they're not started in sync, if you only use the IN from the first one it will also be synchronized, I think.

  • So I could put wxpin with 192,000.
    OK again learned something about it.
    Thanks

  • cgraceycgracey Posts: 13,580
    edited 2021-03-05 14:56

    @Wuerfel_21 said:

    @cgracey said:

    @Wuerfel_21 said:
    Note that you don't need (and indeed don't really want) to use WAITCTx to time DAC sampling - Just set the DAC period (Pin's X register) to the desired value and IN will be raised when it is ready for another sample. If you're using multiple DACs, just use the IN from the first one.

    Yes, if you start them all up at the same time using ADDPINS, they will forever be synchronized.

    Even if they're not started in sync, if you only use the IN from the first one it will also be synchronized, I think.

    They will have different phases. The danger is that when you go to update them and you have some timing Jitter, you're getting inconsistent framing, updating some on the intended frame and others on the next frame. If you start them all on the same clock, you have that whole frame. In which to update them all.

  • ---------------------------------------------------------------------
     Syncronisation
    ---------------------------------------------------------------------
            waitse1                  "wait for new period
    
    ---------------------------------------------------------------------
     Ausgabe
    ---------------------------------------------------------------------
            wypin   x11,#pin1        
            wypin   x12,#pin2        
            wypin   x13,#pin3        
            wypin   x14,#pin4        
    
    ---------------------------------------------------------------------
     ***
    ---------------------------------------------------------------------
    jmp     #.loop        
    

    I don't think the jitters really matter.
    The phase offset actuates just 15.625ns per channel.

  • cgraceycgracey Posts: 13,580
    edited 2021-03-05 15:22

    @pic18f2550 said:

    ---------------------------------------------------------------------
     Syncronisation
    ---------------------------------------------------------------------
            waitse1                  "wait for new period
                                     
    ---------------------------------------------------------------------
     Ausgabe
    ---------------------------------------------------------------------
            wypin   x11,#pin1        
            wypin   x12,#pin2        
            wypin   x13,#pin3        
            wypin   x14,#pin4        
                                    
    ---------------------------------------------------------------------
     ***
    ---------------------------------------------------------------------
    jmp     #.loop        
    

    I don't think the jitters really matter.
    The phase offset actuates just 15.625ns per channel.

    Yes, in PASM, it is less likely to matter. When it comes to setting unique values, you have no choice.

  • What is the maximum current carrying capacity of the DAC?

  • evanhevanh Posts: 10,918

    They are resistive, so all currents are a distortion. Highest current would be GIO output at 75 ohm shorted to VIO: 3.3 / 75 = 44 mA. Which is 145 mW heating in a tiny spot.

  • I expected a little bit.
    At 16 bits, the "wxpin # 256, # pin1" should always be the multiple of 256.
    With a _clkfreq of 256Mhz I do not get the audio traffic light rate of 192k or 320k.
    The values always have a remainder less than 1.
    I will have to use "waitct1" again.
    The error thus remains below 2 μs.

    dat     org
            asmclk
    
    '        wrpin   dac,  #pin1'
    '                   3            2           1            0'
    '                   0            0           0            0'
    '                  AAAA_BBBB_FFF_PPP_PP_PPPPPPPP_TT_MMMMM_0'
    '                                101_VV_DDDDDDDD                          DAC_MODE (%TT = 00 and %MMMMM = 00000) 8-bit flash, registered'
            wrpin   ##%0000_0000_000_101_00_00000000_00_00000_0,#pin1        'DAC on'
    
    '        wxpin   #16,  #pin1                  16 clocks per sample'
            wxpin   #256, #pin1                ' 256 clocks per sample'
    '        wxpin   #4096,#pin1                4096 clocks per sample'
                                               '00011   = DAC 16-bit dither, PWM'
    
    
            dirh    #pin1 addpins 1            'enable counter'
    
    '        setse1  #%001<<6 + pin1             se1 triggers on adc sample'
             getct     t1
    .loop
            add     p1,f1            ' 4 calculate sample 1'
            qrotate amp11,p1         ' 4'
            getqx   x11              ' 4'
    
    '        waitse1                    4 wait for new period'
            addct1    t1, s
            waitct1
    
            wypin   x11,#pin1        ' 4 Wert ausgeben'
    
            jmp     #.loop           ' ? loop'
    
  • How do I switch from DAC to ADC?
    I would like to measure the load on the DAC in order to avoid an overload on the DAC.

