Network Photonics Inc. closed its doors yesterday, opting to give its remaining $30 million in cash back to investors rather than burn it waiting for a market recovery (see Network Photonics Closes).
Based in Boulder, Colo., the company had 66 employees on its last day.
CEO Steve Georgis sent a note to analysts yesterday, explaining that the company didn't want to wait in vain for a market recovery. ”It has become obvious that the market for photonic switching and reconfigurable optical networks is not going to emerge in any big way soon. When it does emerge, we estimate that the size of the addressable market will be small," he wrote.
Given that outlook and the substantial money the company raised (see Network Photonics Raises $106.5 Million), Network Photonics didn't appear likely to recoup. "We couldn't come up with a good market scenario that was going to give a good return on investment," Georgis said in an interview.
Network Photonics started out building an all-optical switch for metro access networks. As the market bubble deflated, those plans were scaled back, but the company did manage to produce a subsystem in early 2002 (see Network Photonics Scales Back and Network Photonics: A Corvis Copy Cat?).
As with most all-photonic switching plays, Network Photonics based its technology on micro-electro-mechanical systems (MEMS), using tilting mirrors to shunt light to the appropriate ports. But the company's CrossWave system added a twist: a prism-like filter that split a WDM stream into component wavelengths.
This allowed Network Photonics to use just one row of mirrors, one per wavelength. Other MEMS hopefuls -- the still-surviving Calient Networks Inc. among them -- used square grids of mirrors, and the most ambitious startups allowed those mirrors to swivel in arbitrary directions, the so-called "3D MEMS" approach. Network Photonics claimed its way was cheaper (fewer mirrors) and easier to control.
Larger optical MEMS switches never took off, however -- a trend most dramatically marked by the recent closure of OMM Inc. (see OMM: The End Is Near). Now, Network Photonics' time has come.
Georgis and a skeleton staff will remain to sell off the company's assets. That so much technology might never resurface -- not just Network Photonics', but the industry's in general -- is going to be one of the "tragedies" of the recession, he says.
"A lot of breakthrough R&D has been done [in the industry as a whole], and a lot of that is going to just go away," Georgis says. "A couple of years from now, people might look back and say, 'What happened to Network Photonics? They had a product we needed.'"
re: Network Photonics Shuts Down GÇ¥It has become obvious that the market for photonic switching and reconfigurable optical networks is not going to emerge in any big way soon. When it does emerge, we estimate that the size of the addressable market will be small,"
One major misconception within our industry (especially among VCGÇÖs) was that DWDM was the right technology at the right time that solved the important problem of fiber exhaust. The reality is that except for a few highly celebrated initial cases (the creation and success of Ciena notwithstanding), the majority of DWDM deployment was not to alleviate fiber exhaust at all. Instead, the transition was more about EDFA, which was extremely economically compelling compared to OEO 3-R regeneration.
As a result, many of the long-haul DWDM deployment that we have seen in the last few years which fueled the telecom bubble were deployed with just a few wavelengths (some were financially justifiable even with one wavelength). As a result, there are pretty of unlit wavelengths in the ground and as the industry continues to rationalize and eventually starts to recover, the carriers will only need to buy more line cards with limited opportunities for new systems or new architectures.
On the other hand, since there is little need for amplification in metro, the economic driver for metro DWDM will be based purely on the potential for fiber exhaust which would not be a problem for a very long time except in a few unique venues. Therefore the opportunity for GÇ£ReconfigurableGÇ¥-OADM is even further out since not only do we have to justify the need for multiple wavelengths but also the need for re-configurability.
In the last twelve months, we have seen many MEMS or non-MEMS component or subsystem companies shutting down, some of which have been reported and most have disappeared quietly.
The first wave of companies were those that never managed to develop a product.
The second wave were those that have shippable products but went out of cash.
Network Photonics and Transparent Network (both are USVP portfolio companies) are the beginning of the third wave which have products AND cash but the investors decided that they would be better served by taking back the cash, because they finally recognize that GÇ£the market is not going to emerge anytime soon and when it does, it will be small.GÇ¥
Free market capitalism is brutal but it works and like Clint Eastwood once said in one of his movies GǣDeserve has nothing to do with it GǪ Gǣ
re: Network Photonics Shuts Down It looks like that Network Photonics is following Onix Microsystems' model of giving back remaining cash to investors and throwing in the towel. The management must have been forced by investors to give back the cash instead of giving up willingly. It's surprising that they still have $30 million three years after the last financing round of $106 million.
