AnsweredAssumed Answered

vrf Linearity - solved

Question asked by VRFuser on Dec 15, 2003
Yes,

Jay is right, you will not get a better soulution with more MIPS. Asssuming your washer-process ist continous, you will not get any improvement by using a AD-card with burst, FIFO and all the other goodies. It stucks only a few seconds later...

Saw the same problem year ago in a continous process measuring the thickness of ribbon every 1ms. My customer split his application into two: one machine pure data collection and checking the limits for pass/fail, the other machine was human-machine-interface for setting limits, data storage and producing quality statistics.

My recommendation is these days a linux machine - they can arbitrate down to less than 1 microsecond (1e-6 sec)continously off the shelf with a few lines of code. The parameter and data exchange can be VEE-programmed using TCP/IP 'To/From Socket'or, in this case with good performance, through the parallel port to a VEE /W98 machine. (Yes, call me a twit, please - but it works... ;_)))


/Stefan

P.S.: Last comment, I have to disagree stronly: VEE users are the true heroes of realtime-applications !!!

-----Ursprngliche Nachricht-----
Von: Jay Nemeth-Johannes [mailto:jay.johannes@completetest.com]
Gesendet: Freitag, 12. Dezember 2003 03:08
An: VRF
Cc: 'Gross, Jim'; Marc@MeadowFace.net; 'Osborn, Bob'; 'Hopp, Raymond'
Betreff: [vrf] RE: VEE Speed Question


I am going to go out on a limb here and make a blanket statement that you are not going to see the improvements you need simply by moving to a faster processor.

The problem is not a simple one to answer as the exact numbers are tied up in things like windows non-real time behaviour.  This means that you cannot guarantee when you will get access to your card, nor how long you will have access before swapping out.

In order to meet your goals, I believe you need one of two changes.  The first would be to go with a data acq card that had enough reading buffer that you could fill it at your data rate and then burst it out when you got windows access.  Something like a 10 reading memory should suffice.

The second possibility is to go with a real time operating system.  This would guarantee response in a particular time window.  Assuming that the DAQ card can read at this rate, it would solve the job.  However, at this point we are in the wrong group to be discussing this as VEE is not a real time app.

Jay Nemeth-Johannes
Complete Test
720 SW 14th Street
Loveland, Colorado 80537

www.completetest.com
> -----Original Message-----
> From: Brutzman, Bill [mailto:BillB@hkMetalCraft.com]
> Sent: Thursday, December 11, 2003 9:14 AM
> To: VRF
> Cc: Gross, Jim; 'Marc@MeadowFace.net'; Osborn, Bob; Hopp, Raymond
> Subject: [vrf] VEE Speed Question
>
>
> We would like our test equipment [under development] to be able to
> take measurements every 0.5 msec.
>
> Right now, our current Win98 PC, Pentium III, 550 MHz, 128 MB ram
> seems to max out at measurements every 3 msec.
>
> We measure thickness of 1/2" washers using a rotating bowl, shaft
> encoder, laser, and A/D card.
>
> I am writing to request estimates on speed improvements we can expect
> to see by installing a faster PC.
>
> I am wondering about CPU speed, memory for speed, XP for speed.
>
> Dell catalog indicates...
>
>   Price         CPU [GHz]     Up to [GHz]     with Front Side Bus
>   $599     P4     2.4                      2.4          400 MHz
>   $999     P4     2.6          3.2          800
> HyperThread
>   $1099     P4     2.8          3.2          800
> HyperThread
>   $1249 Xeon     2.4          3.2          -               Dual
> Processor Capable
>   $1399     Xeon     2.66          3.2          -
Dual
> Processor Capable
>
> Suggestions would be appreciated.  Of course, we would like to get it
> right the first time.
>
> Regards,
>
> Bill Brutzman, Manager IT
> HK MetalCraft Mfg Corp
> PO Box 775
> 35 Industrial Road
> Lodi  NJ  07644-0775
>
> 973.471.7770 x145 .voice
> 973.471.9666      .fax
>
> BillB@hkMetalCraft.com
> www.hkMetalCraft.com
>
>
>
>
>
> ---
> You are currently subscribed to vrf as: jay.johannes@completetest.com
> To subscribe send a blank email to "join-vrf@it.lists.it.agilent.com".
> To unsubscribe send a blank email to
> "leave-vrf@it.lists.it.agilent.com".
> To send messages to this mailing list,  email "vrf@agilent.com".
> If you need help with the mailing list send a message to "owner-
> vrf@it.lists.it.agilent.com".




---
You are currently subscribed to vrf as: Stefan.Paulick.extern@brose.net To subscribe send a blank email to "join-vrf@it.lists.it.agilent.com".
To unsubscribe send a blank email to "leave-vrf@it.lists.it.agilent.com".
To send messages to this mailing list,  email "vrf@agilent.com". 
If you need help with the mailing list send a message to "owner-vrf@it.lists.it.agilent.com".

---
You are currently subscribed to vrf as: rsb@soco.agilent.com
To subscribe send a blank email to "join-vrf@it.lists.it.agilent.com".
To unsubscribe send a blank email to "leave-vrf@it.lists.it.agilent.com".
To send messages to this mailing list,  email "vrf@agilent.com". 
If you need help with the mailing list send a message to "owner-vrf@it.lists.it.agilent.com".

Outcomes