The human brain has about 100 billion neurons and 100 trillion synapses.
The brain emulation roadmap is a key reference.
It discusses how much more compute power might be needed to emulate a human brain if more modeling is required beyond neurons and synapses to ion channels and proteins.
At a whole brain emulation workshop it was estimated that 5×5×50 nm scanning resolution would be needed. It may also be possible to use embedded processor technology to manufacture large amounts of dedicated hardware relatively cheaply. A study of high resolution climate modelling in the petaflop range found a 24‐ to 34‐fold reduction of cost and about two orders of magnitude smaller power requirements using a custom variant of embedded processor chips (Wehner,Oliker et al., 2008).
If 10**19 operations per second is a likely upper bound to simulate a human mind then we should be able to achieve that with irreversible computing. If we need up to a trillion times more irreversible computing.
Future Computing systems
Exaflop (10^18 operations per second) class supercomputers should be here by 2015-2017. Distributed [email protected] systems should achieve exaflops around 2012-2013.
Memristors will help with faster memory and computing and possibly with AI. Memristor seems to be a better analog model of a synapse.
Two memristors can be used for an analog simulation of a synapse. 200 trillion memristors would be a human level synapse system. Memristors can go 1-5 nanometers in the form of nanowires. Hundreds of trillions of memristors would be compact.
Memristors could be the key to making analog synapse networks at human brain scale or just as a way to make a 3D computer structure that is near nanoscale is structure that has localized computing and memory.
Memristors, optical computing could take us to the limits of irreversible computing and help us take advantage of reversible computing.
Irreversible computing seems to be limited to about 3.5 X 10*20 operations per watt. A wide variety of proposed reversible device technologies have been analyzed by physicists. With theoretical power-performance up to 10-12 orders of magnitude better than today’s CMOS. A likely minimum reversible computing capability for an advanced civilization is 10**29 operations per watt.
Yottaflops at 10^24 operations per second seem reachable using memristors and highly efficient optical computing and onchip communication. Memristors could also be used for 3 dimensional computing systems.
If you have a computing system that has the capability for human level computation then how would you transfer consciousness to it ?
Consciousness transfer – What I think might work. Need to perfect mind machine interfaces and have brain prosthetics and integrated co-processors.
Then if you have your wet mind and the computer mind extension working together for a decade and the computer appliance had the same or higher level of complexity the result could be your consciousness might exist and span the substrates (both in the brain and in the artificial systems). So if you lost the biological substrate then you could have consciousness that continues in the machine portion.