Optical computing systems are computers that use photons to process information rather than electrons via optical transistors, rather then electronic transistors which most modern systems use. Optical transistors use light, AKA electromagnetic waves, or photons, to convey information (the on or off state) through a processor, where electronic transistors use the movement of chains of electrons. Optical transistors have the potential to operate at a much greater speed then conventional electronic transistors and use less power, although as of 2017 they still face enormous technological and economic challenges. There is much controversy over how viable the technology truly is.
Technology
Optical computing does not currently refer to a particular way of doing optical computation or a particular type of optical transistor. There are hundreds of ways that thousands of researchers are attempting to make optical computing a viable alternative to electronic computing, the only thing that unifies them is that they all use electromagnetic waves as the means of information transfer. For example, some processors use light emitting diodes as the means of emittence and then the transistors consist of materials which can be made reflective to allow the signal to pass on, or a non reflective state to stop it. Other ideas involve the transistors themselves using fluorescent chemicals to emit light when both inputs are positive, or more ambitious designs use changes in wavelengths of light as a result of their interaction to create gates.
Transistors:
An example of the design of an optical transistor. (Author: Michelle Copyrights: All rights waived)
The most fundamental part of any computational device is the transistor. The transistor can be most simply described a a switch, or gate. So long as it is receiving signal from one input, it will allow or disallow a signal to pass through on another. Computer scientists can develop vast and sophisticated chains of logic to allow incredibly complicated mathematics to be executed, which when ordered the right way by a programmer allows people to interact with computers to solve problems, record and access information, and simulate deterministic systems. The primary limitations of a computers power is the speed at which the transistors may operate and the number of transistors it uses. Modern electronic computers work at about 2 gigahertz, which means they fire on average about 2000 times per second. They are densely packed with millions of transistors which execute huge swaths of instructions at the same time, allowing them to do the incredible math for things like physics simulations, or light rendering. Optical processors have the potential to fire much much faster since the speed of light is as fast as matter can travel in the universe, In electronic circuit boards made of Polyimide material, the signal velocity is typically about 16.3 cm per nanosecond which is approximately 54% of the speed of light, that is simply the raw speed of movement not including interactions with transistors which can take much longer for electrons to interact with then electromagnetic photons (4). Further, one of the major limitations of electronic computing is that electrons are particularly susceptible to a phenomena known as "Quantum Tunneling" in which when transistors get too small they lose the ability to prevent electrons from passing through them, the electrons act as though they were not there. Modern processors are very near this limit and it is believed to be the stopping point of Moore's law. It is possible that optical computers may be much less susceptible to this phenomena and be capable of becoming much smaller.
Practical Challenges and Limitations, Controversy
At the moment, it is highly debated among computer scientists whether optical computing is a technology worth pursuing at all. Opinions range from believing that it has greater potential then even technologies such as quantum computing, to it being simply to expensive for potential marginal improvement when modern technologies are much more economically viable. The issue is that is its largely unknown just how much of an improvement optical components truly offer. Current technologies are extremely power hungry, are far less dense, and not much faster then electronic components, many of which issues may prove to be insurmountable. Meanwhile alternative technologies, namely electronic computing is fast, well developed, and inexpensive. Optical computers require a great degree of precision to achieve any sort of density, and unlike electronic circuitry may be exponentially more expensive to create more dense chips. The primary challenge to this technology is that there is an already reliable and effective alternative. That said, it may still be very useful for niche applications even if some of these challenges can not be solved. For example, it may be relatively inexpensive to build optical processor so long as transistor density is not an issue, so in applications where space does not matter such as that of servers it may be very viable to build large but efficient optical computers.
Works Cited
Hanson, Jack. “Optical Computing.” Optical Computers, Future For All, July 2017, futureforall.org/computers/opticalcomputers.htm. Accessed 24 July 2017.
Kadiri, K. O., et al. “ Optical Computing: An Overview.” Academia.edu, Optical Computing: An Overview, 2014, www.academia.edu/11501562/Optical_Computing. Accessed 21 July 2017.
Larry Hardesty | MIT News Office. “Toward Optical Chips.” MIT News, MIT, 17 Sept. 2014, news.mit.edu/2014/optical-chips-tunable-light-source-0917. Accessed 24 July 2017
Clayton R. Paul, Analysis of Multiconductor Transmission Lines. Johm Wiley & Sons., New York (1994)
Witlicki, Edward H.; Johnsen, Carsten; Hansen, Stinne W.; Silverstein, Daniel W.; Bottomley, Vincent J.; Jeppesen, Jan O.; Wong, Eric W.; Jensen, Lasse; Flood, Amar H. (2011). "Molecular Logic Gates Using Surface-Enhanced Raman-Scattered Light". J. Am. Chem. Soc.133 (19): 7288–91. doi:10.1021/ja200992x.
Optical Computing
Optical computing systems are computers that use photons to process information rather than electrons via optical transistors, rather then electronic transistors which most modern systems use. Optical transistors use light, AKA electromagnetic waves, or photons, to convey information (the on or off state) through a processor, where electronic transistors use the movement of chains of electrons. Optical transistors have the potential to operate at a much greater speed then conventional electronic transistors and use less power, although as of 2017 they still face enormous technological and economic challenges. There is much controversy over how viable the technology truly is.Technology
Optical computing does not currently refer to a particular way of doing optical computation or a particular type of optical transistor. There are hundreds of ways that thousands of researchers are attempting to make optical computing a viable alternative to electronic computing, the only thing that unifies them is that they all use electromagnetic waves as the means of information transfer. For example, some processors use light emitting diodes as the means of emittence and then the transistors consist of materials which can be made reflective to allow the signal to pass on, or a non reflective state to stop it. Other ideas involve the transistors themselves using fluorescent chemicals to emit light when both inputs are positive, or more ambitious designs use changes in wavelengths of light as a result of their interaction to create gates.Transistors:
The most fundamental part of any computational device is the transistor. The transistor can be most simply described a a switch, or gate. So long as it is receiving signal from one input, it will allow or disallow a signal to pass through on another. Computer scientists can develop vast and sophisticated chains of logic
to allow incredibly complicated mathematics to be executed, which when ordered the right way by a programmer allows people to interact with computers to solve problems, record and access information, and simulate deterministic systems. The primary limitations of a computers power is the speed at which the transistors may operate and the
number of transistors it uses. Modern electronic computers work at about 2 gigahertz, which means they fire on average about 2000 times per second. They are densely packed with millions of transistors which execute huge swaths of instructions at the same time, allowing them to do the incredible math for things like physics simulations, or light rendering. Optical processors have the potential to fire much much faster since the speed of light is as fast as matter can travel in the universe, In electronic circuit boards made of Polyimide material, the signal velocity is typically about 16.3 cm per nanosecond which is approximately 54% of the speed of light, that is simply the raw speed of movement not including interactions with transistors which can take much longer for electrons to interact with then electromagnetic photons (4). Further, one of the major limitations of electronic computing is that electrons are particularly susceptible to a phenomena known as "Quantum Tunneling" in which when transistors get too small they lose the ability to prevent electrons from passing through them, the electrons act as though they were not there. Modern processors are very near this limit and it is believed to be the stopping point of Moore's law. It is possible that optical computers may be much less susceptible to this phenomena and be capable of becoming much smaller.
Practical Challenges and Limitations, Controversy
At the moment, it is highly debated among computer scientists whether optical computing is a technology worth pursuing at all. Opinions range from believing that it has greater potential then even technologies such as quantum computing, to it being simply to expensive for potential marginal improvement when modern technologies are much more economically viable. The issue is that is its largely unknown just how much of an improvement optical components truly offer. Current technologies are extremely power hungry, are far less dense, and not much faster then electronic components, many of which issues may prove to be insurmountable. Meanwhile alternative technologies, namely electronic computing is fast, well developed, and inexpensive. Optical computers require a great degree of precision to achieve any sort of density, and unlike electronic circuitry may be exponentially more expensive to create more dense chips. The primary challenge to this technology is that there is an already reliable and effective alternative. That said, it may still be very useful for niche applications even if some of these challenges can not be solved. For example, it may be relatively inexpensive to build optical processor so long as transistor density is not an issue, so in applications where space does not matter such as that of servers it may be very viable to build large but efficient optical computers.Works Cited