I've got a simple theory for why we are seeing so many, sometimes overlapping wireless consortiums with sometimes overlapping memberships.
This week Amimon announced a group of top consumer companies backing its 5GHz technology for sending uncompressed video across the digital home. It competes directly with the WirelessHD group SiBeam helped launch around 60 GHz technology for in-room links. Last week Sony debuted a group backing its TransferJet for short range synchs.
I am speculating the startups in search of credibility and design wins that can help them bootstrap into the mainstream offer a deal that's hard to refuse: Contribute some time from a few engineers and you get a chance to look at and influence the technology to make sure it works well on your systems, and maybe even a lower cost for adoption if you decide to use it.
There's very little to lose on such a deal, especially for the giant consumer electronics companies who have been jumping into multiple groups. The lesson: When the consortia virus strikes, immunize yourself with a dose of reality. These groups sound important, but they may not be good predictors of who will use what in a consumer wireless market that is still a wide open field where many flowers will blossom.
Thursday, July 24, 2008
I've got a simple theory for why we are seeing so many, sometimes overlapping wireless consortiums with sometimes overlapping memberships.
Friday, July 18, 2008
Just a quick shout out to consultant Eric Bogatin who runs the "Be the Signal" Web site and has recently started blogging on the topic of technical tips for signal integrity engineers. Eric is one of my go-to guys when the questions get too sticky for my limited on-the-job engineering training to handle, so I am sure practitioners and all those folks over at the SI-List will find this blog useful.
Posted by Rick Merritt at 2:33 PM
Thursday, July 17, 2008
Talk about being blindsided. I thought Sony's announcement at CES about its ultrawideband-based TransferJet technology was an anomaly, something that would die a quiet death. Surprise, a slew of consumer companies are saying this morning they will participate in an effort to standardize the technology.
Well at a claimed 560 mMbits/s (PHY rate?) and an easy touch-to-associate user model, they have a good start. And now with giants such as Panasonic, Samsung, Toshiba and a handful of camera companies behind them, they have some clout.
As PR guru Susan Cain who forwarded the TransferJet release to me said, "Just what we need another wireless consortium!" Yeah, we have wireless USB, WirelessHD (60 GHz), WiFi Alliance, WiMedia, more out there now and more yet to come I know for a fact. Gotta find out more…
Posted by Rick Merritt at 9:57 AM
Wednesday, July 16, 2008
Market watcher ABI Research reported today that WiQuest, Alereon and Pulse~Link came out tops among ultrawideband suppliers when measured on a range of factors. All three offer physical layer, RF transceiver and media access controller silicon and the three are tops in numbers of announced design wins.
WiQuest leads the pack largely because it claims it has won sockets at Belkin, Dell, DLink, Lenovo and Toshiba, said ABI analyst Doug McEuen.
McEuen admits the outlook for UWB is not as rosy as when many of the startups were cropping up a few years back, in part because the goal of 480 Mbit/second devices is still elusive. "But its still quite a viable market," he added.
Posted by Rick Merritt at 4:59 PM
Tuesday, July 08, 2008
The folks over at the Consumer Electronics Association are full of Omega-3s these days. They want to define a universal interconnect for media players by running copy protected High Definition Multimedia Interface (HDMI) signals over the CEA's existing Portable Digital Media Interface (PDMI, aka CEA-2017).
That's a great and ambitious idea that would really serve consumer interests. Just imagine being able to take any media player and link it to any car or home docking station or other peripheral. Fabulous.
But, um, you know there's this company called Apple that makes this thing called an iPod. They have this tendency to sort of do their own thing. And sense the typical teenage consumer does not know the brand name for a second media player device, they can pretty much get away with these shenanigans, forcing the industry to follow their ad hoc-ish lead.
By the way, for those other companies--unknown to teenagers--also making media players there's this plug called USB. It's pretty popular. Maybe not quite the thing for a car in its current state, but these USB folks are pretty active in rolling out variantions and upgrades.
So my dear friends over at the CEA, I look forward to watching your ambitious efforts to swim upstream in digital media. If anyone has any observations or insights on this effort, post a comment or drop a note at email@example.com.
Posted by Rick Merritt at 8:04 AM
Tuesday, July 01, 2008
I had a chance to catch up recently with Lou Lenzi (pictured), a veteran consumer electronics designer who currently heads up product management for the accessories business at Audiovox. He talked about several interconnects he is using in his latest consumer products including HD-PLC, WiFi and CEA-909.
Audiovox is rolling out a $399 wireless HDMI connector for flat-screen TVs and projectors using HD-PLC, the powerline technology developed by Panasonic. Lenzi's team surveyed many options including competing powerline approaches, ultrawideband technology from Pulse~Link and Tzero and Amimon's WiFi derivative.
Powerline was the best bet for adding no new wires around the wall-mounted displays and projectors, Lenzi's team concluded. HD-PLC gave his systems a net 90 Mbits/s, less than the advertised 190 Mbits/s but plenty enough to support a 1080-progressive display.
"We looked at them all, and were very impressed by their 128-bit AES encryption which will play well with content owners," said Lenzi. "Their technology spans multiple circuit breakers, and they have an easy pairing method," he added.
The only downside for powerline in general is cost. "I'd like to see the modules get down to $49 to really take off. Right now they are at $99," he said.
Lenzi's group also makes a variety of universal remotes for which WiFi is becoming increasingly important. Startups ZeroG and G2 are doing a good job pushing down component prices to get WiFi into more systems such as remotes, he said.
Finally, Lenzi's big new product for the fall is an indoor flat antenna for over-the-air digital TV that can plug into any analog or HDTV. He estimates some 20 million U.S. homes use over-the-air as their primary TV source. Another 14 million use it for at least one TV in the home.
"That’s a big opportunity for us," with the cut-off of analog signals coming in February 2009, he said.
The antenna can deliver a crisper picture than satellite TV systems which typically compress signals down to a tight 7 Mbits/s. So, even some HDTV users may buy the antenna box to get the best reception for special events such as the Super Bowl.
"If you do a side-by-side comparison of high def from over-the-air versus cable or satellite you would be surprised at the difference," said Lenzi.
Audiovox has patented the layout for antenna plates it uses in its mini-pizza box unit. Lenzi said the new CEA-909 interface that lets users "steer" antennas by electrically exciting different combinations of the plates will be a key feature for an upcoming crop of DTV converter boxes. The interface will eventually appear on set-top boxes and DTVs, too.
I first met Lou perhaps more than a decade ago when he was designing PC-TV systems for Thomson. Times and technologies have changed quite a bit since then, but Lou remains an optimistic and innovative engineer at heart.
Posted by Rick Merritt at 10:09 AM
Monday, June 30, 2008
Somewhere along the line last week this little blog crossed 100,000 hits to date. I know that ain't much in these days of Web 2.0 mass market mania where everybody is angling for millions per week or day, but to this old-school print trade journalist who is still learning how to be a Webbie, it is amazing to think something done in my spare (!!) time has been viewed 100,000 times.
This milestone is especially gratifying given the fact I haven't been able to post daily for some time because I have been extra busy with other work stuff—and life stuff. So thanks, keep coming back and don't hesitate to provide feedback in a comment here or an email to firstname.lastname@example.org
Posted by Rick Merritt at 8:11 AM
Monday, June 23, 2008
A new Frost & Sullivan report suggesting WiMax is in trouble is getting some attention today, latching on to a broader debate my colleague Jack Shandle explored with his recent story on the outlook for harmonizing WiMax and LTE.
There’s been a fair bit of tit-for-tat between WiMax and LTE lately. A group of LTE assembled to drive interoperability testing today. Recently both LTE and WiMax backers created separate efforts to come to grips with patent issues.
One of the big proponents of WiMax has been Intel which has dangled the possibility of rolling a combo WiFi/WiMax chip set into future Centrino notebooks to kick start this market as it did 802.11. It will be well worth watching for any developments on this front at the big Intel event in August. Meanwhile, the debates rage on.
