The typical maximum througput for an 802.11g 54Mbps wireless connection is about 22 Mbps. (The radio works at 54Mbps, but in only one direction at a time.)
At 7m the wall might, depending on what it is made out of, actually block the signal. If it is typical home construction though, it won't, and you will probably get a 54Mbps connection.
Note that as long as you get a wireless connection rate that is higher than roughly 1 Mbps, you will have more "bandwidth" available on your wireless connection than you can possibly use with a 512Kbps Internet link.
It is possible... if your walls are made of metal, cement, or other material that will not pass 2.4Ghz radio signals. (Which is not likely!)
It's not a telephone extension. It's a CAT5 ethernet cable running either 10baseT (10 Mbits/sec) or 100baseTX (100 Mbits/sec). The same cable is used for voice telephony (POTS) but the designation is by the function.
Nope. Your perceived speed is limited by the slowest bottleneck on the system. At 0.5Mbit/sec sec, your ADSL connection approximately matched by the very slowest speed of an 802.11b/g wireless connection (1Mbit/sec) which hopefully, you will never reach.
The enemies of wireless connections are (in order of decreasing importance):
Multipath and reflections
Interference is a problem is you're located in a tall glass building, with a view of the city and it's hundreds of other 802.11b/g users. You could be right next to your access point and still have problems communicating in the presence of interference. Also add microwave ovens, cordless phones, X10 video cameras, and other sources of 2.4GHz noise.
Attenuation is deteremined by the building construction. If the inside walls are made of some material impervious to 2.4Ghz RF, you will not have enough signal to maintain a useable connection. The usual cuprit is aluminium foil backed fiberglass insulation in the walls.
Multipath and reflections are where you have more than one path between the access point and your wireless client. At some positions, they add, at others, they cancel. In both cases, they are not stable and will vary. It is difficult to detect or calculate multipath problems. Wildly varying indicated signal strengths are a good clue. The symptoms are that you can obtain a connection, but might have trouble maintaining the connection.
I am considering buying the Netgear DG834G which will be connected in a wireless maner to my PC so that I do not need to use a telephone extension cable for my ADSL connection - which is currently a trip hazard!!
I was told that this may cause a performace hit, especially for the lower bandwidth connections (I am currently using 512KB). Since I will be only about 7 metres from the wireless access point in another room on the same floor, does anyone think that I may suffer a performance hit using the DG834G over using a telephone extension cable with ADSL modem?
Yes, there is technically a speed difference between wired and wireless. However, when accessing the Internet, you won't see the difference at all as your DSL modem operates at a much slower rate and is thus the limiting factor in your overall speed regardless of which type of "ethernet" (wired or wireless) you use.
Guesswork. It's my specialty. I also like detective stories, mysteries, scientific who-done-its, answering usenet questions with insufficient information, and other forms of logical deduction. I think of it as intellectual exercise. A rather large percentage of questions lack sufficient information for a proper answer. A smaller number supply ambiguous information such as the original reference to "telephone extension" and your reference below to "cable and connectors". I don't really blame people for doing this as I would not expect proper buzzwords from someone with limited experience in wireless or computing. Same with English as a 2nd language. However, that doesn't stop me for pointing out that there is room for improvement.
Most users plant the DSL modem on the floor somewhere and run ethernet to their computah. That's so they don't have a big box DSL modem sitting on their desk. However, I guess there are exceptions.
Amazing. My 10baseT RJ45 connectors are the same size as my RJ45
100baseTX connectors. I guess you mean the RJ11/14 telco connector is a different size as the RJ45 ethernet connector. Incidentally, SBC ships their DSL modem kit with a rather heavy gauge telephone cable in the box. I've seen an amazing number of customers shove it into the ethernet port of their DSL modem and then call me wondering why nothing is working. If the connector fits, someone will try it.
Your biggest difference would be with the authentification. sometimes it may take up to a minute for everything to handshake. Once that's up it should be quick enough although my wife swears she can tell the difference - maybe encryption processing time? cb