Bitcoin Forum
February 26, 2020, 02:34:01 PM *
News: Latest Bitcoin Core release: 0.19.0.1 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: [Theoretical question] Could the whole BTCnetwork run on a handful of computers?  (Read 117 times)
Chironexxx
Member
**
Offline Offline

Activity: 141
Merit: 18


View Profile
January 29, 2018, 09:12:56 PM
Merited by DannyHamilton (2)
 #1

…or in other words: how would a massive loss of hashing power affect the network shortterm/longterm?

Hi!

This question came to my mind, when I read the headline that China might ban BTC mining and I started to think about the ramifications for the whole network. Since I don’t have any programming background, let me briefly elaborate on my thoughts:
If I understand the whole system correctly, the difficulty is adjusted every two weeks adjusting it in a way, that the blocktime stays at ~10mins. This means that a sudden loss of a large amount of global hashing power would lead to considerably higher blocktimes until the next difficulty adjustment. From then on, everything should be back to normal. Furthermore, it should be able to push this even further and run it on a handful of computers, if the difficulty is adjusted correctly to this presumed tiny global hashrate.

Obviously, I understand that a massive drop in hashpower would attract tons of new miners who want a larger piece of the cake and the global hashrate would drop shortly, but rise back up quickly. Hence, running the network on a small amount of computer defeates the purpose of decentralization, I know that. However, I had the feeling that this is a really good thought experiment for learning more about the BTC network.

I would really appreciate if more knowledgeable people could let me know if I’m on the right track or if I misunderstood something, or if I failed to take all important aspects into account.

Thanks, Chiro

nothing to see here!
100% First Deposit Bonus Instant Withdrawals Best Odds 10+ Sports Since 2014 No KYC Asked Play Now
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1582727641
Hero Member
*
Offline Offline

Posts: 1582727641

View Profile Personal Message (Offline)

Ignore
1582727641
Reply with quote  #2

1582727641
Report to moderator
Minnie1928
Member
**
Offline Offline

Activity: 140
Merit: 14


View Profile
January 29, 2018, 09:16:21 PM
Last edit: January 29, 2018, 09:27:03 PM by Minnie1928
Merited by Xavofat (1)
 #2

If all the computers in this world would focus only on the mining algorithm they still couldn't reach even that small percent that the Bitcoin network is offering. If you could swap somehow to CPU(general) instead of ASIC the electricity would have to cost trillions of dollars, euros(any strong FIAT) to match the number of calculated hashes per moment.

Bitcoin network has supreme power and it is unreplaceable right now!

▶▶  [ WHITEPAPER ]  ◀◀              i v y              ▶▶  [ JOIN US NOW! ]  ◀◀
Ivy Disrupts The US$5 Trillion Per Day Money Transfer Market
TELEGRAM   ●   FACEBOOK   ●    TWITTER    ●   YOUTUBE   ●   LINKEDIN
ManaMan
Member
**
Offline Offline

Activity: 238
Merit: 32


View Profile
January 29, 2018, 09:18:27 PM
 #3

…or in other words: how would a massive loss of hashing power affect the network shortterm/longterm?

Hi!

This question came to my mind, when I read the headline that China might ban BTC mining and I started to think about the ramifications for the whole network. Since I don’t have any programming background, let me briefly elaborate on my thoughts:
If I understand the whole system correctly, the difficulty is adjusted every two weeks adjusting it in a way, that the blocktime stays at ~10mins. This means that a sudden loss of a large amount of global hashing power would lead to considerably higher blocktimes until the next difficulty adjustment. From then on, everything should be back to normal. Furthermore, it should be able to push this even further and run it on a handful of computers, if the difficulty is adjusted correctly to this presumed tiny global hashrate.

Obviously, I understand that a massive drop in hashpower would attract tons of new miners who want a larger piece of the cake and the global hashrate would drop shortly, but rise back up quickly. Hence, running the network on a small amount of computer defeates the purpose of decentralization, I know that. However, I had the feeling that this is a really good thought experiment for learning more about the BTC network.

I would really appreciate if more knowledgeable people could let me know if I’m on the right track or if I misunderstood something, or if I failed to take all important aspects into account.