Posted by Rick Merritt at 1:49 PM
Thursday, June 19, 2008
A brief trip to IEEE Secon raised more questions than answers for me about the status and outlook for wireless sensor networks.
A workshop on WiFi meshes suggested a few small companies are seeing some market traction with the technology despite the fact the 802.11s standard has gotten bogged down and there appears to be a lot of fundamental research still being worked out. Eric Brewer made it clear the technology has both huge potential and tremendous challenges in what is its potentially biggest market—getting the developing world on the Net.
A fire hose of a keynote from David Culler (pictured) raised other questions about work in the lower power, lower bandwidth 802.15.4 world. Here too it seems there are still unresolved standards and software issues for the industry, particularly around how meshes and routing are being defined. Culler’s efforts hold promise of a way forward, despite questions about what are the driving apps for sensor nets.
I suspect the apps are diverse ranging from the factory floor to health care at home, as Intel has shown. BTW, whatever happened to that national sensor net programs folks around the U.S. Homeland Security Department were talking about a few years ago?
Posted by Rick Merritt at 10:55 AM
Vendors cranked the volume on Infiniband at the International Supercomputing conference in Dresden this week with news on products from QLogic, Voltaire and a new roadmap for the IBTA. Likewise IB proponents were quick to jump on the bandwagon--along with AMD--for their part in the success of IBM’s Roadrunner cracking the petaflops barrier.
But an editor of the Top 500 list was quick to point out Gbit Ethernet still dominates the world’s biggest systems, appearing in 285 of the supers compared to just 120 for IB. The IB installations tended to be high-end academic machines while GbE was more often used in commercial systems, he added.
And as for AMD, despite its high profile appearance in Roadrunner, its presence has diminished in the latest Top 500, a casualty of its missteps and Intel’s success with quad-core chips.
Posted by Rick Merritt at 10:24 AM
Thursday, June 12, 2008
I am headed to IEEE Secon next week, a technical ground zero for work on wireless sensor networks. I see participation in the event from a wide range of universities as well as top tier tech companies including Alca-Lu, Cisco, Intel, Microsoft and Nokia.
From anyone engaged in the event or the sector, I’d love to hear what are you consider the key technology and market struggles as I get myself oriented to cover this event. Drop a comment here or at email@example.com.
At an Intel Research event yesterday I heard a lot about wireless sensor nets for use in elder care, an emerging market Intel is trying to enable with its Shimmer sensors that currently use 802.15.4 and Bluetooth.
I also found out Intel is about to hand off to OEMs technology to create WiFi backhaul systems that can deliver 6 Mbits/s over 30 kilometers without using fancy high gain antennas or pushing the limits on regulatory power regimes. If they use better antennas and push power limits, they can send 4 Mbits across 100 km, said David Taylor (above right) who helped refine the software. The Intel technology involves software for a modified WiFi MAC protocol that uses slotted TDMA, enabling products that could be of use bringing the Web to the developing world.
Taylor was quick to point out the so-called Rural Connectivity Platform is a point-to-point link only and won’t compete with WiMax, Intel’s current wireless hobby horse. My view: between what WiFi and LTE will do, Intel would be better served to ease up on its whole WiMax religion—but that’s fodder for another story.
Posted by Rick Merritt at 10:28 AM
Tuesday, June 10, 2008
I look forward to popping into the annual PCI SIG meeting in San Jose tomorrow and am already hearing significant news about the release of the IOV specs. It ain’t the full PCI Express for the embedded world that ASI hoped to be, but there is nevertheless real momentum to use the standard to further drive PCI Express into comms and embedded gear, so look out RapidIO.
If there are any other importat wrinkles in the Express world I need to know about—issues with Express 3.0 or whatever—now’s the time to chime in as I get ready to dive into this pool.
Posted by Rick Merritt at 6:29 AM
Tuesday, June 03, 2008
Thanks to Patrick Mannion of TechOnline for tipping me off this morning to the conflict over 60 GHz. I had no idea the 802.15.3c folks and the 802.11 VHT crowd were about to have a big powwow to figure out how to parse out separate standards in this spectrum peacefully.
Craig Mathias says there is plenty of room for multiple standards and product approaches in this promising area, so bring on the competition. I’d like to hear other opinions about multi-Gbit wireless nets so please chime in here or at firstname.lastname@example.org.
What role do you think WiFi could play at 60 GHz, especially given the market lead SiBeam has with its .3c-like approach? Is there a real opportunity for the VHTL6 concept of an aggregate Gbit WiFi net at less than 6 MHz? And what does all this hubbub mean for Wireless USB and other ultrawideband wannabees?
And while you’re at it, what do you think of an Obama-Clinton ticket in 2008?
Posted by Rick Merritt at 10:06 PM
Monday, June 02, 2008
It’s a reasonable speculation from where I sit in the blogosphere. In a recent interview with CEO Ahbi Talwalkar (left) I found out the company lacks standard Ethernet chips to support its nascent drive into networking, but it does has an investment in the startup that was early to the long-coming 10G party.
It’s clear a good 10G capability would be key for the company that wants to live sat the space where storage and networking meet. However, there are plenty of available startups out there, LSI has some custom capabilities that could spawn products and there is always the question of when the timing is right for the 10G ramp to really start.
I’m thinking pretty darn soon. But I’d love to hear any good insights you may have in a comment or at email@example.com
Posted by Rick Merritt at 3:37 PM
Startup Ozmo debuts today with a novel idea for scaling back Wi-Fi chips for personal-area net connections in mice, keyboards, headsets and more. Intel will enable these and other kinds of WiFi PANs under its Cliffside project as I noted below.
I talked with Todd Antes, vice president of marketing for Atheros recently about a low cost- .11n router design from the company. He said Atheros demonstrated a Cliffside like capability at CES last January, though he would not comment on whether the company has any chip in the works similar to those from Ozmo.
“Philosophically we believe in [WiFi PANs],” he said. “Notebooks are getting smaller, so it makes a lot of sense to let one radio accomplish more tasks,” he added.
Posted by Rick Merritt at 3:28 PM
Wednesday, May 28, 2008
I am playing catch up with Cliffside, an Intel project to put WiFi in the personal area network disclosed at IDF Shanghai. Turns out this work and another effort (stay tuned for a Monday story) has a lot of people excited about WiFi someday becoming the one radio you need to handle everything from mice to linking to the Net.
One of my fave wireless analysts, Craig Mathias, claims the time has come for WiFi PANs. "In a couple years a WiFi PAN business could be very successful," Mathias told me on the phone today. It won't kill Bluetooth--nothing dies in high tech--but it could stunt its long term growth worse than a bad cigarette habit.
Intel claims its technology can connect up to eight WiFi devices to a notebook on a PAN while the computer is on a WiFi LAN. The technology could improve the quality of media streamed between a computer and a TV or other display because it eliminates the latency of going through an access point.
The Cliffside program uses a modified Intel WiFi chip with additional buffers to rapidly switch between PAN and LAN modes and is expected to ship within 12 months, said Gary Martz, a marketing manager for the program. He said use will probably be limited to consumer notebooks for the first year or so.
"Corporate IT managers will probably have a heart attack, perceiving this as enabling notebooks to be a bunch of rouge access points that impact performance of their WiFi networks," Martz said.
Mathias was animated in his enthusiasm for WiFi. "WiFi's success is unquestionable. It will be in everything. Twenty years from now people will still be talking about WiFi."
Good, we tech reporters need the job security ;-)
Posted by Rick Merritt at 5:18 PM
Monday, May 26, 2008
I haven't had my eye on the ball of the high-speed Ethernet standards effort over at the IEEE 802.3ba for awhile, so I talked to the chair of the committee, John D'Ambrosia late last week for an update. This is big complex stuff about the future of Ethernet, the data center and converged telecoms at mind-bending 40 and 100G speeds.
