James,

The problem with ARQ is latency of the audio stream and there are many stages when this can occur; Firewire or USB, audio buffers, VAC, DSP, etc...

All it takes is one point in the signal path to inject a lot of latency and the whole process falls apart.

What version of VAC are you using? This component is the one I would suspect first.

I do not think this is an issue with PTT or the RigCAT configuration, as PowerSDR is going to react very quickly the the ZZRX command once sent by the software and the radio will be transitioned from TX to RX. Even if the audio stream is still sending audio. I do not know of a way of introducing an artificial delay in processing CAT commands from the PowerSDR or RigCAT side of the CAT connection.

I have heard that some ARQ programs have a parameter to delay the sending of the RX CAT command to compensate for this condition.

Also, the issue may be inherent in the 1500 itself. a 48K sampling rate may be too slow to process everything in the time frames needed for ARQ operation. I know this works on the FLEX-3000/5000 at 96K with the proper buffer setting. I can not recall anyone trying to do this with a 1500.

-Tim
---
Tim Ellison, W4TME
Product Management, Sales & Support
FlexRadio Systems^(TM)
4616 W Howard Ln Ste 1-150
Austin, TX 78728
Phone: 512-535-4713 Ext. 223
Email: [email protected] <mailto:[email protected]>
Web: www.flexradio.com <http://www.flexradio.com>

logo


On 12/14/2011 11:47 PM, James Nelson wrote:
Hi Everyone,

When using FLdigi/FLarq on the Flex1500 Psdr Ver 2.2.3 on either XPx86 or Win7x64 a phenomena of early 
audio truncation occurs with the PTT being released too early OR the audio arriving too late depending on 
how you want to interpret the culprit. This presents a problem with fast ARQ such as "MT63" as 
the "<EOT>" frame marker becomes incomplete/damaged therefor a connect ack is not 
possible. Another indication/symptom of this problem shows up in the FLdigi CW Ident at approx 20 wpm or 
faster so in my case the call sign becomes VK2JT in place of VK2JN or another example VK2ACL becomes 
VK2ACA.

Thinking about latency general in recent time ARQ modes such as pactor&  RMS 
Express(Winlink 2000), these problems through settings have been mostly worked out. 
In this case could it be the XML file file requires an update somewhere to counter 
this ? I see the Author of the XML file is our always at the helm Tim Ellison.

A good 10 hours between 2 of us has been spent changing buffer sizes&  settings 
in many areas of both Psdr, FLdigi and VAC to no avail.  Another interesting aspect 
is there have been some occasions (that have been rare) that the truncation has not 
been evident that makes this even more puzzling.

This issue does not appear evident on testing with the 5K

Would really like some ideas on this baffling little nasty.

73'

James  VK2JN

_______________________________________________
Flexedge mailing list
[email protected]
http://mail.flex-radio.biz/mailman/listinfo/flexedge_flex-radio.biz
This is the FlexRadio Systems e-mail Reflector called FlexEdge.  It is used for 
posting topics related to SDR software development and experimentalist who are 
using beta versions of the software.
_______________________________________________
Flexedge mailing list
[email protected]
http://mail.flex-radio.biz/mailman/listinfo/flexedge_flex-radio.biz
This is the FlexRadio Systems e-mail Reflector called FlexEdge.  It is used for 
posting topics related to SDR software development and experimentalist who are 
using beta versions of the software.

Reply via email to