x
Optical components

Sony to Play in Transceiver Market

While its Playstation2 game boxes sell like hotcakes, Sony Corp. (NYSE: SNE) is hoping to use technology developed for the PS2 to capture 30 percent of the optical transceiver market for LANs and WANs.

The initiative isn’t a samurai-sword-swinging, kung-fu-kicking virtual fighter, but mounting technology used in the company’s DVD, CD, and PS2 players will make the modules “significantly cheaper” than those of rivals, claims Kazuyoshi Onouchi, an assistant manager at the company’s semiconductor laser division. The modules in these consumer products include front-end technology that eliminates the need for active alignment of the laser diode (LD), photodiode (PD), ball lens, and other devices – a slow and expensive process that calls for the optical performance of the subassembly to be monitored while the position of each part is adjusted.

Kiyoshi Tanaka, a product manager at Sony’s semiconductor company, has made a sleeve that holds the focusing lens and the optical fiber into one unit and has mounted the LD and photodiode in a package. The front-end module, therefore, now contains the connector, lens, LD, PD, monitor PD, and transimpedance amplifier in a single unit. The result, he says, is a six-to-one simplification of the assembly that can be slapped together with the LD driver or the limiting amplifier using passive mounting.

“Our technology makes an all-in-one module – it suppresses manufacturing costs, it’s reliable, and it’s cost competitive,” Onouchi says.

“Sony is the world’s largest LD maker because of our optical pickups for our CD, DVD, and Playstation,” says Kenji Nagashima, general manager of Sony’s integrated device department, part of the company’s device solutions company. “We have been wanting to use this technology to build a bigger market for our LDs. Our members have researched this and the conclusion is that the optical communications area shows the biggest potential.”

Sony feels it’s a good time to muscle in because, while the Japanese market for modules is treading water, the company's market analysis predicts a 300,000 unit per month demand emerging – largely in the U.S. – from 2005, according to Nagashima. Sony wants at least 20 percent and is aiming for 30 percent of that market, he says.

So, Sony is commercializing two optical transceiver modules: a small form factor SAS-100A and a pluggable SAS-101A version, both of which can handle 1.25 Gbit/s for Gigabit Ethernet and 2.125 Gbit/s for Fibre Channel, running on the 850nm wavelength. The company started sampling both last December at ¥10,000 yen (US$82.25) per unit. And Sony will be shipping samples of 1310nm versions in the first quarter of 2004, says Tanaka.

Sony has impressive technology, but the jury’s still out on whether that’s enough to compete against existing players such as Agilent Technologies Inc. (NYSE: A), Finisar Corp. (Nasdaq: FNSR), and Infineon Technologies AG (NYSE/Frankfurt: IFX), according to Dennis Gallagher, equity research analyst at SoundView Technology Group.

“I am not at all sure how much of the value in a datacom transceiver is dictated by the alignment in the package,” he says. “Other companies have offered manufacturing efficiencies in subcomponents – e.g., LightPath Technologies Inc. [Nasdaq: LPTH] – and this has not driven enough of the total module's value to prove decisive in the end market; or the value has been fleeting.”

Gallagher has “very modest assumptions” about the 500m and shorter networking market and the price competition, and he questions why Sony should bother. Success requires a direct sales force, expensive marketing, and the level of investment and research needed to bring the product to market – which likely leads to a pretty low ROI. And, he asks, “Where does datacom fit with Sony's corporate directions? They may see these as an entrée to next-gen PON [passive optical network] transceivers – higher volume, consumer-like products – and these first products are mostly intended to get their feet wet in the market."

— Paul Kallender, special to Light Reading

Page 1 / 2   >   >>
metroshark 12/5/2012 | 12:23:58 AM
re: Sony to Play in Transceiver Market $82 for a GigE SFP? They must be smoking something. You can get these parts for less than $40 today from at least 10 different vendors.

If they want to make an impact in the market, their prices would need to be 50% of the average market price today. Not 200%!
ke6udl 12/5/2012 | 12:23:58 AM
re: Sony to Play in Transceiver Market This is great news for the FTTH fans like me.
Network optical interfaces on REAL appliances.
Note: They are not PON transceivers. PON won't meet mass market cost requirements.
Raymand 12/5/2012 | 12:23:53 AM
re: Sony to Play in Transceiver Market In my opinion, Sony's strategic interests are exactly aligned with this move. These interests go well beyond optical modules and they are tied in with the evolution of networks as a whole.