Network Photonics' technology should be extremely difficult to make, since it relies on a big lens to transform all incoming beams. In the end, they can only make 4 or 8 channel WDM mux/demux. It's pathetic.
If there is any lesson to be learned, one of them may be that elegant device design does not automatically translate into successful product. Network Photonics is probably killed by the near impossibility of device alignment and very high fiber insertion loss, among other things.
re: Network Photonics Shuts Down The word is that Network Photonics had a shot at using their OADM engine in combination with xtera's raman long haul, however the board decline the proposed merger...
re: Network Photonics Shuts Down "The reality is that except for a few highly celebrated initial cases (the creation and success of Ciena notwithstanding), the majority of DWDM deployment was not to alleviate fiber exhaust at all. Instead, the transition was more about EDFA, which was extremely economically compelling compared to OEO 3-R regeneration."
__________
y2k
There is a whole new round of long haul cost reductions out there ready to start deployment. Just like the EDFA based multi wavelength systems prove in at just a few wavelengths in comparison to single channel systems, so do extended long haul systems prove out over short haul systems. As you say - it is extremely economically compelling.
Deploying a 1500 km 3R system to replace a 500 km 3R system with 2 OEO regens pays off with only a few incremental channels.
Vendors gave carriers low common system pricing during the bubble years to get them into their products with the trade-off being higher cost transponder and regen cards. 3R regens are not cheap.
Each incremental channel added costs a fortune. Once channel and regen card inventory drops to low enough levels (already a problem in a few networks), the best-cost solution is Extended Long Haul overbuild.
No capital budget this year for any long haul in carrier network, but 2004 is goin to be better than most people think.....
re: Network Photonics Shuts Down Just for the record...
Managment had several opportunities to keep "milking" the $30M down, but didn't think it was wise. The investors and Management reached a joint decision.
4 to 8 lambda's is totally wrong, Try 96 lambda's with complete mux/demux/switch in one device.
As for a big insertion loss, I guess that's true...if you consider 4dB (fiber to fiber) to be big.
As for allignment, how about flat performance form 0 to 70C, .2 PDL, 45dB of channel isolation, 38GHz of bandbass on 50GHz centers.
re: Network Photonics Shuts Down Problem with NP was not technology. They were doing good progress on that front but inability of management to decide what to do. They were a system company to start then a sub-system company then trying to roll-up with Xtera to become a sytems company again. No leadership, no vision, burning money like crazy, why would the investors trust the CEO with the 30M left in the bank?
re: Network Photonics Shuts Down Deer_in_the_light writes:
"Problem with NP was not technology. They were doing good progress on that front but inability of management to decide what to do. They were a system company to start then a sub-system company then trying to roll-up with Xtera to become a sytems company again. No leadership, no vision, burning money like crazy, why would the investors trust the CEO with the 30M left in the bank?"
And if they had the "leadership" you claim they didn't have do you think they would be a smashing success now?
Network made these moves because they were trying to adapt to a rapidly changing market. They were unsuccessful because the market they were in crashed.
I'm not affiliated with NP. I'm just tired of all the blame I read on this site.
I've been expecting this announcement for some time. I could be wrong, but I always saw the NP switch as a MEMS version of the Corning PurePath, namely, a wavelength selective add/drop. Corning got a good product out quickly and ended up dropping it completely because there was no market. If they couldn't sell a device I wondered how anyone could. It seems like only JDS is left in this particular sector.
I sure would like to see a summary of startups and failures.
You have a point on NP being a victim of the market, I agree that the management team cannot be blamed for not anticipating the downturn. I still blame them for having burned $80M!!! and having very little to show for it. No qualified product, no customers, no revenues, end of story.
On the market side, I do not think that JDSU has anything like a 1xN wavelength selective switch. The only wavelength selective product they have is their wavelength blocker, see link below
re: Network Photonics Shuts Down If JDSU doesn't have a wavelength selective switch now, they could have one soon since the waveblocker is essentially such a switch allready.
In regards to all the other MEMS switches made my Tellium, Xros, OMM and the likes: is it correct to say these were never to be wavelength selective? Does each MEMS mirror switch an entire set of channels?
To save this item to your list of favorite Light Reading content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
(15) | 
Digg
Del.icio.us
Reddit
Tweet This