John reports all goes well with technical progress at the last meeting, as many as 200 expected at the next meeting and appropriate debates on the floor. I haven't heard much from the committee at large about this effort, so if you have some observations or opinions on this big leap forward post a comment or drop me a line at firstname.lastname@example.org.
While this standard is gestating, John has a new puppy at home (pictured while snoozing) that is also growing fast. John says it's possible the .3ba work could be in an early draft stage by March and a late draft in November. By that time little Buster D'Ambrosia will be chomping down dog food at a 40 or maybe 100G rate.
Posted by Rick Merritt at 10:32 PM
I spent a couple days researching GDDR5 for an EE Times story last week. From what I learned, here are a few quick and free from-the-hip predictions. Take them for what they are worth.
--AMD will be the first to use GDDR5 on a new graphics controller to be announced soon.
--Archrival Nvidia will wait as long as possible before moving to GDDR5, opting to use the more economical GDDR3 at 1.3 GHz+ as long as possible
--Intel will use GDDR5 on its much-anticipated Larrabee controller when it debuts probably in late 2009, but the part won’t make that big of a splash in the market in its first generation.
--Rambus will struggle to get traction with XDR in HDTVs and elsewhere, crossing its fingers for better luck in next-generation videogame consoles with a second or third generation technology.
Think I'm wrong? Share a prediction of your own here or at email@example.com. What to learn more? Check out this white paper.
Posted by Rick Merritt at 10:07 PM
Wednesday, May 14, 2008
You get interesting glimpses into where a company is at trolling the job boards. I saw a recent ad for Hua Wei on the SI-List for an engineer that indictaed where this company is on its climb from a copier of Cisco products to an innovator of modest capabilities in its own right.
The ad called for an SI designer who ideally has "experience in taping out multi-million gate CMOS ASICs in 130nm technology and below." They are willing to take someone with no more than a bachelor's degree and two-years industry experience to handle "evaluation and choice of 3rd party IO buffer and SerDes IP for ASIC design."
Clearly, they are not at the exalted level of Cisco's 90nm Quantum Flow Processor or even its FCoE work, but few ASIC designers on the globe could pull off a design like that. Nevertheless, the China's answer to Cisco is clearly on the rise, and I'd like to—and expect to--hear more from them.
Posted by Rick Merritt at 10:57 PM
Tuesday, May 13, 2008
In the relatively slow and steady pace of the Fibre Channel world, competition is starting to heat up around the transition to 8 Gbit/s products. Storage switching giant Brocade rolled out its portfolio of 8G switches and server cards today.
Emulex was quick to point out it has been there, done that with server cards already qualified by EMC. Indeed both Emulex and QLogic were early to the 8G party with cards they rolled last summer.
Despite a little bit of temperatures rising, I recall what Renato Recio of IBM told me during an interview for a recent story about Fibre Channel over Ethernet. For users that what high performance and need to move this year FC is the safe bet, but for 2009 the FCoE community should have its ducks al in a row to be a solid competitor.
Posted by Rick Merritt at 10:34 AM
Thursday, May 08, 2008
There's plenty of work to do at the IEEE 802.3az, but Intel's proposal for a low power idle version of Ethernet has already got the nod to be part of the group's spec for 100 Mbit and Gbit chips. The next big debate will be whether it or a Broadcom proposal for a so-called subset PHY wins out for 10 Gbit chips and for backplane Ethernet.
Next week's meeting is the final call for proposals, so if you have a hot idea for green Ethernet better find the meeting and come with Powerpoint in hand or forever hold your peace. I'd be happy to host a debate on the trade off between these two approaches if anyone wants to take one side or the other in comment posting here.
Posted by Rick Merritt at 11:17 PM
Wednesday, May 07, 2008
No big surprise that Conexant sold off its powerline networking group today. Linking the dots on its future, it's not a big shocker HomePNA specialist CopperGate Communications was the buyer.
The Multimedia over Coax spec is Conexant's current direction in home nets these days. A Conexant exec told me at the MoCA conference last fall that Conexant plans to ship discrete MoCA RF and baseband chips in the second half of 2008, then follow them up with chips that integrate MoCA support as a block on its silicon for DSL, MPEG and passive optical network chips for set-tops, routers and optical network terminals.
Conexant and Broadcom both joined the MoCA board recently, telegraphing plans for MoCA silicon. Conexant had already given up on earlier plans to make HomePlug AV chips, believing the technology will have difficulty keeping pace with the bandwidth needs of digital media.
For its part, CopperGate is very bullish on the ITU's G.hn standard as part of its road to the future. CopperGate just announced it has delivered five million chips for HomePNA over phone and coax lines to date. It bought Connexant's powerline group so it could add to its mix that technology which European carriers demand.
Looking to the future, CopperGate foresees the day when a completed G.hn could allow it to deliver a chip that handles all three wired media in a standard way. The company also plans to offer chips that support G.hn as well as the version 4.0 of HomePNA now in progress. One chip, all home net markets--that's the Holy Grail.
I wonder why Conexant didn't see that broader win, hang on to its power line group and prepare for the same single-chip future.
Posted by Rick Merritt at 5:59 PM
Monday, May 05, 2008
Tom Cox pointed out to me that I missed a milestone back in February when RMI Corp. (formerly Raza Microelectronics) announced a processor for wireless base stations using a native serial RapidIO interface. It was the first MIPS chip with the interface that has previously been used in PowerPCs and DSPs.
No big surprise that anything linking into the wireless base station might go RIO, given the traction the group has gotten in DSP farms there with Ericsson, TI and others. But will the MIPS win expand beyond RMI?
Cox, executive director of the RIO trade group, suggests the right dynamics are in place. Cost reduction, the need for better communications links and the rise of multicore processors will drive out FPGA-based bridges in use today between MIPS chips and RIO farms, he said.
The RIO group has gotten several new members recently including Nortel and Qualcomm. No word on exactly what these folks are up to yet. If you have some insights, post a comment or drop me a line at firstname.lastname@example.org
Posted by Rick Merritt at 5:07 PM
Thursday, May 01, 2008
Jockeying for position in the home network circa 2010 will be a major consumer sport with the HomeGrid Forum jumping into the fray. The players all will want to position themselves as the best solution for everyone from carriers to OEMs to consumers.
So far MoCA, HomePlug and HomePNA have been pretty quiet about their plans to leap toward the 400+ Mbits/s carriers are calling for in the next year or so. But everyone has heard that call including the HomeGrid folks who want to goose the G.hn effort to deliver that over coax as well as a reach goal of Gbit links when possible.
Writing an analysis of the latest news revealed a few interesting wrinkles. For example, the wired (MoCA, Powerline and HomePNA crew) represent just 19 percent of today's home nets which are dominated by WiFi and wired Ethernet. And most of HomePNA's installs are on coax, not phone lines.
To get one overarching standard will require a tussle between the big vendors who want a unified market and the small vendors with silicon skin in the game. Already Pulse~Link is suggesting G.hn should re-think its choice of OFDM (which it does not use), while HomePlug folk praise the group for the choice (because they made it, too).
Meanwhile the powerline folks at IEEE 1901 have yet to be able to muster support for a confirming vote on a HomePlug/Panasonic home net proposal. The vote was put off at meetings in October, December and March, but may come up in July. Not a good sign for consensus building in this community.
Posted by Rick Merritt at 12:03 PM
Tuesday, April 29, 2008
Could we be headed for one spec that delivers the best home networking over coax, phone or power lines? That's what they hope over at the HomeGrid Forum that launched yesterday with 11 members companies. They aim to accelerate work on the ITU-T G.hn standard-in-progress.
Turns out only a couple percent of the estimated 140 million home nets today use MoCA or HPNA and maybe 10 percent use some form of powerline, according to market gazers over at Parks Associates. It's mainly Wi-Fi and wired Ethernet at home today.
Still, it could be quite a boost for the industry to get behind one interoperable standard. I'm gathering opinions on this effort for a print wrap up story, so if you have something insightful to offer, drop a comment here or at email@example.com. Can this work? Will people get behind it? What will it take? Sound off!
