Archive for the ‘Semiconductors’ Category

SIA January Sales Report Indicates Strength for 2010
Friday, March 5th, 2010

The SIA released their report for January sales. The numbers look dramatic with a 47% year-on-year increase. That, however, is more an indication of how bad January 2009 was than how good January 2010 looks. January 2009 was a $15B month in the industry making the run rate look like the late 1990′s. The SIA reports January 2010 as a $22.4B month which would be a harbinger of a $250-$270B year for the industry. I believe this is likely and well above the current SIA forecast of $240B. The February report, which should be available in early April, just ahead of earnings, should be telling. February typically shows the “seasonal pattern” oft cited in company earnings announcements and mentioned by several in this quarter’s guidance. If those numbers show only modest decline or, better, sales flat with January, its another harbinger for a good year. We’re well into the dynamics of an up cycle and there may be some additional “running room” for revenue and valuation growth. We should be looking now, however, for what will make it “run out of steam”, probably in early 2011.

SIA Report Mar-10 table

100% of the growth is in Asia!

SIA Report Mar-10 graph

Did anyone mention "V-shaped" recovery? ...

Tags: ,
Posted in Forecasts, Semiconductors | 1 Comment »

Semiconductor Q4 2009 Results
Monday, February 22nd, 2010

As predicted, it was a great quarter for the semiconductor industry. Overall, average results were up over 5% Q/Q with very strong quarters reported by INTEL and AMD, both benefitting from the Windows 7 upgrade cycle that is still early in its progression. The “Xitera” FPGA pair delivered spectacular results with Q/Q growth approaching 25% and, in a significant way, outpacing overall industry growth. A little too early to tell but the much heralded design capture phenomena may be seeing the light of day. Very strong growth reported by Atheros as it continues to benefit from 802.11 spreading to connect all things digitial to the network. And this reported growth includes a negligible contribution from the just completed acquisition of Intellon.

Guidance remains conservative, a healthy view given the uncertainty in the overall economy.   I would speculate, however, that the combination of the PC market benefiting from the Windows 7 release, the strong onset of the “smartPhone” product category and the somewhat delayed but necessary investment in communication infrastructure will allow semiconductors to have some “strong sailing in the rough seas” of the global economy for 2010.

2009 Q4 Results

Tags: , , , , , , , , , ,
Posted in Forecasts, Semiconductors | No Comments »

… and for 2009 the winners are …
Friday, January 22nd, 2010

It’s the tough times that separate the weak from the strong and 2009 was tough enough to buckle the knees of the most stalwart. Given that semiconductor industry revenue dropped by about 10-11% in 2009, it’s interesting to look at how the top fabless semiconductor companies faired. This article from Semiconductor International lists the results from 2009. It’s a little premature, as many of the 4th quarter results aren’t fully out yet, but it’s a pretty good snapshot. Given the 10-11% average drop it’s logical that the average for the top 25 was significantly less, about 5%. However, even in this group there was a considerable spread.

Top 25 Fabless Semiconductor Companies by 2009 Revenue

Top 25 Fabless Semiconductor Companies by 2009 Revenue

Looking at the table a few things standout. First, it’s interesting to separate the list by country of origin. As there is only one company from each of Japan and Europe, it hard to conclude anything on average except that neither region has embraced the fabless model. Note, however, the stark difference between the results in the U.S. and in Taiwan. The average decline for a U.S. based company in the top 25 was 8.7%, pretty close to the industry average. A company in Taiwan in the top 25, on the other hand, grew an average of 13%!!

As, or maybe more interesting, is to look at how the companies did against their regional average. Seven companies beat the regional average in the U.S.: Atheros, Silicon Labs, Qualcomm, Avago, PMC-Sierra, Broadcom and Semtech. Is there any doubt that that list represents the “cream of the crop” of fabless IC companies. More unique, however, is that two U.S. companies managed significant growth in this environment. Congratulations to Atheros and Silicon Labs, among my most admired companies in the fab-less industry.