WHY?:

Sony & Microsoft want to dominate the home entertainment god box (PS3/4, X-box2/3).

The RBOCs (and specifically Verizon has clearly said) that it needs to drive towards FTTH and 20Mbps and faster services so as to provide converged services that can compete against cable operators.

The recent FCC ruling allows the RBOCs to not have to share their high speed infrastructure which opens the door for the RBOCs to make such investments. These investments will have to be possible under a sustainable level of CAPEX. There will not be another debt financed adventure.

NOTE 1: Sharing POTS going forward is acceptable since POTS will increasingly become an isolated bundle of services with declining demand. The RBOC's surely would rather not share POTS but by keeping this unbundling in place, it forces them to make strategic (presumably agressive) decisions.

NOTE 2: Presumably, DSL is not likely to support 20Mbps, but to the extent that it does is will simply be part of the delivery portfolio vs FTTH, wireless and other TBD.

As the cost of optics comes down, the delivery requirements for FTTH may simply become the fiber to a passive garage wall demarcation point with the SONY/MS/etc multimedia god boxes consolidating and providing the optical module and protocol delivery. (Much like today's POTS where you simply have a wire and every phone contains the necessary smarts interface and signal to the network.)

This now also gets to the larger issue of packetization of the networks which is also clearly underway. This trend is inexorable though it also will occur under a sustainable capex build out. The MPLS card is being played but the largest cards (electronic vs optical networks) have not and the RBOCs and IXCs (at least AT&T seem to be keeping their powder dry for now).

As the network operators hold off on adopting new technology while they work on enhanced control methods such as MPLS....new 3rd generation networking paradigms will mature and displace the future that many people are thinking is going to be Optical Circuit Switching.

Electronic Vs Optical Networks

-- Clearly optics are cheaper than electronics when dealing with high speed tranmission -- regenerators are not going to be re-introduced into the networks. However, optical circuit switching (i.e. OCS aka Next Generation Optical Networking) may be seeing its market window collapsing.

The reason is that OCS is not well suited to packetized networks with high levels of connection density and rapid (bursty) levels of spacial and temporal path demand. OCS likes to establish persistent paths that are slow to change. This limitation will make the cost of scaling an OCS network too high and its overall bandwidth efficiency too low when you require high numbers of high speed connections. OCS cannot statistically multiplex its resources. Electronics however, can rapidly support these fast demand changes and electronics are getting cheaper.

WHICH ONE WINS.....Electronics or Optics?

OPTICS WINS!....but it won't be OCS....nor will it be the long hoped for Optical Packet Switching. Instead, a third generation, packet optimized optical networking approach called Optical Burst Switching (OBS) and more specifically its MPLS compliant version Labeled Optical Burst Switching (LOBS) will win.

NOTE 3: Optical Packet Switching (OPS) optically replicates the adaptibility of electronics BUT its presently impossible. The impossible part is O-RAM (optical random access memory) and the impractical and expensive parts are optical-synchronization and optical header recognition.

NOTE 4: Optical Burst Switching, also optically replicates the adaptability of electronics AND it is presently very possible and very much cheaper than electronics or OCS. In essense, by keeping a single wavelength as OEO, you allow every other wavelength 100's to be optical and switch at electronic speeds. In some ways, OBS is very similar to a packet fast version of SS7 and POTS. At the network edge data headers and packets are aggregated into bursts. An OEO system signaling channel then sends a burst control packet along the MPLS (or GMPLS) path ahead of the burst. This control packet arranges (without acknowledgement) for the network to switch the burst (following a decreasing distance and offset time behind) though the network all optically. All OEO's are avoided except at burst assembly and at burst disassembly. The switched path is essentially a circuit with a temporal and spacial footprint that matches the burst, i.e. a burst length circuit.

There will be invited papers and presentations on OBS as well as a technical session at OFC in Atlanta. This networking paradigm is still maturing (much like Linux has had to mature) but it clearly has the legs to go the distance.

Cheers,

Everybody Loves Raymand



splitEndz 12/5/2012 | 12:23:52 AM
re: Sony to Play in Transceiver Market re: This is great news for the FTTH fans like me.


