Hello Stan,

> With WLR2-26 on a standalone computer, this time a shuttle, 3 Ghz Pent4, 
> 1 GB RAM, delta44
> With 2 folders, one set up for the WLR2-26 as a MASTER, the other folder 
> WLR2-26
> setup as a SLAVE. 2 Icons on the main screen for ease of starting.
> All using common group and port numbers, RAW16 bit data

> Master is set up and operation is orderly by itself, close it out.
> Setup the slave, it listens to the network. Close it out.
> Start up the MASTER, and start up the SLAVE. memory error in linrad, 
> fft1, fft2, close out both.
Do you mean that the slave prints the message memory error in linrad, 
fft1, fft2 ? (It should also give two numbers.)

Linrad uses several large arrays for processing fft1 and fft2.
All of them are allocated as a big block with small gaps
between each array. There is some code in the background
that monitors these gaps and notes if the data has been
changed. This should never happen but if something
goes wrong and some process writes outside its intended
memory area this error message is produced. One of the
numbers would tell me which gap is written into and thereby
which array is likely to be incorrectly set up.

At this point the master should contiue as if nothing happened.
You exit the slave using any key while the master continues
to run. Then you make a normal exit from the master?

> Start up SLAVE, it listens to the network, Then start up the MASTER, 
> getting the memory error in linrad.
> I could not duplicate the window that said the memory error was in 
> linrad and fft1, fft2.
Do you mean that the memory error is now in the master?

Please save the complete sets of parameter files for
the two Linrad set-ups and send them to me. I want to
see how it works with exactly the same settings here. 

Maybe something is wrong with disk swapping? Try
to set all storage times to their minimum values
and de-select the second fft. If this works, check
memory and cpu usage. Then go gradually back to
the original settings to see if you can draw any

> failures are repeatable.  Sorry I broke it.......
Since it is repeatable this bug can be found.
I have verified that Linrad works with both master
and slave in the same computer under Windows 2000 
with a Delta 44 as input. I do not have Win XP 
on any stationary computer so I can not check it 
with a Delta44.

I have also verified the operation on a laptop where
I set 96 kHz, 24 bit and 2 channels I and Q to simulate
a Delta44. This laptop has a Pentium 4 and had Windows XP
Home Edition in its original state without any update 
or service pack. (The hard disk had a crash recently)

By selecting silly parameters I can provoke Linrad to
have more than 500 megabytes of processing data on
the hard disk with a cpu load of about 80% but still
both copies run reasonably well with their summed
audio in the loudspeaker. I see occasional errors
(overrun and network errors reported by Linrad) but 
nothing that affects processing much.

After one hour with Microsoft Update (and 66 updates)
everthing is still fine but a new visit to Microsoft
update shows that service pack 2 is not yet installed.
After spending one more hour with Microsoft Update I
have SP2 installed.

The default settings, Firewall enabled, does not allow
me to start Linrad. I get a pop-up window asking whether 
I want to keep some functions blocked. I answer Yes
and the master runs normal. The slave comes as far as 
"listening to network". When I exit and restart I get
no pop-up window. The network stays blocked for Linrad.
I reset the firewall, get the pop-up window and tell
Windows to allow Linrad to do what it wants. Everything
still runs fine. 

A new visit at Windows Update tells me that there are 
now 28 new updates. After installing them all I find 
that Linrad still runs fine with a master and a slave 
on the laptop.

I suggest you have the ethernet card enabled but unplug
the cable. Disable any WLAN or other network access.
Then it will be safe to disable the firewall and see
whether Linrad runs with both master and slave on the

My E-mail addres can be found at the bottom of this page:
I look forward to test your par_* files under Windows 2000
on a very similar computer with a Delta44. If there is
a bug inside Linrad, it should happen exactly as on your
machine, Linrad does not know anything about what Windows
version it runs under.


Leif / SM5BSZ

This message is sent to you because you are subscribed to
  the mailing list <linrad@antennspecialisten.se>.
To unsubscribe, E-mail to: <[EMAIL PROTECTED]>
To switch to the DIGEST mode, E-mail to <[EMAIL PROTECTED]>
To switch to the INDEX mode, E-mail to <[EMAIL PROTECTED]>
Send administrative queries to  <[EMAIL PROTECTED]>

Reply via email to