But let’s turn attention to Taiwan. With the regional average being 13% growth, these companies are operating in a different (more condusive to growth?) environment. Within that environment, three companies beat that average, Mstar, Mediatek and Realtek. These names have and should strike fear in the management of U.S. based companies. These companies continue to grow on a business model focused on cost based design for volume applications. They make strong ties to local foundries and easily capture design wins with local ODMs. Given that TSMC is destined to become the 2nd largest semiconductor company in the world in just a few years we are only going to see the strengths of Taiwan based companies increase. U.S. management and investors, take note …

Top 25 Fabless Semiconductor Separate by Country and Sort by 2009 Growth

Top 25 Fabless Semiconductor Separate by Country and Sort by 2009 Growth

Tags: , , , , , , , ,
Posted in Semiconductors, Uncategorized | No Comments »

Pre-announcements on the way …
Tuesday, January 5th, 2010

The SIA released November sales estimates for the semiconductor industry on Monday. The table in that report is shown below. The press release highlights that it’s the first month of year-on-year growth since the downturn began in 2008. That’s a great milestone but there’s also a couple of other interesting tidbits here. First, it’s close to a record month in the semiconductor industry with the run rate returning to a >$250B/ann level. More interesting to me, however, is the quarter on quarter progression that the November sales show, >18% according to the table. That should auger well for the outcomes of this quarter for most companies in the industry as that, by definition, means that the average semiconductor company should report a 15% or greater growth from the September ending quarter.

SIA November 2009 Sales Estimate

SIA November 2009 Sales Estimate


SIA Historic Sales Estamates

Semiconductor Industry Revenue

That’s not quite what the industry guided to, however …

Xilinx and Altera, both broadly exposed geographically and across application segments did revise guidance in December, guiding guided to up 16-20% and up 15-18% respectively … so they’re consistent with the predicted average. Intel, however, provided guidance in the up 6-8% range in their October earnings release and did not modify that guidance when they provided an update in November. Looking at a couple of other large semiconductor companies guidance:

Hmmm … I’d say that INTEL, Broadcom, Nvidia and TI are all fairly comfortable in their guidance. I expect that we might see a few more pre-announcements and revisions here shortly as well. And I’d also say, that Altera and Xilinx might actually be outgrowing the industry … but we’ll have to wait and see on that one.

Tags: , , , , , , , , ,
Posted in FPGA, Forecasts, Semiconductors | No Comments »

Semiconductor Growth and FPGAs … hmmm
Tuesday, October 27th, 2009

FPGAs are one of the seminal inventions of the Semiconductor Industry.    They have and continue to be a hugely productive and generally cost effective way of exploring designs in complex electronic systems.     Theory says, in fact, that they are increasingly a good choice for full production implementation of systems electronics as the cost of full ASIC design becomes prohibitive.   A whole body of pundits (1, 2)  predict the demise of custom ASICs as the capabilities of FPGA continues to progress.   In fact, there’s a very lucid presentation of why the numbers say that FPGAs will subsume ASIC development over time from the perspective of 2003 and I, generally, agree with this analysis.

OK, so now that we’re about 6 years removed from this analysis what’s happened?    I’ll put out some data.    The chart below shows three revenue sequences since 1994.    One is the SIA data for overall semiconductor industry revenue, the second is the same for fab-less from the GSA and the third is the revenue history on a yearly basis of Xilinx and Altera (summed).    The assumption is that Xilinx and Altera ARE the FPGA market, or at least an excellent proxy as they comprise some 85% or more of the market.    The graph might be a little hard to understand as I’ve scaled the three sequences differently to allow them to plot conveniently on the same axis:  “Xitera” is plotted in Billions, Fab-less is $10′s of Billions and overall Semi is $100′s of Billions.    Regardless, do you note anything unusual?

Semiconductory Industry Revenue

Two things stand out:   First is the fact that the fab-less industry has continued to outpace the overall industry in the last 15 years, not surprizing … but a trend that has slowed dramatically in the last few years.     Second, of course, is the anomoly of the year 2000.    It was a banner year for FPGAs (and many others in the industry).   But if you study the overall trend there’s something quite surprizing and in direct contrast with the predictions of the pundits in the last several years.     Take the same data for the overall semiductor industry and the “Xitera” sequence and normalize for 1999 (i.e. Revenue in 1999 = 1).   Plotting, the graph below results.  Conclusion:  the FPGA market has not grown faster than the overall semiconductor market for the last decade!    In other words, despite the theory that FPGA’s should be capturing more content because they are a more overall total cost effective solution when including development cost, there has been zero content capture relative to the overall industry …

image001

I discussed this with a long time executive in this industry recently and he was quite aware.    Some of this is ‘self-imposed’.    Xilinx and Altera are among of the best financially performing companies in the industry and, uniquely in the fab-less space, pay a dividend to their shareholders.    Capturing new applications, no doubt, requires business model innovation and sometime aggressive price moves.    Arguably, they are making a choice for margin and profitability over growth and, certainly, the current investment mood favors that.    I think, however, the numbers in the analysis are all too real.    ASIC and ASSP development at advanced nodes is simply becoming too cost prohibative, FPGAs will become a better choice for overall economic cost.   

I remain convinced that FPGAs will re-assert their relative growth against the industry.    This may, however, look like another effect of the Moore’s Law exponential sequence.   It will ride underneath the measured data until the the order of magnitude of the effect makes it immediate.   Suddenly, a step function will occur.    As is often the case in these “steps” it provides an opportunity for some to “step-up” … and some to “trip”.

Tags: , , , , , ,
Posted in FPGA, Forecasts, Semiconductors, Uncategorized | No Comments »

IEEE Spectrums 25 “Earth-shaking” chips …
Saturday, October 24th, 2009

I ran across the IEEE Spectrum article “25 Microchips that Shook the World” this morning.   I love history and especially tech history so I enjoyed the article very much.   However, as is often the case in microelectronics, the article is written from a compute centric viewpoint and misses the mark widely on communications.  

Let’s put out my bias.   The most earthshaking invention of the last 25 years is the internet.   The internet and it’s ‘world-wide’ reach has done more to re-shape our lives;  the way we use time, the way we perceive other cultures and the ability to acquire new knowledge; than any other modern invention previous or since.     After the internet, I put the cell phone in second.     For 100 years the basic measure for economic development has been telecommunication infrastructure; bringing the ability to communicate to the population.    Cellular has brought the ability to make a simple phone call to more people in the last 10 years then the entire history of telephony.    I’m a communication zealot, these technologies have had huge and unquestionably positive impact on the human condition.

The internet was enabled in 1994 by a standard and the resulting chip development, the V.34 modem.     The V.34 ITU standard was ratified in July of 1994.    The V.34 standard introduced the breakthrough of Trellis coding and broke the 10Kbps bandwidth barrier over the phone line.    In fact, with all features implemented, V.34 neared the Shannon theoretical limit for digital bandwidth over the phone line (~35Kbps).    This was enough to make more than e-mail viable through the internet and, not coincidentally, the 1st version of the Netscape browser was announced in October of that same year, 1994.     These events were followed by quick and significant commercial success.    Rockwell, then one of the leaders in modem chip technology, generated about $3B in revenue in V.34 sales between 1995 and 1996 alone and Netscape completed one of the most sought after IPO’s in history on August 9, 1995.    The next four years, through 2000, were the heyday of the internet as its reach and application exploded, eventually leading to the infamous ‘tech bubble’ of 2000 and 2001.

The Spectrum’s article seems to have missed this one completely.    They do put the DSL modem from Amati on the list of the top 25 and also recognize the 802.11b WLAN chip from Lucent when the list is expanded to 40.     But these chips succeeded only because the basic architecture of the internet was established.    These chips then enabled broadband application, albeit opening up a significant new world of applications and impact, but the tale was told in 1994 with V.34 and Netscape.    I therefore submit, the RC336 modem chip from Rockwell as one of the top “Earth-shaking” chips of all time.

My second submission to this list is the C50 from TI.    The C50 from TI was the follow-on implementation from their seminal fixed-point DSP the C25.    The C50 built on the C25′s instruction set but was implemented in a low-power architecture specifically for digital cellular applications.   The C50 became the heart of a large array of cell phones that took full advantage of the GSM systems launched in Europe.     The digital cellular system combined with the low-power features of the C50 enabled more compact handsets with significantly longer battery life than any previous analog implementation.    It was these breakthroughs in size and usability that made cellular phones a technology that was easy to adopt and the numbers quickly showed it.    Again, it was a core chip implementation that enabled communication to the masses.

