How to use the full 256M sample memory with 8-bit data on the Digital Discovery

[Jope]

Hello,

I have a problem concerning the Digital Discovery using the Waveforms SDK.

 

I thought it is possible to use the full 256MBytes on-board recording memory for 8-bit sampling, but I am only able to read 64M samples, no matter if I use 32 bit, 16 bit or 8 bit sampling (which I configure with the function FDwfDigitalInSampleFormatSet).

The maximum size I can set via FDwfDigitalInBufferSizeSet() is 64M (i.e. 67108864), no matter what sample size I set via FDwfDigitalInSampleFormatSet(). That is, if I try to set a higher value for the buffer size, FDwfDigitalInBufferSizeGet() will still return 67108864, regardless which sample size is set (8, 16, 32).

So I thought this buffer size value is the value for 32-bit sampling, and therefore 4 times as big when using 8 bit sampling. But unfortunately, when I try to read the data via  FDwfDigitalInStatusData(), with a countOfDataBytes value that is higher than 67108864, all bytes after the 67108864 byte are zero.

I am using single aquisition mode, and start an aquisition with FDwfDigitalInConfigure(handle, true, true) and then wait until the device state becomes DONE.

 

I would be very thankful if someone could explain what I am doing wrong here. The Waveform GUI seems to be possible to do that (64M samples for 32 bit sampling, 256M samples for 8 bit sampling), so I thought it is possible with the SDK, too. Thanks.

 

[attila]

Hi @Jope

See the following example: DigitalIn_Acquisition_8x256M.py
The captured data looks good from first to the last byte.

>python DigitalIn_Acquisition_8x256M.py
DWF Version: b'3.11.4'
Opening first device
Generating gray counter
Configuring DigitalIn
DigitanIn base freq: 800000000.0
Sample rate: 100000000.0
Buffer size: 268435456
Waiting for acquisition...
   done
Writing to file...
   done

[Jope]

Many thanks! It works.

I made a stupid mistake. I've programmed a class in C++ for talking to the device, and in the method for setting the sample buffer size I made a sanity check against a maximum value. The max value I checked against was obtained before from calling FDwfDigitalInBufferSizeInfo(), which gives the 64M value for 32-bit sampling... So my buffer was never set higher than 64M.

Thanks again.