April 24, 2006 6:50 PM PDT
Wi-Fi consumers cautioned to wait on new gear
- Related Stories
-
Faster Wi-Fi standard gets draft approval
January 19, 2006 -
Intel leads Wi-Fi chipmakers in new forum
October 7, 2005 -
New Wi-Fi standard takes the slow road
May 19, 2005
On Monday, the Farpoint Group and the technology trade publication eWeek released results of tests conducted on new products using draft versions of the 802.11n standard. While eWeek's assessment is not nearly as negative as the analysis of the testing from the Farpoint Group, both groups said they felt it was still too soon for consumers to buy products using 802.11n.
"I've always been a harsh critic of selling equipment that is compliant with a draft," said Craig Mathias, an analyst with Farpoint Group. "But besides that I was reasonably underwhelmed in terms of the throughput and range of the draft compliant products."
eWeek was also critical of the new products.
"It is not advisable to invest in these products lock, stock and barrel," eWeek said in its article. "Enterprise-grade WLAN manufacturers continue to wait for the standard to fully bake, and enterprise customers should do the same."
The new 802.11n standard, which is expected to be finalized later this year, will allow notebook users to connect to wireless access points at much faster speeds than currently available with 802.11g technology. 802.11n will use a technology called MIMO (multiple-in, multiple-out) , which should improve the range and throughput of 802.11n products so that it can be used as a replacement for Ethernet cabling in an office and as a way to transmit video around a house without interrupted playback.
In January, the IEEE approved a draft version of 802.11n, after much controversy and infighting among chipmakers. In the last few months, several products have emerged on the market claiming to comply with the 802.11n draft.
Problems with the technology
But now that products are out in the market, groups testing draft 802.11n are finding that the technology has some problems. The Farpoint Group compared the performance and interoperability of Buffalo Technology's AirStation Nfiniti router and client, which use Broadcom's draft 802.11n Intensi-fi chipset, and both versions of Netgear's RangeMax Next client and routers, which use draft 802.11n chips from Broadcom and Marvel, with Linksys' Wireless G and SRX400 equipment.
What the Farpoint Group found during the testing was that the Linksys SRX400, which uses Airgo's third-generation MIMO technology that isn't compliant with the draft version of 802.11n, offered higher throughput at longer distances than all three of the other products tested, which used draft N technology.
The report also indicated that the "draft compliant" products did not connect at any faster speed or across any greater distance than existing 802.11g products, which typically transmit data between 20 and 24 mbps.
eWeek tested Linksys' new WRT300N Wireless-N Broadband Router and the WPC300N Wireless-N Note-book Adapter, which both use Draft 802.11n chip technology from Broadcom. The article said Linksys' draft 802.11n gear was the fastest wireless equipment at short distances the magazine had tested to date, "besting even a pair of products based on Airgo's Gen 3 True MIMO chipset."
But when it came to long distances, a key reason for developing the 802.11n standard, eWeek, like the Farpoint Group, found gear based on the draft standard fell short. eWeek said that performance at 50 feet lagged considerably when compared with the products using Airgo chips.
"I wouldn?t read too much into these early tests," said Bill Bunch, director of marketing for wireless LAN for Broadcom. "The testing we have done has gotten results more like the eWeek test."
Bunch said the true value of Broadcom's draft 802.11n technology is that it interoperates with equipment from other draft N suppliers. But the Farpoint Group found in its testing that this was not the case. Mathias said that he was unable to get equipment from Netgear and Buffalo Technology to talk to each other. What's more, he wasn't even able to get the two versions of the Netgear products to work together.
Bunch said the Farpoint Group's results are flawed.
"I can guarantee you that it's a problem with the test," he said. "I have tested this myself at home, and it works. I look at these results and can see right away something was wrong with this test."
Mathias said that companies such as Broadcom are overhyping the capabilities of their products.
"These products have not been verified by the Wi-Fi Alliance," he said. "It's all marketing right now, and it's marketing out of control."
See more CNET content tagged:
IEEE 802.11n, Airgo Networks, Broadcom Corp., throughput, MIMO
10 comments
Join the conversation! Add your comment
So, here's my question: Does N take care of this? It's supposed to be optimized for multipath, high interference environments, right? Did the testers test even test that? IOW: G to N in a clean environment may not gain much, but how about G to N in a spectrally dense environment? I didn't see in the article any mention of that.
mark d.
Given that high speed internet is on the order of 1.5MB/s, it is clearly going to be the weakest link of any home network for speed. Unless you share files and other resources between computers, the biggest concern you should have with wireless is range and reception. Wireless G will certainly xmit fast enough, but too often I've encountered problems with range and certainly with walls and ceilings killing the signal faster than anything else. This article makes me think that wireless N isn't ready enough when it comes to range. Anybody willing to dispute their test results on wireless N (not Pre-N), I'd like to hear it. My experience is that a wall will kill about 30% of wireless G signal. How does wireless N fare?
tweak your netowrk for only the "N" standard and nobody else can hijack your connection!
wonderful!
technology as a whole is suffering from the age old adage of "To
May Chefs Spoil the Broth" syndrome. No central control, just a
forum that likes to think it's in control.
1) Standards take to long to approve and decide on by the
current Wifi panel.
2) Companies like Linksys and DLink are exploiting 802.11(x) as
well as Bluetooth Technologies to extremes that weren't meant
to be.
IE. The art of using all 11 channels to send data over a Wifi
network not only hinders the affects of your neighbors wireless;
it also puts the original wireless device at a risk of security
exploits. This in turn has an end result of leaving the consumer
at risk to future software exploits.
3) Average consumers (non-geeks) don't know what a RADIUS
server is, so don't put the option in the box. Making a sub-class
of wireless devices geared towards everyday consumers that
leave out these very option like Port Forwarding, DMZ Zone, and
Port Triggering.
Login, set wireless security, and leave! This not only slims down
the software needed to run the wireless device, it creates a
barrier of security in the software. Security that trickles down to
the end user.
4) There is a serious need for Channel Identification Software
built into the wireless devices. This software would ensure the
best channel to choose for your wireless network by scanning all
local 802.11x channels within' 1000 ft. in every direction. Noting
the channels that are in use buy your direct neighbors that WILL
affect your wireless network if you chose to be on the same
channel as your neighbors. The fore technology would ensure
that this would never happen.
5) In the case that someone want's to be on the same channel as
your wireless network, you should be informed. Because this
kind of activity is not normal unless your trying to sniff your
neighbors packets.
In that case that a neighbor is looking for trouble and I want to
know about it in the form of a pop-up window that tells me
there is somebody using the same channel on my Wifi network,
would you like to change the channel.
The above is just suggestion and should be taken with a "grain
of salt".
~Justin
Tech01.net
I couldn't be happier with my choice.
Any help would be appreciated.
jc