I again commend Spectrum for its article and very much enjoyed the perspective on history.    It missed the mark on communication chips, however, and the V.34 modem and the C50 have to be considered on the list.     The argument could have been made on commercial significance alone.     One of my favorite questions is to ask is to name “billion dollar chips”, chip designs that have generated $1B of revenue in a year.     It’s a short list: there’s memory chips, there’s each generation of the x86, there’s the 68000 from Motorola and then … 

… there’s the V.34 modem from Rockwell and the C50 from TI.    The first three (or their precursors) are all part of the Spectrum article.    The last two belong there as well.

Tags: , , , ,
Posted in Chip Design, Semiconductors | 6 Comments »

Foundry forecast … real and meaningful
Tuesday, October 20th, 2009

In my previous post, I noted the strong forecast for the pure play foundry and observed that it requires a structural change to be true.   In asking a few questions to colleagues the suggestion that this is a result of the “hybrid” supply model, or from the design companies perspective “fab-lite” strategy seems to hold water.    The core dynamic, well known and examined in detail in this April article from The Economist magazine, is the fact only IDMs with significant scale can afford leading edge fabs.   While increasingly true at each previous node this seems to have “hit at wall” at 65nm and very few IDMs made the investment.     At this node then there is a “transfer” effect as large IDMs begin to participate in earnest in product outsourcing.    These design decision were made over the past 3 years and are now showing up in product shipments.   Announcements recently by ST, NXP, Infineon and AMD are indicative of the trend and show participation of four of the top ten semiconductor companies in the world.    

Estimating the size of the effect is shown in an update of the graph from the last post.   Two assumptions are made, one that the IC Insight pure play foundry forecast is about right and two, existing fab-less companies, in aggregate grow at the rate of the current SIA industry forecast.    If both those assumption are true then there is $32B of semiconductor revenue in 2012 that would be revenue “transferred” from IDM in-house supply to outsourced, or fab-less supply.     To put it in context, that implies that the foundries have just gained a new design-in customer that grows from zero to almost the size of INTEL in a 3-4 year period of time!

New revenue from hybrid model transferring to foundries

New revenue from hybrid model transferring to foundries

Having lived through the transition from IDM, to “fab-lite”, to fab-less at Conexant I can say that this is going to be a very tough transition for the ‘fab-lite’ companies.     Though a foundry customer, their existing fabs still need filling to make bottom line numbers work.    This tends to pull internal product lines in multiple direction up until the point where the design entity and the fab separate completely.   I commend AMD for making the decision to separate from the fab business immediately and in one step.     In addition, the creation of Global Foundries and their immediate movement to gain scale through consolidation is, while highly risky in execution, the only possible path to long term survival.   

The long predicted maturation and consolidation of the semiconductor industry is showing up in the numbers.     It is very clear that there will be three “mega-entities” in the industry 5 years from now:  INTEL (PC), Samsung (memory)and TSMC (pure-play foundry).    The real question now is who else has the business model, IP scale and plan that delivers sustainability for long term growth.

Tags: , , , , , , , ,
Posted in Semiconductors, Uncategorized | No Comments »

3 year Foundry forecast looks good … but wait a minute!?!
Friday, October 16th, 2009

Semiconductor International is highlighting the release of a couple of forecasts on the foundry industry.   Those forecasts are  quite optimistic for that segment of the semiconductor market.    The two reports, one from iSuppli and another from IC Insights, are quite consistent in the actual revenues through 2008 and are fairly consistent in their view of the years ahead.    Both show foundry revenue of about $20B in 2008, about a 15% decline in 2009 to around $17B, followed by three solid years of 20%/annum growth.    Doing the math, at that rate, there will be a pure-play foundry industry of about $30-35B in 2012 or nearly double the business of the segment in 2009 and 75% above 2008 levels.    Wow!  Corks must be a-poppin’ in Taiwan and around Asia!!

159406-IC_Insights_Foundry_Forecast

159408-iSuppli_sees_three_good_years_ahead_of_the_foundry_industry_

But wait a minute, something’s up here … !?   