  • evanhevanh Posts: 10,918
    edited 2021-03-11 12:27

    @pic18f2550 said:
    ' wrpin dac, #pin1'
    ' 3 2 1 0'
    ' 0 0 0 0'
    ' AAAA_BBBB_FFF_PPP_PP_PPPPPPPP_TT_MMMMM_0'
    ' 101_VV_DDDDDDDD DAC_MODE (%TT = 00 and %MMMMM = 00000) 8-bit flash, registered'
    wrpin ##%0000_0000_000_101_00_00000000_00_00000_0,#pin1 'DAC on'

    That enables the DAC with a set value of zero. In that pin mode you have to repeatedly issue WRPIN to set new DAC values. Also, you could use the built in constant for the mode setting, eg: P_DAC_990R_3V

    ' wxpin #16, #pin1 16 clocks per sample'
    wxpin #256, #pin1 ' 256 clocks per sample'
    ' wxpin #4096,#pin1 4096 clocks per sample'
    '00011 = DAC 16-bit dither, PWM'

        dirh    #pin1 addpins 1            'enable counter'
    

    Those are for smartpin control but the smartpin is off. You'd have to set a smartpin mode first, ie: M > 0

    ' setse1 #%001<<6 + pin1 se1 triggers on adc sample'

    This suggests you are desiring the smartpin setup for ADC samples. M = %11000 or P_ADC. In this case the DAC and ADC modes can be combined because only the ADC needs the smartpin. eg: wrpin ##P_DAC_990R_3V | P_ADC, #pin1

    Note: To enable the bitstream from the ADC to the smartpin you need to set OUT high.

  • evanhevanh Posts: 10,918
    edited 2021-03-11 12:50

    PS: Here's a block diagram of the pin pairs. It shows there is a number of sections involved in using each I/O pin. In your case above you can set the DAC level from the cog through the upper DAC bus select and DAC network sections of the pin diagram ... and at the same time have the smartpin continuously consuming the ADC's bitstream in the lower sections of the pin diagram.

    1390 x 1593 - 268K
  • evanhevanh Posts: 10,918
    edited 2021-03-11 13:02

    Oh, the mode you're using right now is actually slightly different again, It uses WRPIN instruction for setting DAC level. The block diagram doesn't show the data paths for WRPIN. Instead, the diagram just indicates where each part of the config word is applied block by block.

    To use the cog/streamer DAC bus select there is one config bit to set and then you can use SETDACS instruction for updating the DAC level instead of WRPIN.

        wrpin ##P_DAC_990R_3V | P_ADC | P_TT_01 | your_cogid, #pin1
    

    It's sorta six of one, half a dozen of the other though.

  • I'll sum up everything I imagine.

    The DAC receives a 16bit value from the COG and must output it for an unknown period of time.

    The output value of the DAC is read back from the COG via the ADC.

    Different COGs will write values to a smartpin "simultaneously" (DAC) and read (ADC).

    Is there a document to the PASM commands of the P2?
    For spin2 I found something.

  • evanhevanh Posts: 10,918
    edited 2021-03-11 14:50

    A smartpin can only perform one mode at a time. In this case the smartpin is handling the ADC capture.

    The DAC is set by the cog directly, bypassing the smartpin. And since the smartpin isn't handling the DAC you also don't get the 16-bit dithered mode, you only get the native 8 bits of the DAC.

  • evanhevanh Posts: 10,918

    You could use two smartpins instead. Have the local smartpin dither the DAC in a smartDAC mode, then have a nearby smartpin routed to this pin's ADC using the %AAAA bits in the "Logic Input" block.

