Bitcoin Forum
December 17, 2017, 12:35:27 PM *
News: Latest stable version of Bitcoin Core: 0.15.1  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 [2] 3 4 »  All
  Print  
Author Topic: If Core had to hard fork and use another mining algorithm, what would it be?  (Read 3155 times)
Carlton Banks
Legendary
*
Offline Offline

Activity: 1848



View Profile
September 19, 2017, 07:21:06 PM
 #21

Yeah, because super-majority miner activated hard-forks away from the proven development team, with zero technical justification == letting people choose


Done with you DooMAD, your shill account is burned

Vires in numeris
1513514127
Hero Member
*
Offline Offline

Posts: 1513514127

View Profile Personal Message (Offline)

Ignore
1513514127
Reply with quote  #2

1513514127
Report to moderator
1513514127
Hero Member
*
Offline Offline

Posts: 1513514127

View Profile Personal Message (Offline)

Ignore
1513514127
Reply with quote  #2

1513514127
Report to moderator
1513514127
Hero Member
*
Offline Offline

Posts: 1513514127

View Profile Personal Message (Offline)

Ignore
1513514127
Reply with quote  #2

1513514127
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1513514127
Hero Member
*
Offline Offline

Posts: 1513514127

View Profile Personal Message (Offline)

Ignore
1513514127
Reply with quote  #2

1513514127
Report to moderator
DooMAD
Legendary
*
Online Online

Activity: 1456



View Profile WWW
September 19, 2017, 08:13:12 PM
 #22

Yeah, because super-majority miner activated hard-forks away from the proven development team, with zero technical justification == letting people choose


Done with you DooMAD, your shill account is burned

Could have sworn the argument used to be that it wouldn't be a valid hardfork unless there was a super-majority.  Now that it might happen, a super-majority is a bad thing because it suits you?  How the goalposts suddenly move as people get more and more desperate.  And then when you run out of arguments, just call everyone a shill.

And you still can't square the circle that it's supposedly only freedom when you personally agree with it.  Real freedom involves accepting that sometimes people won't agree with you and want to do their own thing instead.  Another reason why I'm more than satisfied with a change of algo if the hashpower plummets.  The developers have the freedom to do that and I have no intention to deny them that freedom.  I'm not in the habit of telling other people what they can or can't code.  I'll leave that to you, Napoleon.

achow101
Moderator
Legendary
*
Offline Offline

Activity: 1246


17kKQppUsngUiByDsce4JXoZEjjpvX9bpR


View Profile WWW
September 19, 2017, 09:10:05 PM
 #23

Could have sworn the argument used to be that it wouldn't be a valid hardfork unless there was a super-majority.  Now that it might happen, a super-majority is a bad thing because it suits you? 
No, the argument is still that a valid hardfork requires a super majority of all users of Bitcoin (meaning miners, businesses, and users), in fact, it requires consensus. A super majority of miners is not enough; miners do not and cannot dictate what Bitcoin is. You need a super majority of everyone involved in Bitcoin.

QuintLeo
Hero Member
*****
Offline Offline

Activity: 910


View Profile
September 19, 2017, 09:12:10 PM
 #24

Any coin that "regularly changed it's algo" would be cutting it's OWN throat - users are NOT going to tolerate the disruption of having their coins become UNAVAILABLE on a regular basis due to algo-forced wallet changes and NO MINER HASHRATE TO CONFIRM TRANSACTIONS during these changeovers.

It could be implemented in way that wouldn't involve huge breaks in the hashrate, so I would forget about that angle

 I would NOT forget about it.

 As soon as you change the algorithm, you have most or ALL of your miners suddenly not mining 'till they change the mining software.

 (edit) and if it's an ASIC-based algorithm like the SHA256 used on Bitcoin you have lost your hashrate and ability to process a significant number of transactions for a VERY LONG TIME since the miners CAN'T just "change their software".



Carlton Banks
Legendary
*
Offline Offline

Activity: 1848



View Profile
September 19, 2017, 09:36:14 PM
 #25

Any coin that "regularly changed it's algo" would be cutting it's OWN throat - users are NOT going to tolerate the disruption of having their coins become UNAVAILABLE on a regular basis due to algo-forced wallet changes and NO MINER HASHRATE TO CONFIRM TRANSACTIONS during these changeovers.

It could be implemented in way that wouldn't involve huge breaks in the hashrate, so I would forget about that angle

 I would NOT forget about it.

 As soon as you change the algorithm, you have most or ALL of your miners suddenly not mining 'till they change the mining software.




Yes, that could be avoided, depending on how it's done. Don't worry, it's fairly simple to achieve.

Vires in numeris
Wind_FURY
Hero Member
*****
Offline Offline

Activity: 574


★Jetwin.com★


View Profile
September 20, 2017, 04:33:27 AM
 #26

I believe the thread was hijacked by the Segwit2x vs. Core topic. They are still good to read though, but please answer my original question first.

