XPost: comp.mobile.android, alt.os.linux, alt.comp.os.windows-10   
   From: dan@djph.net   
      
   nospam wrote:   
   > In article <4kn9fcdssa4jc7pirqp7b7cpgv6sg05gog@4ax.com>, Stephen   
   > wrote:   
   >   
   > [...]   
   >> But a device which doesnt have a direct wireless path to the far end   
   >> device the traffic is going to, mesh will increase the overhead for   
   >> that path by well over 50% because of the forwarding at intermediate   
   >> nodes.   
   >   
   > that's an overgeneralization and depends on many factors. in general,   
   > it won't be that high.   
      
   Due to the vast majority of "mesh" APs (exceptions exist) utilizing the   
   same radio for backhaul as they do client access; and  that WiFi is   
   half-duplex (only one device can transmit at a time), the losses for   
   adding "mesh" hops hover around 50%.  Sure, newer wifi protocols may   
   drop that a bit; but you're not gonna get less than about 40% loss on   
   each hop.   
      
   > [...]   
   >> Things can be a bit better than that if you use 2 different radios and   
   >> different frequencies for each hop so that the mesh devices dont self   
   >> interfere / block.   
   >   
   > which is what most of them do, sometimes more than just two and in some   
   > cases, there's a wired backhaul.   
      
   I've not seen "mesh" setups with multiple 5 GHz radios (i.e. "backhaul +   
   client access").   
      
   Far as wired backhaul is concerned, those APs are either the edges of   
   your mesh (and not the source so much of the losses), or your network   
   doesn't use "mesh" at all.   
      
   --   
   |_|O|_| Registered Linux user #585947   
   |_|_|O| Github: https://github.com/dpurgert   
   |O|O|O| PGP: 05CA 9A50 3F2E 1335 4DC5  4AEE 8E11 DDF3 1279 A281   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)