Posted by Rick Merritt at 3:03 PM
That's what Nick Ilyadis, CTO of Broadcom's Ethernet networking group claims in a story I posted today.
It was noisy on his end at Interop and he may have been pumped up about all the 65nm products he was rolling in Vegas, but he seemed to feel with Broadcom's recent move to back a so-called Multi-Link Distribution proposal, the IEEE 802.3ba work on 40G Ethernet might reach consensus on a proposal at their next meeting.
If you have some insights into the issue at hand, post a comment here or drop me a line at firstname.lastname@example.org. I want to stay on top of this work and need a little of your citizen journalism reporting to help me out!
Posted by Rick Merritt at 3:01 PM
Tuesday, April 22, 2008
More 10G news flowed this week in the after glow of the Storage Networking World conference where FCoE had a big coming out party.
I was interested to see 10GBase-KR get a boost with Fujitsu's new 10GE switch chip, but analysts were quick to point out there are plenty of signal integrity problems ahead getting today's 3 Gbit backplanes to 10G.
Good also to see Aquantia joining SolarFlare with a 5.5W 10GBase-T transceiver. It will be interesting to see how long Broadcom and Marvell stand on the sidelines waiting for a market to develop here.
On the server side, Mellanox is promising new firmware so it can carry a wealth of protocols on top of its ConnectX cards that handle all sort of net and storage functions on Infiniband and 10GE. It marked the first time I had heard Mellanox use the increasingly widespread term from Cisco—Data Center Ethernet—referring to the lossless version of the net still going through the standards process.
Posted by Rick Merritt at 11:40 AM
Sunday, April 13, 2008
I got an interesting reality check on 10GBase-T from Ethernet giant Cisco Systems last week. The company's hot new data center spin-in, Nuova Systems said it will not use the standard for 10G over 100 meter copper in its latest switches.
"10GBase-T is great in terms of compatibility and simplicity, but the additional power, cost and latency means it is not really feasible for us and I don't think we will use it," said Dante Malagrino, director of product marketing at Nuova for a story that should be posted to EE Times shortly.
Instead Nuova/Cisco will use a hybrid solution based on a new copper cable terminated by SFP+ optical transceivers which it claims has lower power and latency than the 10GBase-T options. Malagrino pegged 10GBase-T at 2-3 microseconds in latency and 4-8W per link in power consumption. The new hybrid option, one of several alternative cables to emerge in the last year, will initially be limited to 1, 3 and 5 meter lengths but could shave 30 percent off the overall costs of an optical fibre link, he said.
SolarFlare Communications, one of three startups working on 10GBase-T transceivers, is announcing a single 65nm CMOS chip that transmits 10G over copper up to 100 meters while consuming 5.5W. But the chip is just back from the fab and has not yet passed testing.
Competitors Aquantia and Teranetics may follow suit with similar products before the end of the year. I've yet to hear anything from established players such as Broadcom and Marvell. (If you know something, drop me a line at email@example.com)
SolarFlare claims 10GBase-T switches, aggregation boxes and server cards are in the works, some of which will ship before the end of the year. But the Cisco comment makes me think the new crop of transceivers while major accomplishments in design still may not extend very far the reach of this emerging market.
Posted by Rick Merritt at 9:54 PM
Thursday, April 10, 2008
I've been amazed to see how rapidly the broad group of companies backing Fibre Channel over Ethernet has been moving, given how much there is to do to create this converged data center network. But after some further digging yesterday I see progress is not quite as fast as their marketing departments would have us believe.
The first FCoE products released this week are still pre-standard and many are less than elegant. The reality is the real product wave will hit sometime early next year, as I reported last night.
Still, the FCoE group has been gaining momentum in monthly meetings. A separate CEE Authors group is trying to crank up the pace on the (notoriously slow) IEEE standards efforts.
I give special kudos to Cisco for driving in the diamond lane on this initiative. Their Nexus 5000 shows not only fast time-to-market with an ASIC-laden design but an innovative business approach with its spin-in of start-up Nuova Systems. This tactic, first tried with Andiamo, helped Cisco attract outside data center expertise it needed with people such as Ed Bugnion, the former CTO of VMware who is now Nuova's chief technology officer.
A tip of the hat from the Interconnects blog for the nice work by the folks in Little Italy up on San Jose's Tasman drive.
Posted by Rick Merritt at 7:55 AM
Tuesday, April 08, 2008
My hat is off to the many people driving Fibre Channel over Ethernet. The breadth of support at the Storage Networking World event today was mind blowing considering the technology only hit the radar screen a few months ago. Several big companies are being fast and nimble on this one—particularly Cisco Systems which seems to be reprising its success with the Andiamo spin-in Nuova Systems.
But there's plenty here I do not yet understand. How will these folk roll with the punches when IEEE advanced Ethernet and T11 FCoE standards finally close? Can it all be handled in software?
Is it true as competitors say that Emulex and QLogic are doubling up with Ethernet, Fibre Channel (and glue logic) silicon all crowded on to their adapter cards? Sounds expensive.
Meanwhile, my 10GBase-T questions from last week remain unanswered. I understand two companies may be ready to talk about new and better transceivers, but I have heard precious little from giants Broadcom, Intel and Marvell on this issue.
Will there be a hangover after the party in Orlando?
Posted by Rick Merritt at 5:28 PM
Thursday, April 03, 2008
Holy acronymity, Batman! No wonder this interconnect world drives me whacky sometimes. What's really scary is I actually know what this stuff means...and I want to learn more!
I have briefings coming up Monday on the latest and greatest in Fibre Channel over Ethernet and 10 Gbit/s Ethernet over copper. If you have any perspective, hot tips, leaks or heads ups about what you or others are doing in these spaces, please drop me a line at firstname.lastname@example.org or post a comment here ASAP.
So if you have anything to say on these technologies, sling it now or plan to hold your peace for awhile.
BTW, sorry I have neglected this little plot of the blogosphere lately. I was on a two-week trip in Asia that ended with a quick tour to the MultiCore Expo and Ocean Tomo Spring auction. There's a lot on my plate these days—-check out the Intellectual Property Symposium--so I appreciate your bearing with me as I get to this little side interest of mine whenever I can.
Posted by Rick Merritt at 4:40 PM
Monday, March 31, 2008
The pace of growth for Infiniband is on the rise, but so far the interconnect is not gaining much traction in mainstream business applications. That's the conclusion from an updated report from market watchers at International Data Corp.
IDC now expects Infiniband sales on end devices will rise from about 500,000 adapters this year to more than a million in 2011 with revenues jumping from about $125 million to about $275 million. Sales of switch ports will increase from about 600,000 this year to more than one million by 2010 with revenues on that side rising from about $250 million to about $400 million. High performance computers (HPC), database systems running on clusters and financial systems "with HPC-like characteristics" continue to be the drivers, according to IDC.
Mellanox, the sole maker of Infiniband chips, notes that the shift to 20 Gbit/s links is moving quickly with 40 Gbit/s products on the near horizon for this year. "In our last quarter, 79 percent of Mellanox's business was for [20 Gbit/s] DDR products," said Thad Omura, vice president of product marketing at Mellanox.
The chip company naturally wants to drive Infiniband into more mainstream data centers. Omura pointed to three users employing the interconnect as a cost-saving way to consolidate multiple Gbit Ethernet links on to one IB cable, a kind of unified networking approach Sun said it will drive starting later this year probably with the rollout of 40 Gbit products.
Support for Infiniband on VMware 3.5 and for NFS over RDMA in Linux could help grease the way for more such deployments. But for the foreseeable future most users will choose Ethernet if they are cost sensitive rather than a technology upgrade to Infiniband.
Posted by Rick Merritt at 9:18 AM
Tuesday, March 25, 2008
I admit I was skeptical about near field communications. I ignored comments from Intel wireless USB exec Jeff Ravencraft who said contactless cards enable a wonderfully intuitive usage model. Colleagues at EE Times said NFC had an uphill battle. And I thumbed my nose at a fledgling trial in the Bay Area.
But my experience seeing in action the Octopus card in Hong Kong has modified my views. Run by the company of the same name, the service uses Sony's FeliCa NFC technology and is widely popular here. It even comes embedded in watches, stuffed toys and key chains as well as the popular cards that can all be purchased at the local subway station (see picture, right).
It may never take off in America, but I understand it is doing well in Japan, great in Hong Kong where I am presently visiting and likely to spread elsewhere based on this beachhead.
Posted by Rick Merritt at 6:22 PM
Thursday, March 20, 2008
Call it the jealous wife syndrome. The Ultrawideband community has been rumbling since the Bluetooth SIG announced it would create a version of its spec that works over both 802.11abg and UWB. Jack Shandle, editor of the Wireless Net Designline got the folks at Staccato to give the first technical analysis of their concerns.
Jack makes it clear part of the issue may be the competitive politics of the emerging UWB crowd fighting to retain as much as possible of what was going to be an exclusive application. The technical part of the issue, according to the Staccato paper, is that Bluetooth over Wi-Fi may create interference with IMT-2000 and WiMax traffic at distances up to 16 meters. The paper says the industry should conduct more tests and consider defining mitigation schemes or recommending use only of the 5GHz 802.11a spec.
Jack is seeking broader industry comment on the issue, and I'd like to hear what you have to say about it, too.
BTW, I am in the middle of a ten-day Asia trip, thus fewer posts than usual. But ping me if I am missing something big and I'll try to make time to post.
Posted by Rick Merritt at 3:54 AM
Thursday, March 13, 2008
Sam Lucero, a senior analyst for ABI Research released a study this week predicting home automation is finally poised to take off. Zigbee and Z-Wave products are gaining traction as de facto standards and lower cost approaches than yesterday's custom installations and retailers like Best Buy are getting on board with offerings from the likes of 4HomeMedia, iControl Networks, Portus, uControl and Xanboo, he states.
I haven't seen Lucero's report but I am remaining a show-me skeptic. I agree with Lucero who foresees service providers rolling out home automation and security systems as a value-added service. But I don't believe it will happen for a looong time. This year enabling the multi-room DVR is the big focus for many service providers, so automated lights and web cams from Comcast and Verizon are still a way off in my view.
Got some hard evidence to the contrary? Let me know with a posting here or drop me a line at email@example.com
Posted by Rick Merritt at 2:27 PM
Monday, March 10, 2008
The HomePlug crew will deliver a fleshed out proposal for a powerline standard by June, in time for a possible first vote on at a July meeting of the IEEE 1901 group. That was the promise from Oleg Logvinov, chief strategy officer of the HomePlug consortium in a recent chat.
HomePlug is responding to charges from competitor DS2 which said the 1901 effort has stalled and the HomePlug group has yet to articulate details of its proposal. HomePlug claimed victory in the standards battle back in October when its proposal was down-selected, but has been slow to deliver a detailed plan. Yet to come is any report from the 1901 group chairman, but that may appear this week on the group's Web site.
Posted by Rick Merritt at 8:58 AM
Friday, March 07, 2008
Keith Cowal, a PC marketing manager at Dolby Labs told me today he sees a rising tide of HDMI links going into notebooks. The reasons are two-fold.
Notebooks are incorporating Blu-Ray drives and its AACS technology requires content protected interfaces such as HDM. And OEMs see users plugging notebooks into their TVs and stereos (which already have HDMI) to play Internet and packaged movies and music on the big screens and speakers. Makes sense to me.
A few glitches ahead: To support Dolby's TrueHD audio, you need HDMI 1.3, which is only supported in the latest chip sets. Drivers don't often do a good enough job of identifying HDMI audio interconnects to Windows, and there are still a wide variety of content protection technologies implemented on the PC including Windows' Protected Media Path. But this should all shake out in the next year, Keith says.
Posted by Rick Merritt at 3:14 PM
Tuesday, March 04, 2008
One of the key points Cisco Systems makes about its 20 Gbit/s Quantum Flow Processor, officially launched today as secret sauce in the company's new edge router, is that it has plenty of headroom because it is designed for a 40 Gbit/s upgrade. Turns out the Interlaken interconnect Cisco defined with Cortina Systems is the key ingredient in that upgrade.
Cisco will spin a version of the QFP that replaces four SPI 4.2 ports with four Interlaken ports to get the 40 Gbit/s throughput. The QFP internal architecture is already plumbed for 40G, the company claims.
Perhaps this high profile role will help the Interlaken technology see broader industry use beyond the walls of Cisco. Tundra and Altera already support Interlaken, a transport independent protocol that some already run at up to 60 Gbit/s. Perhaps it's time for members of the Optical Internetworking Forum to consider making some version of Interlaken an approved standard along with its own SPI-S.
Posted by Rick Merritt at 10:55 AM
Saturday, March 01, 2008
Just got an email from Chano Gomez, vice president of technology and strategic partnerships at DS2, who says the IEEE 1901 powerline standard is gridlocked. You will recall from earlier reports, DS2 strongly opposes proposals from the HomePlug consortium to create standard that can use either groups' physical layer technology.
Chano says the standards effort is still in the early days of a process that could take another two years. He notes the HomePlug proposal only allows separate technologies to co-exist not truly interoperate. And he says the HomePlug folk have yet to define the Inter-PHY Protocol at the heart of their proposal.
"Everybody at IEEE P1901 is very disappointed with the lack of progress in the last months," Chano writes.
I have emails out to the 1901 chair and HomePlug reps. If you have a view, post a comment or drop a line at firstname.lastname@example.org
Posted by Rick Merritt at 4:14 PM
Monday, February 25, 2008
Thanks to Richard Mourn from Quantum Parametrics for providing some color about the recently announced VersaPHY specification from the 1394 Trade Association.
Mourn says the spec aims to help a single 1394 link typically used for high bandwidth streaming media apps branch out to cover less demanding apps that might range from speakers and security cameras to sensors.
"If a car seat has an LCD display in the head rest and the display is 1394, through VersaPHY the PHY in the LCD display [also] could service the seat belt sensor, the occupied sensor, the temperature sensor, etc. All of this would be done using the one 1394 cable into the seat," he writes.
The technique works by creating addressing labels so devices can plug into a link automatically. It requires no new hardware, however users need to agree to some standards. An effort is underway to set such standards for power management uses in cars, Mourn says.
Posted by Rick Merritt at 6:40 PM
Neterion rolls its X3100 series chips today at VMworld Europe, the first chips to support the NetQueue technology of VMware and the PCI SIG's I/O Virtualization. One analyst said the move could accelerate the slow ramp for 10G Ethernet.
But as reported today, competitors such as ServerEngines and NetXen are poised to leap to the IOV standard and Gen2 Express soon. And they have an edge in supporting storage capabilities such as iSCSI.
Beyond the battle of Ethernet and Infiniband controllers, the competition to define the optimal cabling for data centers and central offices is heating up. Startup Lightwire's release at the Optical Fibre Conference of a new CMOS optical transceiver with low power characteristics is one of many new active cable components to roll out in the past year.
The sheer variety of chip and cable options suggests there are still quite a lot of unknowns about how to best build and wire big back end facilities for 10G.
Posted by Rick Merritt at 2:58 PM
Wednesday, February 20, 2008
Yowza, Eric Broockman's from-the-hip analysis of Moto's cellular problems stings true.
"Ed Zander’s mistake was to believe that a simple reorg or two together with a few interviews and snappy slogans would be sufficient to fundamentally remake a non-competitive corporate culture," he writes in his blog.
He notes how Moto designers slavish copied the successful Razr rather than dare to innovate with follow on products, and recalls how Moto flubbed a chance to co-develop a music phone with Apple before Steve Jobs & Co. went on to define the iPhone.
"The Schaumburg guys and corporate types from the cornfields outside of Chicago knew better than Steve Jobs did. They didn’t get the vision. So, Apple went elsewhere. Ouch. What a lost opportunity," he quips.
