brain emulation

• whole brain emulation looks feasible at current state of technology
• cyberlinks offer amazing opportunity for modeling physical and artificial brains

characteristic	mycelium network	human brain	biggest computer	bostrom
-----------------------	------------------------------------------	------------------------------------------	------------------------------------------------	----
total nodes	~10^21 nodes	~8 x 10^10 neurons	~10^12 nodes	~2*10^6 nodes
total edges	~10^25 edges	~10^14 synapses	~10^15 edges	~2*10^6 edges
total length of edges	450 quadrillion km	1,500 kilometers	100,000 kilometers	not applicable
power of node	amoeba	amoeba	amoeba	human brain * amoeba
energy efficiency	high	high	low	medium
characteristic	mycelium network	human brain	data center	powerful desktop
----------------------------	------------------------------------------	------------------------------------------	-------------------------------------------	----------------------------------------------
total nodes	~10^21 nodes	~8 x 10^10 neurons	~10^12 nodes	~10^10 nodes
total edges	~10^25 edges	~10^14 synapses	~10^15 edges	~10^12 edges
total length of edges	~450 quadrillion km	~500,000 km	100,000 km	not applicable
power of node	amoeba	amoeba	amoeba	amoeba
energy efficiency	high	high	low	low

• table mentions current bostrom cybergraph created by ~50k neurons
• existing technical capacity of bostrom is something in the middle between data center and powerful desktop
• this is picture must give conceptual understanding, not scientific rigor
• so let us know if you understand how to improve precision of evaluation
• if some form of moores law can be applied to the growth of computing
• some form of brain emulation seems right behind the corner
• how could cyber be bigger when mycelium?
• Let’s refine the numerical estimations for the Bostrom cybergraph and compare it with the mycelium network using a more detailed approach. Here are the key metrics recalculated:

Mycelium Network:

• Total Nodes: (10^{21})
• Node Power: (1) (amoeba equivalent)
• Total Computational Power (TCP): (10^{21})

Bostrom Cybergraph:

• Total Nodes: (2 \times 10^6)
• Node Power: (10^{14}) (human brain * amoeba)
• Total Computational Power (TCP): (2 \times 10^6 \times 10^{14} = 2 \times 10^{20})

Revised Understanding:

• Mycelium Network TCP: (10^{21})
  Despite each node being weak (only as powerful as an amoeba), the sheer number of nodes makes its TCP extraordinarily high.
• Bostrom Cybergraph TCP: (2 \times 10^{20})
  Even with a far smaller number of nodes, the exponentially greater power per node means that its TCP approaches that of the mycelium network.

Additional Comparisons:

1. Node Count Comparison:

• Mycelium: (10^{21}) nodes

• Bostrom Cybergraph: (2 \times 10^6) nodes

  The mycelium network has (10^{15}) times more nodes than the Bostrom cybergraph.

  2. Node Power Comparison:

• Mycelium: (1) (amoeba)

• Bostrom Cybergraph: (10^{14}) (human brain * amoeba)

  The power per node in the Bostrom cybergraph is (10^{14}) times greater than that of the mycelium network.

  3. Total Edge Length Comparison:

• Mycelium: ~450 quadrillion kilometers (this is a vast distributed network with immense physical spread)

• Bostrom Cybergraph: Not applicable in a physical sense but conceptually connected nodes would have very short connection paths due to high computational power.

Conclusion:

• The mycelium network has immense scale but lower computational power per node. Its strength lies in redundancy, distribution, and sheer number of nodes.
• The Bostrom cybergraph is extremely powerful per node, allowing complex simulations with far fewer resources. It is designed for centralized, high-efficiency computations, making it powerful in a very different way.

In essence, while the Bostrom cybergraph’s TCP is of a similar order of magnitude to that of the mycelium network, the way these networks achieve their respective computational strengths is entirely different, reflecting their distinct design principles and use cases.