IF Core decided that a Proof of Work change is needed, what kind of mining algorithm will they use in your opinion? Will they develop something new, or will they use something that is used now like Scrypt?

There is also a new white paper for another mining algorithm out there called Proof of Space. https://www.coindesk.com/proof-of-space-bittorrent-creator-publishes-eco-friendly-mining-paper/


▄▄▄████████▄▄▄
▄▄███▀▀▀ ▄  ▄ ▀▀▀███▄▄
▄██▀▀ ▄▄████  ████▄▄ ▀▀██▄
▄██▀ ▄███████    ███████▄ ▀██▄
██▀ ▄████████▀    ▀████████▄ ▀██
██▀ ██████████      ██████████ ▀██
██▀ ██████████        ██████████ ▀██
▄██                                ██▄
██ ▄                              ▄ ██
██ ███▄                        ▄███ ██
██ ██████▄                  ▄██████ ██
██ ▀████████              ████████▀ ██
▀██ ███████                ███████ ██▀
██▄ █████▀                ▀█████ ▄██
██▄ ████        ▄▄        ████ ▄██
██▄ ▀█      ▄▄████▄▄      █▀ ▄██
██▄    ▄▄██████████▄▄    ▄██▀
▀██▄▄ ▀▀██████████▀▀ ▄▄██▀
▀▀███▄▄▄ ▀▀▀▀ ▄▄▄███▀▀
▀▀▀████████▀▀▀
 

    [    ]
buwaytress
Sr. Member
****
Offline Offline

Activity: 462


I bit, therefore I am


View Profile
September 20, 2017, 05:08:48 AM
 #27

I believe the thread was hijacked by the Segwit2x vs. Core topic. They are still good to read though, but please answer my original question first.

IF Core decided that a Proof of Work change is needed, what kind of mining algorithm will they use in your opinion? Will they develop something new, or will they use something that is used now like Scrypt?

There is also a new white paper for another mining algorithm out there called Proof of Space. https://www.coindesk.com/proof-of-space-bittorrent-creator-publishes-eco-friendly-mining-paper/

Would still be good if someone could also trace the quote (if there's one), since I couldn't find it either. CB speculated above that if this algorithm change(s) were to happen, then it would usher in a real a gold rush (home users, typical Bitcoin users finally able to contribute to mining with a decent computer). That's definitely a situation I'd look forward to. Miners might feel this reduces profitability and hence the playing field, but the super miners can always move on to alts, right?

And certainly a lot of new mining algos I've now seen - eco-friendlier, asic-resistant, etc. While I doubt the snazzier ones like Proof-of-Time are technically solid, I do like some aspects of Proof-of-Importance... amount of work important, but so is quality. Amount of coins important, but less than usage, for example.

Carlton Banks
Legendary
*
Offline Offline

Activity: 1848



View Profile
September 20, 2017, 12:36:16 PM
 #28

I believe the thread was hijacked by the Segwit2x vs. Core topic. They are still good to read though, but please answer my original question first.

IF Core decided that a Proof of Work change is needed, what kind of mining algorithm will they use in your opinion? Will they develop something new, or will they use something that is used now like Scrypt?

There is also a new white paper for another mining algorithm out there called Proof of Space. https://www.coindesk.com/proof-of-space-bittorrent-creator-publishes-eco-friendly-mining-paper/

And certainly a lot of new mining algos I've now seen - eco-friendlier, asic-resistant, etc. While I doubt the snazzier ones like Proof-of-Time are technically solid, I do like some aspects of Proof-of-Importance... amount of work important, but so is quality. Amount of coins important, but less than usage, for example.


A vaunted algo was something called EquiHash (I think), as it's design is amongst the more difficult to produce an ASIC circuit for, and presumably it isn't too new or based on exotic math. But it will always be possible to produce an ASIC eventually, Scrypt was always touted as "ASIC resistant", until someone produced an ASIC unit with the requisite RAM modules to vastly outcompete PC/GPU Scrypt miners.



The key will clearly be to alternate hashing algorithms for PoW, either individually, or as a series of hash functions with the order of the series alternated randomly from a large set of hashing algos (Meni Rosenfeld's idea). And while that would decentralise mining a great deal, it would still mean those with access to the most resources would dominate mining (i.e. money, access to large premises with suitable MW rated electricity supply, cheap electricity, suitable environment for cooling, political favour etc).

But small miners would suddenly get back into the game even still; right now, mining ASIC manufacturers deliberately set the price of complete miner units at a level that guarantees their continued dominance in the actual block reward market (something like 5-8 times the market price of what electronics of a similar level of sophistication and material constituency typically would be). With that huge economic disadvantage removed, home miners can of course purchase regular CPU or GPU computing devices at the same prices as large organisations can. It would be good for Bitcoin, despite being somewhat disruptive at first.