Thanks, Chiro


First things first miners are not the ones who host btc nodes, they might but most of them simply connect to 3rd party pool because  this is how you mine. Which means that if there is sudden drop in hash rate there shouldn't be a problem with the number of nodes that are there in the network.

You are correct we will have to wait for difficulty to adjust, but however knowing that this is multi billion industry I think that miners will see to get out of "China" in this case before they ban mining.

Other disadvantage with low hash rate is that it makes it "easier" for 51% attack to occur and this is why btc big hash rate matters so much - for the security of whole network.
DannyHamilton
Legendary
*
Offline Offline

Activity: 2338
Merit: 1718



View Profile
January 29, 2018, 09:32:03 PM
Merited by Xavofat (1)
 #4

If I understand the whole system correctly, the difficulty is adjusted every two weeks adjusting it in a way, that the blocktime stays at ~10mins. This means that a sudden loss of a large amount of global hashing power would lead to considerably higher blocktimes until the next difficulty adjustment.

It is only adjusted every 2 weeks if the blocks are happening on average every 10 minutes.

The ACTUAL trigger for the adjustment is that the block height be a multiple of 2016.  If the blocks are happening on average every 10 minutes, then that is:

2016 blocks X 10 minutes = 20160 minutes
20160 minutes / 1440 minutes per day = 14 days = 2 weeks.

However, lets imagine that 75% of the hash power were to just disappear immediately after a difficulty adjustment.  Since only one-fourth of the original amount of hash power exists, the blocks will take 4 times as long, or 40 minutes on average.  This would continue for 8 weeks, before the next adjustment would bring the average time back to 10 minutes per block.

Additionally, there is a limitation on the difficulty adjustment that prevents it from adjusting by more than a factor of 4.

Therefore, if more than 75% of the hash power were to just disappear immediately after a difficulty adjustment, the problem would persist even longer.  Lets imagine that 93.75% of the hash power were to just disappear immediately after a difficulty adjustment.  Since only one-sixteenth of the original amount of hash power exists, the blocks will take 16 times as long, or 1600 minutes on average.  This would continue for 32 weeks, before the next adjustment would bring the average time back to 40 minutes per block. This would continue for 8 weeks, before the next adjustment would finally bring the average time back to 10 minutes per block.


12tribes
Member
**
Offline Offline

Activity: 378
Merit: 11


View Profile
January 29, 2018, 09:40:24 PM
 #5

Yes. In theory that could be possible. but practically, that removes the 'decentralizing' quality/characteristic of cryptocurrency.

Bolt Brownie
Sr. Member
****
Offline Offline

Activity: 700
Merit: 350


Betting Championship betking.io/sports-leaderboard


View Profile
January 29, 2018, 10:08:39 PM
 #6


It is only adjusted every 2 weeks if the blocks are happening on average every 10 minutes.

The ACTUAL trigger for the adjustment is that the block height be a multiple of 2016.  If the blocks are happening on average every 10 minutes, then that is:

2016 blocks X 10 minutes = 20160 minutes
20160 minutes / 1440 minutes per day = 14 days = 2 weeks.


Thanks, didn't know it worked like this at all. I usually thought that it was actually based on time, and it was always adjusting difficulty every two weeks. From your example, we could see that it would take some time for everything to get back to normal, but nothing really special. Of course that if we were to consider that BTC had already global adoption, or at least it was largely used by people all over the world, then even 8 weeks would be a lot of time.

I don't think this will ever happen though, but couldn't this mean that mining farms could be potencial targets for terrorists attacks to cause global caos?
If BTC were to get global adoption, farms would need to be kept really secure.

Also, what about the 51% attack? I keep hearing about it, although I still don't understand it quite well. But with a low hash rate it would be easier for that to happen, so if we actually had a huge drop like that, wouldn't it be the end of the bitcoin network?
Chironexxx
Member
**
Offline Offline

Activity: 141
Merit: 18


View Profile
February 01, 2018, 09:04:50 PM
 #7

Hi!

Thanks for all your answers and sorry for my late reply! Yesterday, while I was typing my response, my computer blue screened on me, and never was able to recover from this hickup... Embarrassed..anyway...