Let’s look at a few structural facts that underlay this forecast.    First, let’s look at the pure-play foundry industry itself.      As noted in the SI article, this industry is characterized by increasing consolidation.    The market shares in the industry as reported for Q2 of 2009 are shown below.   Note that TSMC is more than half the market and the top 5 foundries which include UMC, Chartered, SMIC and CRM (China Resource) comprise just over 90% of the pure-play foundry segment (source: GSA).     The SI article also highlights the increasing consolidation pressure going forward as the investment cost in leading edge technology winnows out the weaker players.   Prima facia, then, these five leading foundries should outgrow their industry segment and be enjoying about 20-25% growth for the next three years!!   Does that mean a $20B TSMC in 2012!?   Wow!  More champagne (?) … or maybe something stronger.

Q209WebsiteFinancialSpreadsheet_001_30179_image001

Let’s look further.     We can compare these pure-play foundry actuals and forecasts to actuals reported by the Global Semiconductor Alliance (GSA) and forecasts by the SIA.     The graph below shows three independently developed actual and forecasted series on the same graph.     First is the series that is the IC Insights view of the pure-play foundry market, second is the actuals for the fabless industry as reported by the GSA and the third is the percent that the fab-less segment represents of SIA actuals and forecast of the overall semiconductor industry.     Observe some relationships:     First, the fab-less segment has grown to 20% of the overall semiconductor industry and been fairly stable at that percentage for the past 4 years.        Note also that the pure play foundry business has, at the same time, been in “lock step” with the fab-less segment in the years 2006 through 2009 at 40%.  This makes perfect sense as the foundries basically serve the fab-less segment’s cost of goods.   But if the IC Insight and iSuppli forecasts are right then something structural is about to change and change dramatically.     Assuming the IC Insight and iSuppli forecasts are right and assuming that “40% of fab-less” generally holds, the resulting forecast for the fab-less segment is shown (yellow “hatched” columns).    That forecast as a percent of the SIA forecast is also shown.    If right, that says the fab-less segment will expand dramatically to become about one third of the overall semiconductor market!!    Again, wow!!

Industry Workbook_13465_image001

Fab-less forecast implied by IC Insights pure-play foundry forecast

  

Frankly, I doubt it.    But IC Insights and iSuppli are not naive and two independent forecasters have developed surprisingly consistent views of the market.    Something else is happening.    Is this because IDMs are entering hybrid supply models?    Certainly the ATOM initiative at INTEL is a strong indicator in that direction.     Regardless, these numbers are dramatic and more questions need to be asked.    Simple solution is to buy the reports and find out what’s driving it.     Unfortunately, not in my budget this year but I will ask some more questions and find out what’s going on.

Tags: , , , , , , , ,
Posted in Forecasts, Semiconductors | 1 Comment »

Welcome to October: Enjoy the ride!!
Wednesday, October 14th, 2009

Rabid anticipation? Check. Wild uncertainty? Check. Heart-pounding tension? Check.  Thrill of Victory? Check.   Agony of Defeat?  Check.   It’s October and there’s so much excitement in the air.   Dodgers playing the Phillies starting tomorrow night; Angels playing the Yankees starting on Friday.   Back to the tried and true names delivering solid and exciting results!!

… and so it goes in the semiconductor market.   Intel beats estimates and guides to a quarter exceeding $10B, a mark not hit since Q3 of 2008.   And “hitting for the cycle”, Xilinx, Altera and Avago all meeting or exceeding current quarter results and guiding to higher than analyst expectations.     A new Microsoft operating system released for the PC in time for the holiday selling season and it’s getting good reviews this time!    Yeah baby!!  Upgrade cycle!!   Homerun!!  

It’s going to be an exciting few weeks!   The market is receiving the tech news well.    The Dow closed above 10,000,  and the Nasdaq is near its 52 week high.    But the playoffs and the earnings season will end and we’ll return to reality to re-ask the fundamental questions.    Is the housing market ready to recover?    Has the need for “de-levering” been completed in Asia?   Is the market optimism sustainable through the end of the year?  

Enjoy the ride for October … we may still be in a ’re-building’ mode come early 2010.

Tags: ,
Posted in Semiconductors | No Comments »

Home Networking … and now what
Wednesday, October 14th, 2009

Well, well … all things in their time I suppose.

