Author Topic: Rigol DS1000Z series buglist continued (from: 00.04.04.03.02)  (Read 47723 times)

0 Members and 2 Guests are viewing this topic.

Offline JohnPi

  • Regular Contributor
  • *
  • Posts: 73
  • Country: us
Re: Rigol DS1000Z series buglist continued (from: 00.04.04.03.02)
« Reply #325 on: September 03, 2018, 10:36:03 pm »
I looked through that code (I don't have a convenient system to build/run it on)-- it appears to download either 1200 or 250000 bytes. My problem is when downloading 8244+n*64 only -- 250000 is not an issue in that.

My problem arose when I was specifically downloading different lengths in the range below 30000; it arose from old habits of being frugal with memory. I changed my own code to download 8000 byte chunks, but I guess I could just download a single 30000 byte block and truncate it.

In the end, I don't know if this ia Rigol or a pyvisa bug.

p.s. I updated my example above to select the USB instrument more easily.
 

Offline MarkF

  • Frequent Contributor
  • **
  • Posts: 814
  • Country: us
Re: Rigol DS1000Z series buglist continued (from: 00.04.04.03.02)
« Reply #326 on: September 03, 2018, 10:54:33 pm »
It looks like you're trying to use a USB connection?

I don't know if it will have any effect.  But, you could try changing the OS low level receive buffer size.  The default size for a LAN connection would be different than a USB connection.
 

Offline JohnPi

  • Regular Contributor
  • *
  • Posts: 73
  • Country: us
Re: Rigol DS1000Z series buglist continued (from: 00.04.04.03.02)
« Reply #327 on: September 04, 2018, 02:19:50 am »
@PeDre - thanks for checking. I see it works for you at 8244. What is that program you used ?

When I run my pyvisa program with different values for chunk_size:
DS1054Z = GPIB.open_resource([_ for _ in GPIB_Resources if re.search('^USB.*:DS.*INSTR$', _)][0],chunk_size=3000)

it doesn't appear to fail as often, but there are definite slow downs at some points (watch the results scroll by) -- they appear to be at 64-count intervals. In my case, starting at 8200:
for NPOINTS in range(8200,30100):# fails at 8244 first

gives a slowdown (but not a fail) at 8226.

3.6.5 |Anaconda, Inc.| (default, Mar 29 2018, 13:32:41) [MSC v.1900 64 bit (AMD64)]
visa 1.9.1
 

Offline JohnPi

  • Regular Contributor
  • *
  • Posts: 73
  • Country: us
Re: Rigol DS1000Z series buglist continued (from: 00.04.04.03.02)
« Reply #328 on: September 04, 2018, 02:42:22 am »
It looks like you're trying to use a USB connection?

I don't know if it will have any effect.  But, you could try changing the OS low level receive buffer size.  The default size for a LAN connection would be different than a USB connection.

Yes, USB. I have tried with the LAN with longer lengths with no problems. I haven't tried this code over the LAN -- will do (it appears you have to reboot the scope to switch from USB to LAN).

-- update: - I can't get my pyvisa to work over the LAN. Anyone know the magic resource name ? IP = 10.1.0.54, port = 5555
Code: [Select]
GPIB = visa.ResourceManager()
DS1054Z = GPIB.open_resource('TCPIP0::10.1.0.54::5555::INSTR')
print("DS1054Z:", DS1054Z.query("*IDN?"))
fails

Don't know how to change the OS buffer size. There is a chunk_size option in pyvisa which has an impact -- it moves the starting point of the issue around, but doesn't eliminate the issue.
« Last Edit: September 04, 2018, 03:17:03 am by JohnPi »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf