• Elpis on Will a Broadband Booster help Skype Users?

  "I have a question regarding the results. The intrinsic downlink bandwidth is larger with HBB1 compared to without it. I am not able to understnad why that is the case. Is this the reason for the smaller intrinsic latncy? I would have anticipated that when there are no other Internet activity, presence or absence of HBB1 should not matter."

  "Also I don't understand the packet loss statistics. Obviously I am misreading it, because in my reading HBB1 introduces packet loss in the receive direction."

  "Thanks for any clarification you can provide."

  Baseline numbers with "no other internet activity" were obtained by using the MaxSpeed test at www.numion.com. By "no other internet activity" I mean that all applications that make use of the internet were stopped. That is, applications I know about and have control over. What Windows was up to at the two different test times I can't say (automatic updates, wireless zero configuration hopping ... God knows what! Then there's the firewall, anti-virus, etc, etc that I didn't switch off for obvious reasons.). As I point out, the results are only informal, but I think they are indicative that the HBB1 does do what it says it does.

  Some people are surprised by a packet loss of more than 100%. But let me explain how I understand that number. Suppose some internet activity requires just a single data packet per second to get the job done. At t=0s, the packet is sent, but network conjestion means it's "lost". That's a 100% loss! At t=0.25s the packet is resent, but network conjestion means it's lost again! That's a 200% packet loss! And so on ... I think you can see that if both the original packets and the resends are "lost", then packet loss can exceed 100%.

  As to why the HBB1 can affect both incoming and outgoing packet loss is, I think, a reflection of the fact that outgoing packets are being given priority "on the wire" and so incoming packets can momentarily back up and get lost. The internal buffers (up and downlink) of the HBB1 are finite in size, and so if incoming traffic is held back to give outgoing packets priority, then the incoming buffer can overflow and so incoming packets can be lost.

  Bear in mind that the connection from cable/ADSL modem to HBB1 to rounter to PC is ethernet and incoming and outgoing packets therefore share the same wire on a time division basis. Therefore, high priority outgoing packets can most certainly nudge incoming packets off-the-wire, if that's how the HBB1 decides to manage traffic. Of course, this is just my take on how the HBB1 might do its job, but as it's undocumented, I don't know for sure. Andrew Sheppard

• Elpis on Will a Broadband Booster help Skype Users?

  "Interesting to see how it will be able to shape the traffic. Will it be able to make way for/prioritise Skype packets and/or can it blog/limit the Skype traffic to basic band with for voice?"

  The HBB1 really doesn't have any knobs or levers that allow any sort of real control. Presumably (as the internal workings of the HBB1 are not documented), in firmware, the unit builds up a statistical picture of data packet traffic and then applies some heuristic rules to prioritize them. Andrew Sheppard

• Aswath on Will a Broadband Booster help Skype Users?

  I have a question regarding the results. The intrinsic downlink bandwidth is larger with HBB1 compared to without it. I am not able to understnad why that is the case. Is this the reason for the smaller intrinsic latncy? I would have anticipated that when there are no other Internet activity, presence or absence of HBB1 should not matter.

  Also I don't understand the packet loss statistics. Obviously I am misreading it, because in my reading HBB1 introduces packet loss in the receive direction.

  Thanks for any clarification you can provide.

• jyden on Will a Broadband Booster help Skype Users?

  Interesting to see how it will be able to shape the traffic. Will it be able to make way for/prioritise Skype packets and/or can it blog/limit the Skype traffic to basic band with for voice?

• continued in the Skype Journal Comment River: 200 fresh comments

Skype Links

Author's Posts

• Phil Wolff
• Jim Courtney
• Martin Geddes
• Guest Bloggers

Thanks to our Founding Alumni:
• Stuart Henshall
• Bill Campbell

Archives

• November 2006
• October 2006
• September 2006
• August 2006
• July 2006
• June 2006
• May 2006
• April 2006
• March 2006
• February 2006
• January 2006
• December 2005
• November 2005
• October 2005
• September 2005
• August 2005
• July 2005
• June 2005
• May 2005
• April 2005
• March 2005
• February 2005
• January 2005
• December 2004
• November 2004
• October 2004
• September 2004
• August 2004
• July 2004
• June 2004
• May 2004
• April 2004
• March 2004
• February 2004
• January 2004
• December 2003
• November 2003
• October 2003
• September 2003

Featured Events