[comedi] Data aquisition rate

------=_Part_2002_1635055471.1645529500715
Content-Type: multipart/alternative; 
	boundary="----=_Part_2003_329125536.1645529500715"

------=_Part_2003_329125536.1645529500715
Content-Type: text/plain; charset="UTF-8"

Hello everybody, it's me again

Consider the following system:
Debian 10
Ryzen 3600x (SMT disabled)
16 GB RAM
Nvme SSD
NI PCI-6031E (100 kS/s)

We are running a C program with comedi, which is looping over for(;;) for a 
measurement of a voltage and we react to it by setting a new output voltage 
according to a very simple logic. (Program is executed on an isolated cpu 
core)
The reaction time (executing the loop once) is around 15 micro seconds and 
at max 24 micro seconds. Theoretically with this card 10 micro seconds 
should be possible.

My question is, how on earth do we not reach the maximum potential of a 20 
years old card with such modern hardware? Surely there must be a way to 
reach the maximum? Is there a special way to code for those cards? Is 
comedi a limiting factor? Do I need to use two cpu cores and therefore 
parallelize the code?

Thank you very much in advance

-- 
You received this message because you are subscribed to the Google Groups "Comedi: Linux Control and Measurement Device Interface" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
To view this discussion on the web visit https://groups.google.com/d/msgid/comedi_list/e5c06021-4aa5-4d71-87dc-feaa6c7f60cfn%40googlegroups.com.

------=_Part_2003_329125536.1645529500715
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div>Hello everybody, it's me again</div><div><br></div><div>Consider the f=
ollowing system:<br>Debian 10<br>Ryzen 3600x (SMT disabled)<br>16 GB RAM<br=
>Nvme SSD</div><div>NI PCI-6031E (100 kS/s)</div><div><br></div><div>We are=
 running a C program with comedi, which is looping over for(;;) for a measu=
rement of a voltage and we react to it by setting a new output voltage acco=
rding to a very simple logic. (Program is executed on an isolated cpu core)=
<br></div><div>The reaction time (executing the loop once) is around 15 mic=
ro seconds and at max 24 micro seconds. Theoretically with this card 10 mic=
ro seconds should be possible.</div><div><br></div><div>My question is, how=
 on earth do we not reach the maximum potential of a 20 years old card with=
 such modern hardware? Surely there must be a way to reach the maximum? Is =
there a special way to code for those cards? Is comedi a limiting factor? D=
o I need to use two cpu cores and therefore parallelize the code?</div><div=
><br></div><div>Thank you very much in advance<br></div>

<p></p>

-- <br />
You received this message because you are subscribed to the Google Groups &=
quot;Comedi: Linux Control and Measurement Device Interface&quot; group.<br=
 />
To unsubscribe from this group and stop receiving emails from it, send an e=
mail to <a href=3D"mailto:[email protected]">comedi_=
[email protected]</a>.<br />
To view this discussion on the web visit <a href=3D"https://groups.google.c=
om/d/msgid/comedi_list/e5c06021-4aa5-4d71-87dc-feaa6c7f60cfn%40googlegroups=
.com?utm_medium=3Demail&utm_source=3Dfooter">https://groups.google.com/d/ms=
gid/comedi_list/e5c06021-4aa5-4d71-87dc-feaa6c7f60cfn%40googlegroups.com</a=
>.<br />

------=_Part_2003_329125536.1645529500715--

------=_Part_2002_1635055471.1645529500715--