On 2/11/07, Herbert Graf wrote: > On Sun, 2007-02-11 at 00:52 -0800, Vasile Surducan wrote: > > On 2/10/07, piclist@mmendes.com wrote: > > > I haven't actually done any tests to see if the transfer is trully at > > > 108, but what I have done is try to connect from diferent distances from > > > the router, most of the time outside. As the signal degrades over the > > > distance the speeds (indicated in the taskbar icons) do switch to lower > > > or higher depending on the quality of the signal. This was done for a > > > very picky friend who wanted to be able to use his laptop while on a > > > floter in his pool (I know, I wouldn't do it, but he wanted it) > > > > > > Obviously, you can only get 108 when using an 108 router with an 108 > > > card. > > > > The maximum speed allowed in 802.11g standard (2.4Ghz) is 54Mbps, > > named as 108, page14 of 802.11g standard, downloadable here: > > > > http://standards.ieee.org/getieee802/802.11.html > > > > Do I miss something ? or there is available a newer specification ? > > For pure g, yes, 54Mbps is the fastest. > > However, because the IEEE has taken a stupidly long time to come out > with the next standard (n), the manufacturers didn't want to wait (g > hardware had reach commodity prices, not much margin when prices get > that low). So they started developing extensions to g that could go > faster. Most claim speeds of up to 108Mbps (and often improved distance > specs). > > The rub is they are NOT compatible between different chipsets. This > means if you want a 108Mbps network (often called super g), you must > ensure every bit of equipment uses the same line of chipsets. Easiest > way to ensure this is to buy the same brand, but even that isn't 100% > since sometimes brands change chipsets, meaning the older 108Mbps > hardware may not work with the newer 108Mbps hardware. > > On top of this confusion is the MIMO or "pre n" stuff. Usually you can > tell this stuff separately from the "normal" 108Mbps stuff by the number > of antennas, MIMO usually has three, although sometimes one is > "internal", further mucking up the market. > > Personally, I won't touch any feature beyond 54Mbps until n is fully out > there. My router is capable of 108Mbps, but I've never even enabled it. > > The part that really bugs me is the IEEE members. They are really the > ones to blame for the mess that is WiFi right now. They kept arguing and > squabbling, holding up any chance at a reasonable time period for the > next standard. Consumer confusion is high, interoperability is in the > tubes, and even worse, many of the MIMO technologies out there right now > will KILL your g bandwidth completely. MIMO technologies usually take > two WiFi channels. Despite the numbering, in the 2.4GHz band there are > only 3 non overlapping WiFi channels (1, 6 and 11). If you are using g > on channel 6 and your neighbour powers up most MIMO routers, there's a > good chance that your bandwidth will drop like a stone. Newer MIMO > routers are better then older ones, but the problem is still there. > draft-n products supposedly are MUCH better, but there is still chance > for the problem appearing. > > Anyways, for more information try http://www.tomshardware.com, lots of > articles there on what's out there and what to avoid, if you REALLY want > to go faster then 54Mbps. Me, I run a cable... Herbert, Thank you very much for your nice comments and links. I'm to my second WIFI/WIMAX design project and those different standards are really puzzling me. The same s!it is on 802.[1]16e2003/2005. One of the most interesting situations is WIFI on 802.11b/g with MIMO features at 54Mbps.Those cards (which must be used with routers having the same feature) are increasing the speed and/or coverage keeping low the datarate. Vasile -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist