Aiming to become the global leader in chip-scale photonic solutions by deploying Optical Interposer technology to enable the seamless integration of electronics and photonics for a broad range of vertical market applications

Free
Message: It's raining, no golf :-( but a good time to check this silicon/POET comparison

A little reminder of why we must be patient

• POET technology is applicable to a large portion of the total semiconductor market ($430 billion

2015 Forecast) as it is possibly the most comprehensive solution yet to increasing semiconductor tle reminder why we

performance in an economic and functional manner

• In the near term, POET could potentially address the following markets:

• Optical semiconductor ($37.4 billion 2015 Forecast)

• Sensors and actuators ($14.1 billion 2015 Forecast)

• Analogue ICs ($55.9 billion 2015 Forecast)

• Discrete semiconductor ($28.6 billion 2015 Forecast)

• In the longer term, POET could potentially address the following markets:

• Logic ($114 billion 2015 Forecast)

• MPUs and MCUs ($92.6 billion 2015 Forecast)

• Memories ($86.6 billion 2015 Forecast)

• Quantum computing – Although quantum computing is in its infancy, POET can support

quantum computing applications since it can support fabrication of quantum dot-based spin

qubits and the devices needed to read and write them on the same die in the

semiconductor Market

Forecasts (POET Markets)• CMOS Silicon

• Final CMOS Si geometry (10/11 nm) is under development now; first production 2015?

• POET/PET offer about 100x speed improvement over CMOS silicon

• POET/PET offer 10-100x power efficiency improvement over CMOS silicon

• OE Conversion

• As example, current 10 Gigabit Ethernet transceivers use about 10 individually packaged

ICs on a substrate in a die-cast housing; POET can reduce this to 1 individually packaged

IC

• Depending on application, POET can reduce overall OE transceiver cost by 60 to 90%

• Memory and Storage

• Current memory types include dedicated SRAM, DRAM, and NVRAM devices

• POET/PET memory cell can concurrently support all three memory types

• Massive simplification at system level due to elimination of NVRAM backup/recovery

• Much lower bit error rates than silicon-based memories (several orders of magnitude)

• Sensors and Weapons

• POET provides low-cost optical thyristor arrays that can be used as dual-mode sensor/laser

arrays (same panel can find targets and destroy them)

• Main reason US Government funded research for so long (19 years and counting…)

• CMOS Silicon

• Final CMOS Si geometry (10/11 nm) is under development now; first production 2015?

• POET/PET offer about 100x speed improvement over CMOS silicon

• POET/PET offer 10-100x power efficiency improvement over CMOS silicon

• OE Conversion

• As example, current 10 Gigabit Ethernet transceivers use about 10 individually packaged

ICs on a substrate in a die-cast housing; POET can reduce this to 1 individually packaged

IC

• Depending on application, POET can reduce overall OE transceiver cost by 60 to 90%

• Memory and Storage

• Current memory types include dedicated SRAM, DRAM, and NVRAM devices

• POET/PET memory cell can concurrently support all three memory types

• Massive simplification at system level due to elimination of NVRAM backup/recovery

• Much lower bit error rates than silicon-based memories (several orders of magnitude)

• Sensors and Weapons

• POET provides low-cost optical thyristor arrays that can be used as dual-mode sensor/laser

arrays (same panel can find targets and destroy them)

• Main reason US Government funded research for so long (19 years and counting…)

Four Examples

Share
New Message
Please login to post a reply