Don't think this 850nm transceiver has much impact on FTTH, but instead "fiber IN the home," or, distribution within MTUs. Fiber IN the home seems a pretty miniscule market--especially at 1-2.5Gbps. If they could meet these cost and performance goals with POF (plastic or polymer optical fiber), that would be interesting. POF is touted as being as easy to work with as splicing copper wire. Giving the consumer that sort of capability to add optical cables in their home as easily as copper would be a huge boost it would seem.

re: Note: They are not PON transceivers. PON won't meet mass market cost requirements.

Yep, especially in the operational scenario they'd be using this in.
splitEndz 12/5/2012 | 12:23:51 AM
re: Sony to Play in Transceiver Market re: The RBOCs (and specifically Verizon has clearly said) that it needs to drive towards FTTH and 20Mbps and faster services so as to provide converged services that can compete against cable operators.

Given the string of dozens of broken promises by the RBOCs over the decades (including promises of deploying FTTH in the early 90's and ISDN before that), I wouldn't bet any money on them following through on this "promise."

What has you so convinced they will follow through on this proactively, and in a meaningful way? The "threat from cable" just doesn't seem to cut it as a reason. They've cited this threat several times in the past and have failed in their video endeavors--why is it different now? They can't even deliver a competitive internet offering that people want, despite it being available to over 70% of their collective territories. Though it is fair to say that the threat (of internet) from cable is in fact what got them off their butts to deploy DSL. So I'd be interested in heairng people's reason that believe the RBOCs will deploy FTTH (also considering their financial position).

Raymand 12/5/2012 | 12:23:50 AM
re: Sony to Play in Transceiver Market I am not betting on specific promises from specific companies but rather I am infering a general conclusion from bits of data and what makes strategic sense. The rest is noise. Clearly, the road forward is unpaved and unpredictable but I think the the conclusions are valid.

Strategically, building out an FTTH plant, like building out a copper plant, is a barrier-to-entry to new competitors. Its hard to get there, but its good once you do get there - if your not forced to then share it. Until the FCC decision, it did not make sense to build anything.

I think we all agree however that there are no guarentees and that irrational decisions may carry the day or some number of days. Still, I believe the conclusion is unavoidable -- like death and taxes. Sort of a reverse Alice in Wonderland -- It doesn't matter which way you go...your going to end up at the same place.

Cheers,

Everybody Loves Raymand

ke6udl 12/5/2012 | 12:23:49 AM
re: Sony to Play in Transceiver Market 850nm COULD play a role, but requires an open mind. Corning reported doing GbE over Infinicor SX+ MMF using cheap 850nm VCSELs. That said, in most case the final drop cable wouldn't be anything like that length. More like 100 meters, making 10G possible. That should be enough bandwidth for all except those with the biggest ...
ke6udl 12/5/2012 | 12:23:48 AM
re: Sony to Play in Transceiver Market OBS? Why go there? We already have have what it takes to do it cost effectively as it is. Perhaps we should wait for Solitons to the home?
I think we optical folks need to keep our feet on the ground. Just because one can do something doesn't mean we should. Especially when "what it is offers no particular benefit, or its advantage can't be visualized. If Solitons and OBS are ever to play a role, let's leave that to the next few generations to develop. Right now moving from 56kbs to 100Mbps seems quite adequate, and DOES meet the necessary economics while providing at least a 10:1 upgrade potential.
ke6udl 12/5/2012 | 12:23:47 AM
re: Sony to Play in Transceiver Market Please add options for either 850nm or 1310nm 100Mbps transceivers on your:
HDTVs
PVRs
PS2
PCs
etc.

And don't forget the high-res video camera interface options so I can:
Make video calls to my grandchildren.
Do video conferencing from home for work and play.
Attend Web University classes.
etc.

Thank you for your support of the stagnant optics industry.

PS: Love my 36" Sony WEGA.
moose 12/5/2012 | 12:23:42 AM
re: Sony to Play in Transceiver Market The difference? For the first time, RBOC are actually LOSING voice lines. This will force them to be serious about looking for new revenue streams.

----------------------------------------------
splitEnds said:

Given the string of dozens of broken promises by the RBOCs over the decades (including promises of deploying FTTH in the early 90's and ISDN before that), I wouldn't bet any money on them following through on this "promise."

What has you so convinced they will follow through on this proactively, and in a meaningful way? The "threat from cable" just doesn't seem to cut it as a reason. They've cited this threat several times in the past and have failed in their video endeavors--why is it different now?
Page 1 / 2   >   >>
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE