You sure you're including Overhead in your measurements of 24Mbps? A Wireless G connection is a Half Duplex connection. Unless the lab guys are sending data in only one direction with all Acknowledgements in the other direction disabled, they will by no means see a 54Mbps connection even in perfect conditions. They'll score right where you are hitting. Turn off Acks and at most you'll get 30Mbps in a single direction.
I do agree with the movement to get more up to date specifications in the router, considering how long Gigabit and Wireless N have been out and in formal specs. They don't cost any more than the typical cost of a G/100Mbps unit, either.
As far as not having a guarantee on speeds, to be honest, no one can guarantee speeds on a Wireless connection. Wireless is a whole animal in itself. Think about it this way: I drive up to your house with a laptop and park in front of it in a van. I run a few seconds of data sniffing using the laptop, and notice that you have a Wireless network running on Channel 1. In the back of my van, I have an array of batteries, along with 24 different assortments of Wireless B, G, and N gear each set up to operate on the same channel, with the rest of the settings set to the most optimal settings you could possibly have. I fire up the access points, and then begin to use my laptop to connect to your network (assume that I was given access). Even if I had an Excellent signal, the amount of interference would take the speeds and drop it down to some odd figure below 10Mbps. Link speeds would drop, and due to beacon and "Carrier Busy" signals from the other APs, everything will start to fall. Now, I start to transmit data on a few of the APs in the truck. No know destination, but just as broadcast data meant to saturate the links on the APs of choice. Now we have a case where speeds are falling even farther, as the APs all have to fight for air time as well as the devices operating on the same channel nearby. Additionally, Further away APs creating noise create collisions, further hampering the performance of the network.
See where I'm getting?
Now assume that everything was set up as it should be. I've got several vans located near your home in several key points, an extended wireless site survey was performed and an array of 24 APs are scatters throughout. I have 20 laptops scattered in key locations to be in range of at least 3 APs. Each AP runs on a different channel, 1, 6, OR 11 based on your network's channel as a base. Assume no networks or Wireless devices are in operation from youe neighbors. Each of the Wireless APs are connected to a central network, completely hard-wired and attached is a file server (SAN) meant to serve up 5TB test files over a Wireless link. Assume all APs are Wireless G routers, using 20Mhz Wide channels (spec design). Each AP has a device associated to it, and they all begin to download a 5TB file one at a time. In this scenario, what you will first see is the first device start to download at a rate of 24-26Mbps. As other devices begin to download, but are on different channels, the speeds will remain the same. Once the next device is selected and operates on a channel the first device uses, you will start to see the speeds drop a little bit, as either the device or the access point starts to detect carrier busy signals. As more and more APs fire up, you will start to see speeds start to suffer across all APs, including yours, as noise is induced causing collisions to occur (signal too weak to recognize carrier busy signal). At the same time, your network is starting to suffer performance related issues due to devices in unknown locations and the physics of how radio signals travel.
Now assume we are using Wireless N APs in a mode they should NEVER be set in; 40Mhz wide channels on 2.4Ghz. Each AP is rated for 300Mbps Theoretical in a Half Duplex setup. Same thing as above, as performed, however this time, range and speed suffer at an inconsistent rate due to the amount of noise that is being generated. At the same time, some APs begin to fail to transmit as error counts start to increase beyond an ideal threshold. ACKS fail, the AP stops transmitting data except for keep-alive and broadcast data, and other APs begin to see space freeing up. Speeds go up, and the stalled APs are at reduced link speeds as a result of what happened.
So, now let's assume the APs we use are dual-radio 3x3 Wireless N Access points and our devices support the same setup with two radios. Since a radio can only send or receive data at a single time, this makes a connection half duplex. With two radios, we can now do a full duplex connection. Assume this is your equipment. You'll see if Wireless G could support this setup (N Does), your speeds clocking in around 42-43Mbps assuming some overhead in data in both directions, as each path does not have to wait for another path to speak. Now let's take this same scenario and apply the above. With full duplex connections, we put wireless connectivity to a greater test. Assume everything has been configured to be ideal and work under any situation. We've got the same 5TB file stored on a server, with the proper amount of connectivity to handle the load. So we fire up some downloads from this server. As we download, we'll see the first few devices pull 42Mbps. As other nearby APs start to transmit on the same channels, noise will start to set in. With a full duplex setup, we will start to see a greater amount of collisions occurring, especially as more traffic is put into the air. Speeds will fall sharper, and the risk of stalls further increases.
So, you see why Verizon nor anyone else cannot support a WIreless connection? There's way too much to talk about that can go wrong with a Wireless connection and I don't believe Verizon is in the business of performing site surveys such as those you'd get for an Enterprise environment for the entire country. Verizon Wireless? Well, they pay for their own spectrum, and narturally, they perform a survey of the country in order to ensure they can get coverage to devices but assume that Wireless will have it's downfalls. With Ethernet connectivity, the only factors you have are the gear on either end, and a Cable. The gear on both ends is almost never a factor unless one is defective. From there, you've basically got the cable. A proper cable will be rated to handle 1Gbps (As all Ethernet cables are Twisted Pair, thus noise isn't really an issue), for example. As long as the Cable, and the two pieces of equipment are Gigabit, you will get what you see: 1Gbps all day, all night. If some noise issues do settle in, just add shielding to the cabling or increase the twist amounts. Problem solved. This is how you can see FiOS, being a connection delivered in via a Physical Fiber cable rather than something that comes from the air.
Wireless is a convenience, not a guarantee. If you want speed, you can go for an N router. Yes, it will give higher speeds but you're at the mercy of the radio waves. I pre-wired my home with Ethernet (Two ports per room) knowing the problems WIreless has, and even though 24 other networks show up around me, I get 950Mbps down and up at the same time regardless of what goes on in the air.
