USB and PCI Wireless is very slow

[Gravey]

So I have a wireless network in my house

I didn't transfer files between my pcs so I didn't notice it was maxing at about 10mbits a sec.

It is not my router as my laptops are downloading at 25mbits a sec

My desktop pc's are maxing out at 10mbits

Box#1 E6600@3ghz.. win 7x64
Airlink aww3028 usb adapter

Box#2 Athlon 64 3200+ winxp x64
Airlink pci 54g card


I am going to be clear... Please do not post "Well its a crappy airlink"

I understand it is a low end network device, but wanted to know if it was anything in the settings?

Yes usb is at 2.0...

 

[Electrofreak]

Maximum data throughput of Wireless G is actually about 26 Mbps. Yes, over half of the data that's sent over Wireless G is essentially network overhead, encryption, error correction, TCP connection, etc.

Thus, while the connection itself is 54 Mbps, the actual amount of data you see pushed through under optimal real-world conditions is less than half of that, 20 or so is normal. If you've got computers with weaker signal strength, it wouldn't be surprising at all to see that rapidly drop down to 10 or so.

I'm surprised your laptops are getting 25 Mbps data throughput to be honest. That's really good.

I just posted on another thread on this forum about USB wireless adapters and how horrible they are. Another fellow mentioned that he's heard that USB adapter reliability can increase when they're connected to a powered USB hub instead of a bus-powered hub or port. I'd heard this too but I'd never had a chance to test it out.

http://www.hardforum.com/showpost.php?p=1035421929&postcount=12

...But one very interesting comment in the newegg reviews for the Zonet suggests that a fundamental problem with USB adapters may be a lack of power: "Works great as long as it's plugged into a powered hub and not directly into the computer's USB port. It would drop connection repeatedly when plugged directly into the computer. It appears that the power from the direct connection wasn't enough to keep it going."...

I'd heard that, but I've never actually gotten around to hooking one up to a powered hub. I've just done a bit of reading, and it appears that USB devices connected to a USB 2.0 port will pull 100 mA by default, but if more power is needed, the device driver can request up to 500 mA to be sent to the port. A powered USB hub, on the other hand, provides the maximum 500 mA to each port by default. (In contrast, an unpowered hub will not provide more than 100 mA per port, even if no other devices are plugged in to other ports).

So, likely the USB adapters have drivers that are not requesting the additional power when it is needed, or perhaps are simply not capable of accurately predicting when more power will be needed before the device encounters a situation where it does not have sufficient power and thus drops the connection. Either way, it would make sense that using a powered hub would resolve the problem. I'll have to take a closer look at it some time when I've got the spare time.