Difference between revisions of "Bitcoin Wiki:Proof of work change hard fork"

From Bitcoin Wiki
Jump to: navigation, search
(Proof of work change hard fork)
 
(Proof of work change hard fork)
Line 1: Line 1:
Purpose of proof of work
+
== Proof of work change hard fork ==
To burn electricity in a way that is as efficient as humanly possible and easily verified.
 
This shows how much cost was involved in presenting a set of transactions as valid.
 
If that cost is high enough we can trust that the network was in agreement and that the cost of attempting an invalid transaction was also very high.
 
Because “blocks of transactions are chained together we can be more confident in older blocks that have had additional investment in electricity burn on top of them.
 
Problems we want to solve
 
Miner centralization
 
Accelerate mining hardware commoditization.
 
Proposal to change PoW unexpectedly (one time).
 
This would cause all existing investors in mining hardware to suffer financial loss.
 
The more recent the investment the greater the loss because a recent investment would not have produced any income yet.
 
For example if someone invested 10B in a new chip design, but had not sold a single chip, that entire investment would be lost with a PoW change.
 
If someone invested 10B in in a chip design a year ago and had already earned 10B on that design, they would be at a break even.
 
  
Goals
+
Unexpectedly changing the proof of work algorithm has been suggested as a solution to ASIC manufacturing centralization (a lack of progress towards mining hardware commoditization). It is suggested that by changing the PoW existing hardware manufacturers will suffer great financial loss and this will "make room" for additional competition. In actuality, likely profits, attracts additional competition. By increasing the risk involved in making mining hardware, by showing that their R&D investments could be completely lost by a PoW change, competitors will be more discouraged to make the required investment. This document is an attempt to make these simple, and universal, economic truths clear in this particular application.
Increase commoditization of ASIC hardware.
 
Hard fail. Opposite effect - less competition is the obvious result.
 
To increase the decentralization of mining. More miners spread out in more geographic locations owned and controlled by more individuals.
 
Hard fail. With less commoditization of ASICs, a greater ability for a bad actor to invest in ASICs and dominate the hash rate will not be decentralized.
 
If the PoW was changed every month we *could* get less centralization, but at the cost of being completely open to attack by anyone willing to invest in the unprofitable business of specialty hardware.
 
Long term counterproductive effects on ASIC manufacturing
 
Companies that were about to invest in ASIC manufacturing would realize they would have suffered a greater loss than any other potential investment had they moved sooner.
 
100% loss on a chip design is unusual.
 
Anyone considering entrance into ASIC manufacturing would need to expect much higher profits than before to compensate for the risk that this would happen again.
 
Anyone considering entrance into ASIC manufacturing would need to be more capitalized in order to be able to weather another proof of work change without being financially destroyed.
 
Competition among ASIC manufacturing would be reduced - Mining hardware creation would be more centralized
 
The rapid march towards commoditization, created by competition, would be slowed.
 
Bitcoin would be more vulnerable to an attacker investing into ASIC hardware for the purpose of attacking the network.
 
  
  
 +
=== Purpose of proof of work recap ===
  
Short term counterproductive effects on bitcoin mining
+
* The purpose of proof of work is to burn electricity in a way that is as efficient as humanly possible and easily verified.
The amount of electricity spent at the height of human efficiency would drop to 0%. This is the hash rate that matters.
+
* This shows how much cost was involved in presenting a set of transactions as valid. If that cost is high enough we can trust that the network was in agreement and that the cost of attempting an invalid transaction was also very high.
All existing bitcoin miners would be poorer in proportion to the newness of their equipment. Miners running equipment near end of life (bitmain?) would be affected the least.
+
* Because “blocks of transactions are chained together we can be more confident in older blocks that have had additional investment in electricity burn on top of them.
Supporters of bitcoin would need to mine using inefficient equipment.
+
 
An attacker, such as a corrupt government, could dominate the network by making an investment in ASICs that a private company would consider to risky (unprofitable).
+
 
 +
=== Problems we want to solve through an unexpected PoW change ===
 +
 
 +
* The goal is to accelerate mining hardware commoditization.
 +
* By changing the PoW all existing investors in mining hardware would suffer financial loss.
 +
* The more recent the investment the greater the loss because a recent investment would not have produced any income yet.
 +
* For example if someone invested 10B in a new chip design, but had not sold a single chip, that entire investment would be lost with a PoW change.
 +
* If someone invested 10B in in a chip design a year ago and had already earned 10B on that design, they would be at a break even.
 +
 
 +
=== Problems solved through an unexpected PoW change ===
 +
* Increase commoditization of ASIC hardware.
 +
**Hard fail. Opposite effect - less competition is the obvious result.
 +
* To increase the decentralization of mining. More miners spread out in more geographic locations owned and controlled by more individuals.
 +
** Hard fail. With less commoditization of ASICs, a greater ability for a bad actor to invest in ASICs and dominate the hash rate will not be decentralized.
 +
* If the PoW was changed every month we *could* get less centralization, but at the cost of being completely open to attack by anyone willing to invest in the unprofitable business of specialty hardware.
 +
 
