Data Acquisition speed

Just about everything about Red Pitaya
Post Reply
Daniel_s
Posts: 3
Joined: Fri Mar 25, 2022 7:51 pm

Data Acquisition speed

Post by Daniel_s » Fri Mar 25, 2022 8:08 pm

I am trying to acquire data for 1-2 s (or ~50-100 Msamples) at ~30 MHz with 125-14.
I tried with the example python, and then Jupyter, and can collect up to125 MHz (decimation of 1), but only 16k points.
On the stream server application it seems that the limit is 10.4 MHz rate, whether using local or on TCP. I am using version 1.04-09.
Can a higher rate be achieved on the server application and or Python?
Otherwise, does c need to be used to bypass the SCPI server and achieve 30 MHz for longer measurement length?
Thank you!

pavel
Posts: 789
Joined: Sat May 23, 2015 5:22 pm

Re: Data Acquisition speed

Post by pavel » Mon Mar 28, 2022 10:15 pm

It is possible to achieve higher streaming speeds than those provided by the streaming application.

I think the streaming application is not using the most efficient means to stream the data.

There is also a known issue with the way the Red Pitaya developers patch and build the Linux kernel for their SD card images. As far as I know, the PHY driver was broken in 2017 and never fixed after that:

https://forum.redpitaya.com/viewtopic.php?f=9&t=1880

This affects the speed of network transfers. At least, that was still the case in 2020 when I last tested it.

Here is a link to a patch that I use to correctly set the PHY skew timings:

https://github.com/pavel-demin/red-pita ... #L163-L178

I have a few applications that can stream data from Red Pitaya to a PC using TCP/IP over Ethernet at around 60-80 MB/s.

Here is a link to the code of one of these applications:

https://github.com/pavel-demin/red-pita ... s/adc_test

It should be used with the SD card image available from the following link:

http://pavel-demin.github.io/red-pitaya-notes/alpine/

Daniel_s
Posts: 3
Joined: Fri Mar 25, 2022 7:51 pm

Re: Data Acquisition speed

Post by Daniel_s » Fri Apr 01, 2022 1:50 am

Hi Pavel,
Thank you very much for the thorough reply, and happy to hear that 60-80 MB/s is achievable.
In the past week I was figuring out how to write the card image and compile a C program to run on the card, having had no previous experience with this.
Currently I am able to write the RP provided card image (Debian I think?), and then compile, run and modify a sample C program from Red Pitaya; no problem.

When I try to use the SD card image at the link you provided, the system doesn't boot (only see solid green light). One difference is in the first case I wrote the .img file to the card, while with the zipped folder from the link I extract then copy to SD, or try to write with Win32Disk. Is there supposed to be something more than just copying the files in the card? I have been trying to also create a .img from the folder files, but maybe this is a useless step?
The second question is: Is the patch to set correct PHY skew timings included in the SD card image from your GitHub, or should I include this in?

I'm sure these are basic questions, and will continue to read/learn to understand this more....
Thank you,
Daniel

pavel
Posts: 789
Joined: Sat May 23, 2015 5:22 pm

Re: Data Acquisition speed

Post by pavel » Fri Apr 01, 2022 8:31 pm

I have just added to my notes some troubleshooting instructions based on my previous answers to similar questions. Please take a look. They can be found at this link.
Is there supposed to be something more than just copying the files in the card?
Just copying the files is exactly what needs to be done.
Is the patch to set correct PHY skew timings included in the SD card image from your GitHub, or should I include this in?
The patch is already included in the SD card image.

Post Reply
jadalnie klasyczne ekskluzywne meble wypoczynkowe do salonu ekskluzywne meble tapicerowane ekskluzywne meble do sypialni ekskluzywne meble włoskie

Who is online

Users browsing this forum: No registered users and 30 guests