[App_rpt-users] Beagleboard and USB URI dongle

[Steve Gladden]

I see those same type of audio issues with fast processors & slow processors.

I've been told it's because of how Linux and USB works (or does not work)
for that matter.

And I have been told the USB interface is not a production quality solution.

I wish I better understood this first hand but currently  do not.

I have a lot of personal suspicions of what it might involve but do not
have the software technical know how to test & measure the kernel &
software
activity going on while the audio anomolies occur.. versus while
they are NOT occurring.


I have a lot of those --it might be this or it might be that thoughts but
nothing to back it up or do much about it at this time..
Might even be fault Asterisk itself or Zaptel interaction with the channel
driver for instance :-)

I am at a point where if one (or more) of the developers feels like
digging in I'd happily purchase the EXACT same hardware they are
developing on/working with so that I could be more helpful in
troubleshooting this and reporting issues that could easily be reproduced
 that would happen for the devs the same way.

Although I *think* this is happening across the board to some degree.
(from my own testing on multiple systems and my interaction with other users)
It's subtle under normal use and most operators don't notice it or it's
just not service affecting enough to have become a primary concern.

It's a miracle the software is here and works and as-is
and that's its been made available for free and is getting constant
cool & useful updates at user's requests..
That's very col & exciting.

The Happiness/Grief ratio (H/G) is very high in this area :-)

But yes I'd like to see the audio output to Transmitter work perfectly
And work toward that end If it's possible.
Or at least understand first hand what gives.

I don't want to complain out here I'd like to learn something.
This type of software and building on it is the present and
future of Radio/TeleCommunications amateur and professional.
I plan to be involved, be useful and enjoy the hell out of it.

-Steve










>>BTW- runing full duplex chan_usbradio on a Pentium-3 733MHz box
>>I see 10%CPU continuous (RX) with spikes up to 25-30% during TX-RX-DTMF
>>useage
>
>>That's with Top runing at a 20 sample per second rate (which eats some
>> CPU
>>itself mind you.
>
>>This seems to suggest that an 'ol Pentium-3 can cut it for this app.
>
>> Somebody please beat me with a baseball bat if I'm wrong!!
> I had similar results with my P3/866, but I was hoping that a faster
> processor might help to eliminate the occasional
> "stumble" I've noticed with Allstar. Especially at the beginning of
> telemetry or user audio, the audio will sometimes glitch.
>
> It occasionally happens with user audio also. I've not noticed that before
> when I ran an IRLP or EchoIRLP system.
>
> I'm not sure if it's a USB-related phenomenon, since my previous systems
> have used analog sound cards, rather than USB fobs.
>
> If nodes with faster processors also have that issue, then I won't spend
> any more time or energy trying to upgrade my PC, or change to simpleusb,
> etc. I'll just live with it.
>
> I figured that if the Simpleusb would be less processor-hungry, maybe that
> would minimize the glitches?
>
> 73.
>
> Kyle
> K0KN
> --- Original Message --->>>also seems that the parallel port support isn't
> included in
> simpleusb?
>
> Can you confirm this?
> Might be really easy to get it.
> I'll certainly be wanting to do this. :-)
>
> I'm still getting free Pentium III 1U rack serves lately :-)
> Which are quite sexy looking and convenient to use for this.
> The used 1U Quad Xeons I purchase are going for about $850.  hihi
>
> BTW- runing full duplex chan_usbradio on a Pentium-3 733MHz box
> I see 10%CPU continuous (RX) with spikes up to 25-30% during TX-RX-DTMF
> useage
>
> That's with Top runing at a 20 sample per second rate (which eats some CPU
> itself mind you.
>
> This seems to suggest that an 'ol Pentium-3 can cut it for this app.
>
> Somebody please beat me with a baseball bat if I'm wrong!!
>
>
>
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>
> _______________________________________________
> App_rpt-users mailing list
> App_rpt-users at ohnosec.org
> http://ohnosec.org/cgi-bin/mailman/listinfo/app_rpt-users
>


Michigan Broadband Systems Inc.
"Always Connected"

(734)527-7150

Steve's cellphone: (734)904-1811


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.