 +
=== Long term counterproductive effects on ASIC manufacturing ===
 +
* Companies that were about to invest in ASIC manufacturing would realize they would have suffered a greater loss than any other potential investment had they moved sooner.
 +
** 100% loss on a chip design is an unusually bad outcome.
 +
* Anyone considering entrance into ASIC manufacturing would need to expect much higher profits than before to compensate for the risk that this would happen again.
 +
* Anyone considering entrance into ASIC manufacturing would need to be more capitalized in order to be able to weather another proof of work change without being financially destroyed.
 +
* Competition among ASIC manufacturing would be reduced - Mining hardware creation would be more centralized
 +
* The rapid march towards commoditization, created by competition, would be slowed.
 +
* Bitcoin would be more vulnerable to an attacker investing into ASIC hardware for the purpose of attacking the network.
 +
 
 +
=== Short term counterproductive effects on bitcoin mining ===
 +
* The amount of electricity spent at the height of human efficiency would drop to 0%. This is the hash rate that matters.
 +
* All existing bitcoin miners would be poorer in proportion to the newness of their equipment. Miners running equipment near end of life would be affected the least.
 +
* Supporters of bitcoin would need to mine using inefficient equipment.
 +
* An well funded attacker, such as a corrupt government, could dominate the network by making an investment in ASICs that a private company would consider to risky (unprofitable).

Revision as of 01:01, 8 March 2018

Proof of work change hard fork

Unexpectedly changing the proof of work algorithm has been suggested as a solution to ASIC manufacturing centralization (a lack of progress towards mining hardware commoditization). It is suggested that by changing the PoW existing hardware manufacturers will suffer great financial loss and this will "make room" for additional competition. In actuality, likely profits, attracts additional competition. By increasing the risk involved in making mining hardware, by showing that their R&D investments could be completely lost by a PoW change, competitors will be more discouraged to make the required investment. This document is an attempt to make these simple, and universal, economic truths clear in this particular application.


Purpose of proof of work recap

  • The purpose of proof of work is to burn electricity in a way that is as efficient as humanly possible and easily verified.
  • This shows how much cost was involved in presenting a set of transactions as valid. If that cost is high enough we can trust that the network was in agreement and that the cost of attempting an invalid transaction was also very high.
  • Because “blocks of transactions are chained together we can be more confident in older blocks that have had additional investment in electricity burn on top of them.


Problems we want to solve through an unexpected PoW change

  • The goal is to accelerate mining hardware commoditization.
  • By changing the PoW all existing investors in mining hardware would suffer financial loss.
  • The more recent the investment the greater the loss because a recent investment would not have produced any income yet.
  • For example if someone invested 10B in a new chip design, but had not sold a single chip, that entire investment would be lost with a PoW change.
  • If someone invested 10B in in a chip design a year ago and had already earned 10B on that design, they would be at a break even.

Problems solved through an unexpected PoW change

  • Increase commoditization of ASIC hardware.
    • Hard fail. Opposite effect - less competition is the obvious result.
  • To increase the decentralization of mining. More miners spread out in more geographic locations owned and controlled by more individuals.
    • Hard fail. With less commoditization of ASICs, a greater ability for a bad actor to invest in ASICs and dominate the hash rate will not be decentralized.
  • If the PoW was changed every month we *could* get less centralization, but at the cost of being completely open to attack by anyone willing to invest in the unprofitable business of specialty hardware.

Long term counterproductive effects on ASIC manufacturing

  • Companies that were about to invest in ASIC manufacturing would realize they would have suffered a greater loss than any other potential investment had they moved sooner.
    • 100% loss on a chip design is an unusually bad outcome.
  • Anyone considering entrance into ASIC manufacturing would need to expect much higher profits than before to compensate for the risk that this would happen again.
  • Anyone considering entrance into ASIC manufacturing would need to be more capitalized in order to be able to weather another proof of work change without being financially destroyed.
  • Competition among ASIC manufacturing would be reduced - Mining hardware creation would be more centralized
  • The rapid march towards commoditization, created by competition, would be slowed.
  • Bitcoin would be more vulnerable to an attacker investing into ASIC hardware for the purpose of attacking the network.

Short term counterproductive effects on bitcoin mining

  • The amount of electricity spent at the height of human efficiency would drop to 0%. This is the hash rate that matters.
  • All existing bitcoin miners would be poorer in proportion to the newness of their equipment. Miners running equipment near end of life would be affected the least.
  • Supporters of bitcoin would need to mine using inefficient equipment.
  • An well funded attacker, such as a corrupt government, could dominate the network by making an investment in ASICs that a private company would consider to risky (unprofitable).