Receive new posts as email.
This site operates as an independent editorial operation. Advertising, sponsorships, and other non-editorial materials represent the opinions and messages of their respective origins, and not of the site operator or JiWire, Inc.
Entire site and all contents except otherwise noted © Copyright 2001-2006 by Glenn Fleishman. Some images ©2006 Jupiterimages Corporation. All rights reserved. Please contact us for reprint rights. Linking is, of course, free and encouraged.
Interoperability is not forward compatibility nor a promise of such: The two key chipmakers said they’ve conducted interoperability testing between their Draft N chips, and were able to achieve speeds over 100 Mbps using mandatory Draft 1.0 elements. The release skirts critical issues: First, it doesn’t claim the chips tested are the ones in manufacturers’ devices, although this is probable. Second, it doesn’t identify whether firmware tested is rolled out to their OEMs and then the OEMs’ end users. Third, this doesn’t guarantee upgradability via firmware to the final version of 802.11n.
I don’t deny the validity of the testing. It’s perfectly reasonable, and a nice assurance. But it’s not proven out in real-world testing of current Draft N gear and it’s a smokescreen over the real issue that this equipment is being released well in advance of accord on the standard to which it affirms to comply. Draft N is not a standard. It’s a working draft.
Wi-Fi Planet’s Eric Griffith adds some insight, including a note that Azimuth Systems, a wireless testing equipment vendor, has updated its MIMO test to offer Draft N testing.
US Robotics says Draft N isn’t fully baked: They’re right, of course, and alongside Belkin are in the minority of consumer Wi-Fi gateway makers to take that position. US Robotics will wait until at least fourth quarter to release products. “Products developed with pre-standards technology, or even rushed to market with standards based implementations, put customers at risk of incompatibility,” they said.
Have they been reading this blog—or just being sensible? You decide.
Tim Higgins at Tom’s Networking suggests Draft N is awesome! Higgins writes that it’s the duty of every consumer to go and buy all Draft N devices whether they are called “draft 11n, draft n, nFiniti, TopDog, XSPAN, Intensi-fi, nAncy or nUcular.” He exhorts companies to ship without testing or worrying. “This is the new millennium and competition is everywhere. If you don’t get your product out there first, then you might as well just give up…you loser!”
Becky Waring writes that the first batch of Draft N routers lack interoperability, speed, range: These are the kinds of products that should stay in the R&D department until, you know, the standard on which they are based is finalized and the products have had a full QA (quality assurance) shakedown. Waring writes that Linksys’s SRX400, using Airgo’s third-generation MIMO chips, had better throughput at various ranges than all Draft N devices except NetGear’s 10/100 Mbps line-up at the longest distance she tested. Devices using chips from different vendors couldn’t achieve their highest same-vendor speeds yet, either.
Airgo is trying to promote the notion that the Task Group N vote yesterday was important: It wasn’t. Airgo is reasonably expressing the strong doubts that many equipment testers and pundits (yours included) have voiced about interoperability and interference between the current draft of what will become 802.11n, and the Draft 1.0 currently under consideration. Draft 1.0 was not accepted by a 75 percent supermajority for passing on to the next stages on the road to ratification yesterday—fewer than 50 percent voted that way—but it’s hardly a surprise. I don’t know how many notes were returned based on the circulated draft, but there was no way, according to my sources, that Draft 1.0 would have moved on to the final stages.
This is pretty normal. Almost all IEEE standards have many drafts, and that’s a good thing. It’s part of the process of accommodating different technical viewpoints and producing something that should be able to be implemented in software, firmware, and hardware. The vote yesterday simply affirms that work is ongoing.
Based on what I have heard, I would not expect a draft to be accepted until the September meeting at earliest. It should be two full cycles (four months) from that acceptance to ratification, if that quickly. But it means that an essentially close-to-final version should be settled by summer, a final version by fall, at which point the Wi-Fi Alliance might start building a certification process for several months after that to ensure interoperability and standards compliance.
Airgo has not yet started producing 802.11n-like chips, while three of its major competitors have. I and others thing the competitors have made a huge mistake, as have their manufacturing partners for reasons cited all over this blog, including no guaranteed upgradability to the final 802.11n standard. However, Airgo has every motivation in the world to trumpet any failure to advance. I don’t believe that they were philosophically opposed to early Draft N chips; I think it’s a business decision to reduce costs and possibly an execution decision based on their ability to get chips to market. Regardless of the true causes of their position, they have the moral high ground, as does Belkin, which has chosen to not ship until the draft is further along.
PC World and ABI Research say Draft N gear halfbaked: Yardena Arar writes a strongly worded editorial at PC World against Draft N equipment. The standard isn’t finished, interoperability is unknown, there’s no upgrade guarantee for early adopters, and the 10/100 Mbps Ethernet switches are a bottleneck for top speeds. She also notes the crux: “…a draft standard is a work in progress. That means, for starters, that some technical issues remain unresolved—issues that may seriously impact the way a product works.”
ABI Research also says that devices reaching the market in 2006 will have “wide variability between them” and warns that true interoperability is “wishful thinking.”
Airgo convinced laptop maker to use its chips: The Airgo third-generation chips have a finite lifetime, as they are not promised or expected to be upgradable to 802.11n. Airgo will have its own 802.11n chips within a few months. I can understand wanting greater range, but Asus laptop buyers will need to purchase 3rd-generation-based routers from Buffalo or NetGear to achieve higher speeds.
Belkin conservatively won’t ship gear until June 15: I say, bravo! They’re waiting until the IEEE 802.11n standard has gelled to a state they are comfortable giving to consumers. Since their competitors shipping gear already won’t commit to a full hardware replacement for full final 802.11n compatibility, Belkin has the moral high ground. They’re offering a router ($150), PCI card ($120), and PC Card ($100).