I realize now, that my wording was not optimal and a bit naive to say the least!

If all the computers in this world would focus only on the mining algorithm they still couldn't reach even that small percent that the Bitcoin network is offering....

I never doubted that the BTC network is huge and stable...I was more asking if there is a reason, other than many people wanting a piece of the cake, that the BTC difficulty is so enormous!
 
Quote from: ManaMan
Other disadvantage with low hash rate is that it makes it "easier" for 51% attack to occur and this is why btc big hash rate matters so much - for the security of whole network.

and a 51% attack is exactly the one thing I completely forgot to consider! Thanks for pointing it out!

Quote from: DannyHamilton
The ACTUAL trigger for the adjustment is that the block height be a multiple of 2016.  If the blocks are happening on average every 10 minutes, then that is:

2016 blocks X 10 minutes = 20160 minutes
20160 minutes / 1440 minutes per day = 14 days = 2 weeks.

And wow, this was an amazing explanation! thanks! I really learned several new facts about the BTC network. Exactly the reason, why I have asked this question!

nothing to see here!
WhaleSlayer
Full Member
***
Offline Offline

Activity: 228
Merit: 100



View Profile
February 01, 2018, 09:37:47 PM
 #8

If all the computers in this world would focus only on the mining algorithm they still couldn't reach even that small percent that the Bitcoin network is offering. If you could swap somehow to CPU(general) instead of ASIC the electricity would have to cost trillions of dollars, euros(any strong FIAT) to match the number of calculated hashes per moment.

Bitcoin network has supreme power and it is unreplaceable right now!

But this also means that should the Chinese shut down overnight ALL mining centers which exist in China, the rest of the network would possibly not manage to solve the algorithm of the next block for ages, since the difficulty gets adjusted only every two weeks - IF AND WHEN a block is confirmed. This could freeze Bitcoin forever... right?

TWIST   ⧼   REVOLUTIONISING BLOCKCHAIN ACCESSIBILITY   ⧽   TWIST
║   twist.network   ║
        ANN  Twitter  |  WHITEPAPER  |  Discord  |  Medium
DannyHamilton
Legendary
*
Offline Offline

Activity: 2338
Merit: 1718



View Profile
February 01, 2018, 09:54:43 PM
 #9

since the difficulty gets adjusted only every two weeks - IF AND WHEN a block is confirmed.

That is not correct.

The difficulty gets adjusted ONLY once every 2016 blocks.

If the blocks are being solved on average approximately every 10 minutes (like they usually are), then that works out to 2016*10=20160 minutes.  20160 minutes is 336 hours which is 14 days which is 2 weeks.

That's why people say it is adjusted every 2 weeks, because under NORMAL circumstances that's how long it takes to complete 2016 blocks.

However, if the bitcoin network SUDDENLY lost so much hash power that blocks were only being solved on average once per day, then it could take anywhere from 1 day to 2016 days until the next adjustment (depending on how many blocks it had been since the last adjustment).

If the bitcoin network SUDDENLY lost so much hash power that blocks were only being solved on average once per week, then it could take anywhere from 1 week to 2016 weeks until the next adjustment (depending on how many blocks it had been since the last adjustment).

This could freeze Bitcoin forever... right?

It could stop confirmations for a very long time.  If that time was long enough, then there would probably be demands that a hard fork be created to adjust the difficulty.

WhaleSlayer
Full Member
***
Offline Offline

Activity: 228
Merit: 100



View Profile
February 01, 2018, 10:30:30 PM
 #10


It could stop confirmations for a very long time.  If that time was long enough, then there would probably be demands that a hard fork be created to adjust the difficulty.


Right. I didn't think at the hard fork option, but of course this is the perfectly obvious solution. Actually a big dropdown in hash power could happen also in the case of a solar flair strong enough to put out of orders electrical devices and computers, but in this case I guess mankind would have superior problems to deal with.

TWIST   ⧼   REVOLUTIONISING BLOCKCHAIN ACCESSIBILITY   ⧽   TWIST
║   twist.network   ║
        ANN  Twitter  |  WHITEPAPER  |  Discord  |  Medium
Pages: [1]
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!