Cary,
funny, the tunable laser mania is quite a memory from the past if I remember. This call "opticals" vs. semis is literally off the mark IMHO. It's also quite confusing because "opticals" is so broad, even broader than semis..
For example simple segmentation of "optical" is into "passive" and "active". PAssive is usually meant to concern activities, devices and materials to merely guide light, split it into wavelengths, combine them but not really to interact with it. Comprises of : simple elements, passive integrated assemblys, interconnection (connectors), mux-demux pieces, filters, gratings, etc..
Active usually involves changing a signal's domain from optical to electrical or vice versa (photodetectors, lasers, lightsources). Also the many interaction effects of light with other properties of matter and waves is covered here: acousto-optics, electro-optics etc..
Now my prediction is that passive optics will remain in the crapper longer. Why ? Easier to duplicate, no cleanrooms (barriers to entry) usually involved, many divergent "paths of technology evolution" possible, highly path-dependant. Even China is a strong contender in this segment..
In optics I like the "active" segments way better. What I like best is the area of overlap btw. active optics and semis, usually compound semiconductors like InP, InGaP, InGaAs, CdTe, etc.. For mere "processing" of those fast signals also other compounds like GaAs, InP, SiGe etc.. We have many contenders here like TQNT, VTSS, MLTC, RFMD, ANAD, JDSU, AGRa, ADCT, MSCC, OSIS etc..
This is important becuase the analyst didn't provide this important factor of distinction. I could think that hybrid and Monolithic (InP) "Integration" capabilities would become increasingly important also. Unlike semis, packaging could consume 50% of the cost of optical devices and elements.. Players here are BKHM and ANAD for example apart from GLW, JDSU, ADCT...
however I think it's too early to get into these plays just on their "cheapness". what I would like to see is stabilization of demand. Instead carrier after carrier gets problems. It'S like a phantom market where your customers die away. Awful... Watching great firms like FTHL, WCG, TSIX, STRX, GBLX, CONE and so many others literally fade away into Ch.11. Now even WCOM, Q, BRW and LVLT are cited having problems..
Active optics have another plus: they also partially address the upcoming 10GIGABIT-Ethernet standards (VCSEL, LEDs, lasers, photodiodes: PIN & APD). That will be important, could be more important than telecom applications.. Today's robust gauge of the manufacturing sector (+1.1%) might be an early indication of things to come.. Maybe networking could be the best investment now (EXTR, FDRY, MRVC, ETS, JNPR)
rgrds
CROSSY |