There's no quick solution to the deep-seated culture problems at Moto that extended to the non-competitive wireless chip unit at now spun out Freescale, he claims. OK, here's my vote for Eric as the next CEO of the new Moto.
Posted by Rick Merritt at 6:35 PM
Tuesday, February 19, 2008
The recent Mobile World Congress was ripe with demos of high def video captured on a cellphone and sent to a TV, according to my boss Junko Yoshida, just back from the show. Junko reports on a 65nm HDMI core from MIPS Technologies to make such a link.
The MIPS core competes with an HDMI variant that Silicon Image rolled out at CES. Silicon Image used clever design rather than aggressive process technology to solve the problem.
Junko reports that NXP Semi has licensed the MIPS core. I have noticed other recent cellular chip sets using a full HDMI core. Yet to appear are any design wins for the new Silicon Image approach.
I wrote recently that USB 3.0 might replace HDMI, mbile and otherwise, but other bloggers disagreed. In the march to high def everything, the protected cellular link is perhaps the most bleeding edge of applications.
Posted by Rick Merritt at 8:51 PM
Thursday, February 14, 2008
I was humbled when Valley vet Andy Bechtolsheim sought me out after a Sun Microsystems press conference yesterday. I had just asked Sun's systems VP John Fowler a question about unified data center networking and he pledged to ship such products on Infiniband in 2008.
Bechtolsheim wanted to make sure I knew all the technical advantages of IB as a unified net (high bandwidth, lower latency, built in resilience) and the downside of the FCoE effort (standards not done being the biggie).
It was interesting to hear Bechtolsheim point out weakness in his former employer (Cisco) than could trip it up in the push toward unified Ethernet nets, especially given he had made more than a few million when the networking giant bought his Ethernet startup in the boom days.
Just another example that technologies can look very different depending on where you sit at the moment. I expect Ethernet will become the unified net and that it will be slooooow getting there. I also expect Sun will be finding interesting technology arbitrage positions as long as savvy CEOs like Jonathan Schwartz can keep smart folks like Bechtolsheim on board.
Posted by Rick Merritt at 12:28 PM
Wednesday, February 13, 2008
The vision of a converged Ethernet net in the data center continues to be a moving target, but chip makers are moving toward it. NetXen, ServerEngines and others are moving to 5 GHZ PCI Express and the Express I/O virtualization standard. Their efforts will help deliver better throughput for their 10G Ethernet chips.
But still out on the horizon is the Fibre Channel over Ethernet standard which reported progress yesterday. By this time next year I expect these companies and their counterparts will be announcing support for FCoE.
Meanwhile the software development efforts are “massive,” according to Kim Brown of ServerEngines. Case in point: the company’s chip was complete in March 2006 and its initial software for TOE and iSCSI shipped last month. Oi vay!
Posted by Rick Merritt at 5:27 AM
Monday, February 11, 2008
The bus to the ultrawidband nirvana of Gbit/s+ data rates is taking too long. So the Bluetooth SIG has opted to use the transport independent version of it protocol ride 802.11 as well.
That version of Bluetooth won’t even be available to members until mid-2009, which begs the question of why the group announced this now. My guess is they are trying to hold off interest in other technologies such as 60 GHz radios which had quite a coming out at CES for wired systems. Academics are already working hot and heavy on versions of 60 GHz technology suitable for mobile devices.
Another interesting wrinkle is that the group is only looking to support .11abg and .11n support is not even on the road map for now. That seems to conflict with the stated goal of getting quickly to higher bandwidth, but perhaps broad adoption is a bigger driver than the highest possible speed.
The news amounts to one more sock in the gut for UWB. The wireless USB version has been dogged by low data rates. Regulatory issues mean OEMs need to have different configurations for different markets. Meanwhile design wins have been more a trickle of experiments than the predicted flood.
Posted by Rick Merritt at 12:04 PM
At last week’s ISSCC, I heard engineers in medical electronics pitch for just about every form of wireless networking imaginable.
Medtronic said it’s happy with the 400 MHz MICS standard for its implants. Toumaz Technology said it needs a custom design to hit its ultra low cost and power targets. A UC researcher said he needs high bandwidth to capture brain signals and that is leading him to ultrawideband. Another academic said he needs to resolve underlying issues of capturing individual signals before he can even consider wireless.
This diversity does not bode well for an emerging IEEE effort to set standards for body area networks that hopes to embrace not only medical but consumer uses as well. The good news is there is plenty of room for technology development and adoption where medical meets wireless.
Posted by Rick Merritt at 8:15 AM
Sunday, February 10, 2008
Believe it or not, it’s time to get in a submission for the annual IEEE Hot Interconnects conference at Stanford which runs August 6-8 this year. The committee is looking for “state-of-the-art hardware and software architectures and implementations” in a broad range of areas including but not limited to on-chip, chip-to-chip and network interconnects for silicon, copper and optical transports.
If you got a candidate send it by March 1 to the link found at the conference Web site. I’d be pleased as punch if you sent an FYI on what you are up to my way as well at email@example.com.
Posted by Rick Merritt at 7:31 PM
Thursday, February 07, 2008
Clearly 60 GHz design has hit the radar screen. Three leading research organizations came to ISSCC talking about 60 GHz advances.
The most significant move came from Belgium’s IMEC which described a digital CMOS receiver antenna to limit high-path signal loss. The dual-antenna device uses a programmable phase shift and integrates a low-noise amp and down converter. IMEC’s next step is to use 45nm technology to craft a four-antenna device that integrates a phase-lock loop and ADC.
IMEC is inviting the industry to join its research efforts. Back in the U.S., Berkeley’s labs have already spawned what appears to be the early leader in the field, SiBeam. But the startup’s initial products are aimed at wired-only systems.
Graduate students from Berkeley and UCLA came to ISSCC showing low-power 60 GHz receivers dissipating as little as 24 mW and thus suitable for battery-driven devices. But the 60 GHz transmitters eat perhaps twice as much power, and both researchers are still in early stages of designing full transceiver to bring 60 GHz to mobile systems.
Posted by Rick Merritt at 2:11 PM
It’s an old trick in the semiconductor game, shrinking costs with silicon integration to compete taking products to mass market. Kudos to Atheros, SMSC and STMicro who gave examples in the last 24 hours of doing this with 802.11n, Bluetooth, USB 2.0 and WiMax.
Atheros presented papers yesterday at ISSCC on what it calls it’s the first 802.11n chip with baseband and radio on chip. It knows competit0ors are not far away with its part which should ship soon. Similarly it described its first Bluetooth 2.1 chip, and integrated part slightly smaller than the one from competitors such as Broadcom.
SMSC today shrinks its USB 2.0 transceiver to new size and power lows, hitting a sub-$1 cost for the first time. The company is talking just a little about nits views on the still-emerging USB 3.0, a topic I’d like to hear more about.
For its part, STM showed the first steps in research toward blending .11n and WiMax into an integrated chip suitable for a cellphone. The company has work to do yet before it pulls this one off, but it got plenty of attention at ISSCC for its efforts so far.
Posted by Rick Merritt at 7:37 AM
Tuesday, February 05, 2008
A relatively new IEEE group (802.15.6) has put out a call for proposals for body-area network technology. They seek a “standard optimized for low power devices and operation on, in or around the human [or animal] body to serve a variety of applications including medical, consumer electronics and more.”
Tech requirements include support for a quality of service scheme, extremely low power, and data rates up to 10 Mbits/s. In addition, applications need to consider the so-called Specific Absorption Rate of different body types. They must also respect limits for transmit power in US to less than 1.6 mW and in EU to less than 20 mW.
The group aims to attract to its March meeting in Orlando giant medical electronic and consumer companies to weigh in on what they think will be the big apps for a wireless BAN and the technologies required to serve them. “This is a personal area network with the characteristics of the body in mind,” said Arthur Astrin who chairs the group.
I bumped into Astrin in a medical electronics session at ISSCC this week. Plenty of candidates for this work here i9ncluding developers of wristwatches and disposable band-aids that act as wireless monitors, deep brain implants and much more.
It could be a decade too early, but I think this will be one of the most interesting new wireless areas to watch.
Posted by Rick Merritt at 3:57 PM
Friday, February 01, 2008
That’s what at least one attendee was wondering after the first industry review of the draft USB 3.0 spec. A group of as many as 300 came to the two-day, closed-door event in San Jose last week to hear the latest about the emerging Super Speed USB which aims to hit data rates up to 5 Gbits/second.
“I have attended two or three wireless USB events, and they were not as big as this. Almost everyone I know in the supply chain was there,” said Mark Fu, director of marketing for the connectivity group at SMSC which makes USB transceivers.
“There are a lot of things you can do with 5 Gbits/s such as carry uncompressed video, so I wonder whether this will challenge HDMI,” said Fu.
Interesting thought, given the fact last year Intel bailed on the UDI effort Silicon Image created to get HDMI into PCs and mobile systems. Although HDMI is well entrenched in HDTVs, it is looking less like an industry standard and more like a Silicon Image standard every day. And wouldn’t OEMs prefer one link (USB) that handles everything?
External hard and solid-state drives could use the target 300 Mbytes/s throughput for USB 3.0 as an alternative to external serial ATA, too.
“That’s almost as fast as SATA II,” Fu said.
Any challenge is a ways off. The 3.0 spec is only at a 0.78 version, although Intel hopes to have it complete in June.
“The way they talked about the connectors, EMI issues and other concepts you got the impression they have done a lot of work in the lab and this is not just a paper spec,” said Fu.
Indeed, Intel has some basic USB 3.0 hardware running the lab. There was a lot of back-and-forth in the San Jose meeting about cable issues—such as whether to consolidate USB 3.0 to just one micro-B style connector—the basic PHY, link and protocol aspects of the spec “seemed pretty solid,” said Fu.
I said before USB 3.0 will douse whatever heat Firewire is still generating. Now maybe a broader story about USB uber alles is in the making.
I reached out to the Intel and USB-IF folks for more today, but hey it’s Friday. I didn’t hear back. If anyone else at the big meeting has any comments to lend, go ahead and post them here or drop me a line at firstname.lastname@example.org
Posted by Rick Merritt at 4:46 PM
That’s what IMS Research believes forecasting that the market for cameras with integrated Global Positioning Systems capabilities will grow 200 percent on a compound annual basis over the next five years--coming off a very small sub-million unit base, of course. Makers of digital cameras and GPS chips are both hungry to open up new opportunities. But I think the marketing geniuses have more work ahead to flesh out this concept of geo-tagging pictures and turn it into a compelling use case.
Matia Grossi, author of the IMS report, does well to note that camera makers believe GPS chips are “too power hungry, too expensive and take too long to get a location fix.”
New approaches aim to address those issues, according to the report. The Snapspot software from NXP-spinout Geotate turns on a GPS receiver only for a fraction of a second while the user takes a picture. Air Semiconductor’s Airwave-1 keeps the receiver always on, drawing 1mA average and dynamically balancing accuracy with power based on the application.
Don’t expect many GPS-enabled cameras at the PMA conference in Vegas, but there’s enough activity to suggest they will come along eventually, Grossi concludes.
Posted by Rick Merritt at 11:01 AM
Thursday, January 31, 2008
I’ve heard a hat trick of power problems this week. In 10G Ethernet, signaling and modeling—everywhere you turn—power is increasingly a key constraint.
A chief technologist at Solarflare wrote a somewhat rambling treatise that concludes there is no power budget for running TCP offload on 10Gbit Ethernet chips for the foreseeable future. That’s a somewhat self-serving conclusion, but at least he was specific about the numbers.
TOE-enabled chips will consume about 16W in 130nm or 10W in 65nm—two hot for 2008-class dual-ported products, he said. By contrast, chips that push the TCP stack to a host CPU will dissipate 5 and 4W respectively in 130 and 65nm technology, he said. If anybody wants to take issue with those numbers, let’s have at it. iWarp people, sound off!
Separately, HP fellow Terry Morris said the DesignCon organizers spent much of their time trying to figure out how to cover the hot topic of co-design to handle the increasing merger of power, signaling and timing issues for chip, board and systems engineers. Power effects at multi-gigabit speeds have to be part of the design analysis and there are no standard tools to do this, he said. EDA vendors, whatcha got to say about those beans?
Last but not least I heard Grant Martin, chief scientist at Tensilica, today say that one of the big issues in the new virtual prototyping style of design is a lack of energy models.
"We are much farther behind in this area," Martin said. "We have done some things with energy modeling for our devices at Tensilica, but you really need to know about the energy models for other devices in a system—so we need lots of friends," he added.
OK, so go sign up as friends at Grant’s MySpace page, and let’s get an industry consortium going on this issue, too.
Posted by Rick Merritt at 4:40 PM
Wednesday, January 30, 2008
Sprint is giving 230 people in the San Francisco area cellphones equipped with near-field communications chips as part of a new service trial. Participants will be able to automatically buy tickets for the local mass transit system and pay for food at Jack in the Box fast food joints using NFC.
Users will be able to forget about subway tickets as long as they have their phone in their pocket. They will also be able to waive the phone in front of Jack in the Box smart ads in the subway station to get directions to the nearest outlet.
I suspect participants in the four-month trial will find it a fun novelty, but building a more interesting eco-system is the big issue. The people who use the latest cellphone features are not typically mass transit riders ordering fries at Jack in the Box.
Surely Sprint can learn a few things from these 230 users who will spread the word to others about NFC. But the road to making it interesting to use your phone in place of your debit/credit card is easy to see and hard to pave. Sprint must get a world of service providers to sign on to NFC pay-by-phone services, replace that world of magnetic stripe systems we all rely on these days and put up a bunch of smart ads, kiosks and services.
That’s going to be a lot more crucial than any trial and take a lot longer than four months.
Posted by Rick Merritt at 10:28 PM
Tuesday, January 29, 2008
Intel and Atheros will square off next week at ISSCC when they both describe their latest 802.11n Wi-Fi chips. As usual, Intel has the edge in process technology using 90nm design rules for a 1x2 MIMO transceiver with integrated filters and power amp. But the 130nm Atheros part is no slouch with 2x2 MIMO and at least a similar level of integration.
STMicro will give a peak at the future with papers on a separate transmitter and receiver that handles Wi-Fi and WiMax at 2.4-2.7 GHz bands. Made in 65nm technology, the designs point the way to the kind of integrated comms parts Intel says it is planning for future Centrino notebooks.
Also of note in wireless, I see Georgia Institute of Tech will describe a 60 GHz radio in 90nm CMOS that consumes just 200mW but can kick out data at rates of 7-15 Gbits/s. It’s just an academic design, mind you, but the first 60 GHz radio I have seen described outside of the work of startup SiBeam which was one of the wireless stars of the recent CES.
Posted by Rick Merritt at 6:18 PM
Monday, January 28, 2008
It’s still early days for the next big step in Ethernet networking, the leap to 40 and 100G standards. But the newly approved IEEE P802.3ba task force is on the case and hopes to have a cogent draft by this fall.
A look at the group’s agenda from Portland last week shows they have been drinking double espresso. The agenda included more than 30 presentations from a wide variety of players including top dogs such as Alcatel, Broadcom, Cisco—the ABCs of Ethernet.
Interestingly, HuaWei is stepping up to take a significant role in the committee and is making solid technical proposals. The days of hacking on Cisco code are over for this emerging powerhouse.
Defining a common electrical interface has become one early issue for the standard. Joel Goergen of Force 10 Networks did a good job advocating for using the practical 4x10G and 10x10G links initially, moving to 4x25G in future—and maybe some 10x3 or 8x6 action in the medium term. Leveraging the 10GBase-KR standard for serial 10G links and getting in step with the Optical Internetworking Forum’s efforts on 25G will be key, according to Goergen.
If there are broader issues that need to be aired here, I am all ears. Post a note or drop a line to email@example.com. In the meantime...Man, I am going to have get me some of that good Portland coffee!
Posted by Rick Merritt at 2:44 PM
Saturday, January 26, 2008
10G Ethernet needs some pushing with the costs still relatively high for optical versions and the power consumption still too high for copper. That’s supposed to start shifting in 2008, and the upcoming ISSCC conference in San Francisco may be the place to get the first glimpse of the prospects.
Based on the sketchy abstracts I have on hand, it’s hard to tell what the horizon looks like. Startup Teranetics will describe a 130nm chip that drives 10G Ethernet over 100 meters of UTP, but it burns a fairly hefty 10.5W doing it. And though made in a mainstream 130nm process, it still requires two fairly large blocks--a 55mm-squared analog front end and a 150mm-squared digital processor. What’s more the abstract does not say which flavor of UTP they are using, probably a pricey Cat7.
Startup Aeluros will describe a protocol-independent serial 10G transmitter using a three-tap filter for use in optical cables or backplanes that eats just 165 mW. But it doesn’t sound like it’s for copper cables and there is no word about a full transceiver.
Well, there are a lot of really smart engineers behind these ISSCC papers, but they are engineers after all—not magicians.
Posted by Rick Merritt at 6:35 PM
Thursday, January 24, 2008
That’s what the man said. Admittedly the man was Bob Whitman whose job is to develop new business opportunities for Corning, a leading maker of fibre optical cables. And admittedly it was a minor side point, heavily qualified, as part of a presentation focused on fibre-to-the home (FTTH) networks which are what really have Bob excited these days.
But that’s what the man said. And I repeat: “Until recently we haven’t even considered an in-home market for fibre, but we are beginning to think about it now,” he said.
I wrote on Monday that chip makers are saying the next-generation of systems that terminate a FTTH network will be integrated with residential gateways, making them a sort of Trojan Horse for optical in the home. However, those systems expect to link to copper and wireless home nets.
For ten years home nets have been living under the dictate “no new wires.” I don’t expect the Corning folks to get any reprieve from that guidance until there is such salivating demand for an application so high in bandwidth that no cleaver modulation over wireless, copper or even coax can cost effectively meet it.
Posted by Rick Merritt at 10:30 AM
Wednesday, January 23, 2008
Tomorrow’s cellphones could sport as many as eight radios and require up to 11 antennas, according to Sudhir Dixit, head of network technology for Nokia Siemens Networks, speaking in a presentation at Photonics West today. “There’s an antenna jangle out there,” he said.
The radios included Bluetooth, multi-mode and frequency cellular (with four antennas), DVB-H, FM, GPS, RFID, Wi-Fi and ultrawideband. Dixit, an IEEE Fellow, called for new strides in miniaturization. Without them that future phone might need to dedicate a whopping 25 percent of its area to antennas and radios, he added.
The Nokia Siemens tech exec is admittedly making an extreme case that’s not likely to be on the engineer’s benchtop anytime soon. Similarly his projections of future phones with VGA resolutions on OLED screens and Gbytes of memory was a bit beyond the pale.
That said, we’ve been watching a slow motion explosion in wireless technologies over the last decade. The trend line Dixit outlined is real, even if the details may be a bit exaggerated for affect.
Posted by Rick Merritt at 7:24 PM
Tuesday, January 22, 2008
By 2012, more than 71 million homes worldwide will be outfitted with home networks thanks to a residential gateway box from a major carrier. That’s the prediction from a new study called “Networks in the Home” released by Parks Associates this morning.
Carriers are most interested in providing features such as multi-room digital video recording, streaming media and home monitoring, the report adds. A secondary goal is getting remote management technology into the home so they can more easily fix problems and up-sell services, according to the report.
That all rings true but, as usual for the market research world, probably a little optimistic on the 70 million figure. Anyway, after a decade of talk about home networking, it does look like it is sloooowly beginning to happen with carriers such as Verizon delivering more than one million gateways equipped with Multimedia over Coax and AT&T installing a fair amount of HPNA as well.
One of the next knee-jerk questions is when will they start cranking up the volume on bandwidth. A Verizon spokesman at a November event told me they are planning to move to multi-Gbit optical links soon so they are requesting MoCA quickly shift to 400 Mbits/second. A handful of chip makers are optimistically hoping this shift to Gigabit PON networks and unified boxes that marry the optical link to the home net kicks in big time in 2008.
Separately, over in the data center, Emulex is validating the N_Port ID Virtualization (NPIV) industry standard as the basis for virtualization in its products. At a Cisco event, Emulex is showing the technology across a range of high end data-center networking boards including a 10Gbit/s board for Fibre Channel over Ethernet. Unless I missed something, this FCoE board is still a prototype product. But it’s a good indication of progress on the technology if such products are now broadly getting demonstrated at industry events.
Everywhere I turn this morning there’s just a little bit of hope about new technologies for the new year. While stock markets dive into fears of recession, this is good news.
Posted by Rick Merritt at 10:15 AM
Thursday, January 17, 2008
Flipping through my daytimer, I see DesignCon is just a couple weeks away. So, it’s time for me to get up to speed--as it were--about all the issues pushing backplanes and transceivers to the bleeding edge and beyond. If you live and breathe this fiery stuff, drop me a note at firstname.lastname@example.org to let me know what’s keeping you up at night.
There’s no doubt the need for speed continues to send a plume of smoke up form engineers gathered at this event. Lee Goldberg will host an interesting panel on the move to 20-25 Gbit/second backplane links. My colleague Loring Wirbel will gather thought leaders from the applications world to talk about the drive to 100 Gbit/s networks.
Bring a fire extinguisher.
Posted by Rick Merritt at 5:52 PM
Wednesday, January 16, 2008
I attended a panel of car and plane electronics makers at the Consumer Electronics Show this month that gave some insights into some of their most pressing interconnect issues.
Bill Mattingly, a vice president for electronics at Chrysler, said car makers need standard interfaces for consumer electronics—at least at the physical layer. “We can accept the fact we need to do protocol updates every year,” he said.
Gary Jablonski, a development manager for infotainment systems at Ford, said he expects to leverage more PC standards while waiting for car industry efforts. USB has already become the de factor audio interface for cars, he added.
A representative from an avionics company said car and plane makers need to agree on a video interface because USB is not adequate to the task. Choose wisely because it can cost a whopping $240 million to outfit a plane with new connectors, he added.
One after-market vendor said the industry needs a protocol to pause entertainment devices when the vehicle senses the driver is in a critical phase such as fast breaking or highway acceleration. “We need to get a standard for driver distraction,” he said.
Very basic reconfigurable radios to link to the outside world will also be critical, he added. Cars need to last twenty years and that may require spanning links from GSM to LTE or even WiMax, he said.
As for wireless inside the car, all sides said ultrawideband is a non-starter. “UWB has massive EMC problems in the car. It would be a nightmare,” said one panelist.
UWB and Bluetooth—due to its channel structure--are both not suitable for the airplane, said the avionics rep, although Wi-Fi is a maybe.
As if there were not enough problems, Chris Steiner, OEM sales manager at Garmin, said GPS makers have no common connector for their devices yet. “That’s one of the big challenges we, Magellan and others have ahead,” he said.
Posted by Rick Merritt at 5:19 PM
Tuesday, January 15, 2008
Here we go again. The consumer electronics industry has just gotten to the point where there is a wide range of universal remote controls available to supervise all your gear no matter who made it. But with the rise of digital, high def products the technology is about to take a new twist.
The folks behind the High Definition Multimedia Interface laid out a scheme called HDMI-CEC. The standard lets each consumer company establish a baseline of interoperability with each other’s digital high def remote control products with room left over to add on any proprietary bits if they wanted to put a few special features in their devices.
Posted by Rick Merritt at 6:53 AM