When using FDwfDigitalOutPlayDataSet with large sample values results in corrupted signals

[RRahul]

When trying to play a pattern in an infinite loop with a sample frequency of 8MHz and 6,000,000 samples a corrupted signal appears at the beginning of every period. My pattern is still visible but before the rising edge there are some signals which seem to be random. Additionally what is the format for the data for example to have one line be constant 0 while the other lines are doing some clock signals

CODE:

"""

   DWF Python Example

   Author:  Digilent, Inc.

   Revision:  2022-02-11

 

   Requires:          

       Python 2.7, 3

"""

 

from ctypes import *

from dwfconstants import *

import sys

import time

 

if sys.platform.startswith("win"):

    dwf = cdll.dwf

elif sys.platform.startswith("darwin"):

    dwf = cdll.LoadLibrary("/Library/Frameworks/dwf.framework/dwf")

else:

    dwf = cdll.LoadLibrary("libdwf.so")

 

hdwf = c_int()

sts = c_ubyte()

 

version = create_string_buffer(16)

dwf.FDwfGetVersion(version)

print("DWF Version: "+str(version.value))

 

dwf.FDwfParamSet(DwfParamOnClose, c_int(1)) # 0 = run, 1 = stop, 2 = shutdown

 

print("Opening first device")

dwf.FDwfDeviceOpen(c_int(-1), byref(hdwf))

 

if hdwf.value == 0:

    print("failed to open device")

    szerr = create_string_buffer(512)

    dwf.FDwfGetLastErrorMsg(szerr)

    print(str(szerr.value))

    quit()

 

dwf.FDwfDeviceAutoConfigureSet(hdwf, c_int(0))# 0 = the device will be configured only when calling FDwf###Configure

 

print("Configuring Digital Out...")

 

hzPlay = 10e6

# for infinite playback fill the entire 256MiByte memory

nSamples = 6000000

rgbSamples= (c_ubyte*int(nSamples))()

 

for i in range(len(rgbSamples)):

    if i < 3000000:

        rgbSamples[i]=1

    else:

        rgbSamples[i] = 0

 

dwf.FDwfDigitalOutRunSet(hdwf, c_double(nSamples / float(hzPlay)))

dwf.FDwfDigitalOutRepeatSet(hdwf, c_int(0)) # infinite repeats

 

# enable play mode for the wanted signals

for i in range(8):

    dwf.FDwfDigitalOutEnableSet(hdwf, c_int(i), c_int(1)) # enable

    dwf.FDwfDigitalOutTypeSet(hdwf, c_int(i), DwfDigitalOutTypePlay)

    dwf.FDwfDigitalOutIdleSet(hdwf, c_int(i), DwfDigitalOutIdleLow)

 

print("Samples:"+str(nSamples)+" Rate:"+str(hzPlay)+"Hz "+" Period:"+str(nSamples/hzPlay)+"s")

dwf.FDwfDigitalOutPlayRateSet(hdwf, c_double(hzPlay)) # play sample rate

# set play data array of 8 bit samples

dwf.FDwfDigitalOutPlayDataSet(hdwf, byref(rgbSamples), c_int(8), c_int(int(nSamples)))

 

print("Arming Digital Out...")

dwf.FDwfDigitalOutConfigure(hdwf, c_int(1))

time.sleep(100)

dwf.FDwfDeviceCloseAll()

 

test.py

Edited September 11 by RRahul

[attila]

Hi @RRahul

Fill the entire buffer or use FDwfDigitalOutWaitSet

23 hours ago, RRahul said:

# for infinite playback fill the entire 256MiByte memory

 

[RRahul]

how do I fill the entire buffer ?

 

[attila]

By downloading an array of 256MiB which contains your data repeated times.

[RRahul]

@attilais there a code example of this ?

Edited September 12 by RRahul

[attila]

like use nSamples = 256*1024*1024 in your script

[RRahul]

Thanks 

 

 

[RRahul]

When i change the nSamples to 256*1024*1024 I get this error: 

[attila]

Hi @RRahul

See the following example: DigitalDiscovery_Play8bit.py

Edit: Are you using 64bit Python ? I don't know what limits does the 32bit Py have.

[RRahul]

@attila when I run the example on my machine it exits with code 1 as well

 

[RRahul]

got this error: 

 

 

Edited September 17 by RRahul

[RRahul]

@attila i am running 64 bit python

[RRahul]

@attila when I run it from command prompt there are no errors however I have hooked it up to an external oscilloscope and there the output does not make sense. I expected it to be some sort of square wave

Edited September 12 by RRahul

[attila]

Hi @RRahul

I see the problem now. It will be fixed tomorrow.

[attila]

Hi @RRahul

 

[RRahul]

Thanks, it works now !

[RRahul]

@attila one last question, is it possible to not fill the full 256M and repeat the pattern infinitely without getting garbage/corrupted signals. Is there a work around this issue ?

[attila]

Only by using wait of about 1us, for the DDR RAM read cache to be refilled, before restart. This play features was later added, on used request, and not to sophisticated. 
If 32ki samples / dio is sufficient you can use the simple custom mode.

[RRahul]

@attila could I also have an example of creating a custom pattern using play with an arbitrary array not the "rgbValues[i] = i" to further my understanding of the how the data is stored and used by the signals ?

[RRahul]

Additionally is it possible to make a constant pattern at digital high or '1' using play ?

[attila]

In play, successive samples are output on the specified number of DIOs at a specified rate.
It may be more constructive if you explain what you want you want to achieve. There may be another solution to this.

[RRahul]

For example if my data is [3,1,3,1,3,1] then on DIO-0 it will be [1,1,1,1,1,1] however when I probe that pin with an oscilloscope it's not constant at digital high.

[attila]

Hi @RRahul

Generated with Digital Discovery and captured with ADP2230 DigitalDiscovery_Play8bit.py
DwfDigitalOutIdleLow and DwfDigitalOutIdleHigh

See the WF SDK / samples/ py and manual pdf

7 hours ago, attila said:

It may be more constructive if you explain what you want you want to achieve. There may be another solution to this.

[RRahul]

When I use DwfDigitalOutIdleHigh it results in this:

 

[attila]

43 minutes ago, attila said:

 

[RRahul]

On 9/13/2024 at 10:32 AM, attila said:

Only by using wait of about 1us, for the DDR RAM read cache to be refilled, before restart. This play features was later added, on used request, and not to sophisticated. 
If 32ki samples / dio is sufficient you can use the simple custom mode.

@attila For this 1us delay what are the requirements of the pattern generator. Are all signals held at 0, or held at the last value, etc.