Home Networking Technology is an area of action in a year where most technologies appear at an apparent standstill due to the economic backdrop.     In Home Networking, however, a standard has been published (G.hn), service providers have embraced new technology in the form of MoCA, and in September and October a series of high value semiconductor acquisitions has been announced (Atheros acquires Intellon, Sigma acquires CopperGate).   Certainly some attention deserves to be paid.

First, let’s remind ourselves as to what home networking is and why its a changing landscape.     The image below shows a view of home networking with an automation bias.    Home networks today are mostly computer networks implemented with a combination of traditional Cat 5/6 UTP wiring complemented with wireless networking, almost always using one form or another of 802.11.     Today’s networks are typically “PC centric”, serving a need to connect multiple computers to a shared broadband connection.   Often, however, there’s the additional devices that also benefit from that same broadband connection.   In my household, as an example, in addition to four computers on the network (three desktops and a laptop), we have two game systems (a Wii and an Xbox 360), three set-top boxes (two of them DVRs), add-in the occasional iPod, guest laptop and you have a real networked system brewing.

Home Network

Home Network with Automation Applications

The network technology and applications described probably represent the limit of what can be accomplished with today’s technology and here’s why.    First, for the network described above, I’m lucky, we’re in a new house with Cat 5e wiring pre-installed.   We have a single “wiring closet” where all the connections are “hubbed” and there’s an RJ-45 jack in all bedrooms, the living room and two down in the basement.   That, however, is not common and continues to drive the need for technologies that require “no new wires” and provide the same level of highly reliable connectivity.    Second, my applications are “data centric”, typically broadband connection sharing outside the home and file and printer sharing inside the home.    We do in our house, however, stretch beyond that limit.    We do have two Airport Express products connected into the home network and use them to stream Audio from our computers to the house stereos.   Works OK but we do find ourselves subject to the occasional “glitch” or “hang” in the network.    But the biggie, that causes yelling up and down the stairwell is the Xbox Live connection for the boys.    Their complaints of “lag-gy” connections causing them to lose a battle in Gears of War are without (… further) debate an unacceptable level of performance.   It was the one and only application that caused me to upgrade the broadband connection from 768Kpbs to 3Mpbs and had me working hard to figure out how to configure the gateway as a “pass-through” for that application.

That these limitations are real and being hit by even the most advanced users (of which I count myself among them) means there is opportunity.     The requirements to overcome these limitations can be categorized into a few simple buckets:

  1. Whole House Coverage with “no new wires”
  2. Ability to work with “real-time streaming media”, specifically Audio and Video
  3. Ability to enable “nodes” (especially endpoints) at extremely low marginal cost

Given that framework let’s look at the currently available and emerging technologies and see what all the fuss is about:

Ethernet over Traditional UTP Cable

Ethernet over UTP works great!!!  Every time!!!  And it’s really low cost!  Gosh, I can get an Ethernet hub for about $20 bucks and everyone of my computers comes with the RJ-45 without spending an extra dime.    I’d say 100Mbps to every node is coming for less than $10/node (wiring not included) and I can up that to 1Gbps for selected nodes at significantly less than $50/node.    And note that 1Gbps Ethernet is great, perfect, more than enough for almost all media applications despite Ethernet not being originally designed for media streaming.   It gets back to the difference between packet-based and circuit-based connectivity (which you can enjoy the dialogue) and why VOIP no longer needs a circuit-switch network.    However, it’s a big time FAIL on “no new wires” and the poster child for why home networking technologies are needed.   More than that, even at these very low costs, it’s not low enough for true ubiquity,  simple “just works” networking of “everything” that you want to monitor, control or retrieve or provide content to requires something that can be added to low tech stuff for less than a buck … think lamp switch.

802.11

802.11 has been a hugely successful technology.      And within the context of our three criteria,  it fares well.    For many houses of modest size coverage is quite good depending on the exact implementation and bandwidth required.    The “no new wires” criteria is, of course, intrinsic.      The media piece of it is questionable, however.     Audio can and does work (remember the bandwidth is really quite low).    Video may work, depending on the resolution required and certainly watching video on the iPod over an 802.11 g or n connection should work every time.    It has, however, been thoroughly tested by a number of video service providers such as AT&T, Comcast, DIRECTV and Echostar.   All have universally come to the conclusion that a reliable video service cannot be delivered over 802.11.    As such everyone of them seeks an alternative solution, and currently several are adopting MoCA.     It is the case that 802.11 has probably achieved close to parity with ethernet on cost/node.    The “PHY” chipset is more expensive, 802.11 is $5-8, maybe less, while the Ethernet PHY is $1-2, but remember there’s no wire and that wire cost money to buy and install.    $10-$20/node is a fair estimate, either way (wired or wireless), for 100Mbps technology.