  • evanhevanh Posts: 10,918

    As for simultaneously, it might be a tad fraught. Two cogs accessing the same pin registers on exactly the same clock cycle could be corrupting. Certainly two coinciding WRPINs to the same pin will be a mess. There is a possibility that a RDPIN can reliably coincide with a WXPIN or WYPIN though. I just don't know the details there. I don't remember Chip ever clarifying it.

    SETDACS in one cog and RDPIN in another is perfectly okay.

  • evanhevanh Posts: 10,918
    edited 2021-03-11 15:16

    Okay, there is something in the hardware manual but it's less precise than most of Chip's writings. No one's queried Chip on it before it seems.

    Each cog has a 34-bit bus to each smart pin for write data and acknowledgment signalling. Each smart pin OR's all incoming 34-bit buses from the cogs in the same way DIR and OUT bits are OR'd before going to the pins. Therefore, if you intend to have multiple cogs execute WRPIN / WXPIN / WYPIN / RDPIN / AKPIN instructions on the same smart pin, you must be sure that they do so at different times, in order to avoid clobbering each other's bus data. Any number of cogs can read a smart pin simultaneously, without bus conflict, though, by using RQPIN ('read quiet'), since it does not utilize the 34-bit cog-to-smart-pin bus for acknowledgement signalling, like RDPIN does.

    Each smart pin has an outgoing 33-bit bus which conveys its Z result and a special flag. RDPIN and RQPIN are used to multiplex and read these buses, so that a pin's Z result is read into D and its special flag can be read into C. C will be either a mode-related flag or the MSB of the Z result.

    The first paragraph indicates RDPIN will mess with any coinciding writes ... but the second paragraph also states there is a separate read bus for RDPIN/RQPIN. Which seems to imply a RDPIN can coincide reliably with a WXPIN/WYPIN or even a WRPIN.

    Dunno. Still need Chip to clarify.

  • I can prevent two COGs from accessing the pin at the same time.
    What is important is that the values of different COGs are read and written.
    I had already guessed that I had to connect two pinns.
    Now I have to think a little bit about it.
    Hopefully there will soon be a documentary about the P2 PASM like the P1.

  • evanhevanh Posts: 10,918

    The hardware doc is a Google doc, it's the first link of the list here - https://www.parallax.com/propeller-2/documentation/
    It can be easy converted to PDF, but the paging isn't very tidy.

  • evanhevanh Posts: 10,918
    edited 2021-03-11 23:31

    We were able to notably refine the block diagram above when Chip put together a single sheet for us from his I/O pad-ring design files. There was lots of little details like finding out there is an entirely separate DAC for the comparator. And that "input" is muxed with the ADC bitstream after the feedback route.

  • evanh@
    The only thing I can find here is a table with all PASM commands.
    I miss some explanation how the commands work.

    Or am I just deaf on the eyes? :)

    https://drive.google.com/open?id=1_vJk-Ad569UMwgXTKTdfJkHYHpc1rZwxB-DcIiAZNdk

  • evanhevanh Posts: 10,918
    edited 2021-03-12 12:16

    That's the second link in the list. You probably just mis-clicked. Here's a copy of the correct link https://docs.google.com/document/d/1gn6oaT5Ib7CytvlZHacmrSbVBJsD9t_-kmvjd7nUR6o/edit?usp=sharing

  • AribaAriba Posts: 2,468

    The only thing I can find here is a table with all PASM commands.
    I miss some explanation how the commands work.

    This table has a column called description which expains what the instruction does. It helps if you know a little Verilog for some descriptions. And more complicated instructions are described in the document Evanh has linked.
    If both not helps you normally find explanations in this forum with a search. Or just ask.

    Here is a good thread:
    https://forums.parallax.com/discussion/169069/p2-tricks-traps-differences-between-p1-reference-material-only/p1
    Andy

  • The description is in some cases a little too short.
    And Verilog is not exactly my case.

    Well, I think I'll have to ask questions here several times.

    Googletranslator doesn't make this much better either.

Sign In or Register to comment.