The “No New Wires” triumvirate – Homeplug, HNPA and MoCA

Look, everyone of these technologies is great and each are important.    There are three basic existing wired infrastructure in almost every home.    First, there’s the wires that deliver power, there’s the wires that deliver phone and more recently many houses have co-ax installed to provide for the TVs.    Each one of these provide a decent media (with individual personality quirks) to provide a data connection to all the outlets of that type in the home.     And these are the technologies that are driving the current wave of acquisitions.    These technologies all excel in providing a “no new wires” value proposition, they all have been designed, especially in their second and third generations to carry media as an core data type.    Cost, however, is an issue in all the technologies.    Each one, with appropriate volume, can and should get to at or below 802.11 cost points.   But, be careful, to carry media the 2nd and 3rd generation solutions are near or above 500Mbps capability.     That’s coming with an expectation of many 10′s of millions of nodes that can demand $20-$50/node (at the valuations being given there better be a TAM exceeding $1B at the semiconductor level).      This “triumvirate” is the source of the current heat in the market.    There’s a heated debate as to which of these media will win.    Entropic is the leader in MoCA and MoCA has significant service provider support.   Intellon (now Atheros) is the leader in powerline.     Coppergate is a bit of a dark horse but a strong believer in G.hn as core part of the resolution of the debate (it’s intended to work on all the different media types and their current pre-G.hn solutions work on multiple media types).

Some Predictions

  • Powerline will win.    I’ve said this for a long time and so far, I’ve been wrong.    But I’m not discouraged.   It’s certainly the case that MoCA appears to have the current momentum.    That is due to the fact that the current owners of the streaming media application are called broadcast TV service providers and they own the co-axial cable in the home.   They installed it, they maintain it, and they are comfortable providing a paid for service over it.   It will continue to have momentum as long as they have control of the content that people want to stream.    Two points, however:    1) it is simply the case that the node coverage of powerline is 100x or more greater than co-ax (compare homes with and without times the outlets per home) and that ubiquity will overcome.   It’s just a matter of time.   And 2) the content is becoming ‘un-leashed’.    We won’t go into a debate of what YouTube and Hulu are doing to content control but it is enough to say that people want to see that stuff on the living room TV.    When the content is not controlled by the provider then the user will choose the medium that is most convenient for them … and ‘ubiquity’ will win.
  • There’s a technology missing in the equation that will emerge over the next few years.   I’m sorry, we’re not going to have 802.11 enabled lamps.    It’s killing a mouse with an elephant gun.    There’s a need for an ultra-low cost technology that works well for simple control and monitoring applications.    It needs to cost less than $1/node and it needs to be able to deal with 10′s of Kbps at most, and the applications will have to tolerate latency.    Bluetooth comes to mind but I’m not, personally, convinced.   Maybe X10 will emerge from the hobbyist realm … ? Is everyone paying attention to Zigbee?
  • There will be a revolutionary new ’user interface’ for retrieving and selecting web based content.   Let’s return to content control.   Why isn’t everyone watching YouTube or Hulu in the living room?   Why are your kids huddled around a PC desk and laughing their butts off?    Because to “browse web content” you need a “browser” and a browser runs on a PC … but that can be changed.     How many people were trying to figure out how to monetize the digital audio technology?    It was all there, but the “key” to re-inventing the channel was the user interface.    In this case, it was Apple, not because of the iPod but because of iTunes and the whole end to end user experience of acquiring and watching digital media.    This change in user paradigm is THE core technology that will determine how the Home Network evolves.    I wonder if we’ll all sit around and wait for Jobs to get it right … again.

Stay tuned … or should I say “Keep Browsing” … or should I say …

Tags: , , , ,
Posted in Home Networking, Semiconductors | 34 Comments »