Vires in numeris
thejaytiesto
Legendary
*
Offline Offline

Activity: 1134



View Profile
September 20, 2017, 03:31:07 PM
 #29

Could have sworn the argument used to be that it wouldn't be a valid hardfork unless there was a super-majority.  Now that it might happen, a super-majority is a bad thing because it suits you?
No, the argument is still that a valid hardfork requires a super majority of all users of Bitcoin (meaning miners, businesses, and users), in fact, it requires consensus. A super majority of miners is not enough; miners do not and cannot dictate what Bitcoin is. You need a super majority of everyone involved in Bitcoin.

Full nodes dictate what bitcoin is. At the end of the day the rules are imposed by the software you are running, which is why government agents are trying to push hardfork-attacks that involve big block sizes, so only their corporations can impose said rules and not random people on their basements.

A miner could have massive amounts of hashrate and "prove a lot of work" which is useless if the people isn't accepting that hashrate as valid because they are using full nodes that ignore their blocks.

Similarly, just because you have a ton of machines that can make a huge hole in the middle of a desert and prove you did all of that work, nobody is going to value a stupid hole in the middle of a desert. So don't fall for this "we have all the hashrate so we rule" bullshit agenda that some are trying to push, including McAfee which made new friends Roger and Jihan Wu recently.



Can you fucking believe it?

DooMAD
Legendary
*
Online Online

Activity: 1456



View Profile WWW
September 20, 2017, 05:57:13 PM
 #30

Could have sworn the argument used to be that it wouldn't be a valid hardfork unless there was a super-majority.  Now that it might happen, a super-majority is a bad thing because it suits you?
No, the argument is still that a valid hardfork requires a super majority of all users of Bitcoin (meaning miners, businesses, and users), in fact, it requires consensus. A super majority of miners is not enough; miners do not and cannot dictate what Bitcoin is. You need a super majority of everyone involved in Bitcoin.

I'm aware that it's not just miners who make the decision and there has to be both support from non-mining nodes and economic support from users and third party services.  The real point is, if a fork happens, we'll get to find out for certain where that super-majority lies, rather than than putting the cart before the horse and just assuming based on your own personal preference about which chain each of you happen to like best.  It's largely speculation until we see the actual usage in real time.  You can't prevent the open market from doing its thing here in Cryptoland, and there's undeniably a gap in the market attempting to fill itself if these forks are occurring.  Assign the blame for that where you will, whether it's crying foul of "miner cartels", or "benevolent dictators" developing alternative clients, but there's no stopping it.  Plus, everyone can finally stop bitching about the miners having too much influence if they're going to head off and do their own thing.  People should be happy about all this, but I suspect no one actually will be because it's all about teh dramas(sic)   Roll Eyes



Full nodes dictate what bitcoin is.

LOL NOPE.  No one group does.  That's the whole point.  No one dictates anything.  Everyone does what's best for their own selfish interests and consensus, together with the free market, sorts it all out.  Create the right alignment of incentives and everyone works together voluntarily, don't create the right alignment of incentives and people fork off.  



I believe the thread was hijacked by the Segwit2x vs. Core topic. They are still good to read though, but please answer my original question first.

IF Core decided that a Proof of Work change is needed, what kind of mining algorithm will they use in your opinion? Will they develop something new, or will they use something that is used now like Scrypt?

Technically, the developers don't even have to settle on a single algorithm.  It's possible to use more than one.  I'm pretty sure there's an altcoin out there utilising 6 at once.  This also apparently helps with ASIC resistance due to the extra math involved.  Certainly something to consider.  If the goal is ASIC resistance, maybe leave Scrypt well alone.

d5000
Legendary
*
Offline Offline

Activity: 1582



View Profile
September 21, 2017, 12:34:27 AM
 #31

Caution, another Segwit2x shill coming in Grin

But I am really interested in the original question. For me, the November fork will be far from a "triviality" if the current support for Segwit2x continues the way the block header messages indicate. I don't consider BitPay, Coinbase and Xapo at all "overrated" or so, it is even possible that they are the economic majority, on their own. So definitively it will be hard for "Original Bitcoin" to survive if 90% of the miners switch and the big companies switch, too.

I would like a long-term switch to something like Proof-of-Space, but I think there must be much more research about that topic to only be able to think about it (only 2 altcoins I know about - Spacecoin and Burst - are using it now, but I am not very convinced about their implementations). Proof-of-Importance mentioned by @buwaytress is basically Proof of Stake and thus I would only support it if the nothing-at-stake-problem is definitively solved or it is proven that the derived attacks are so impractical and expensive that they are more difficult to execute than 51%ing Bitcoin.

Anyway - I think the algorithm must be chosen so that Bitcoin cannot be easily attacked by the big mining cartels. Maybe Scrypt isn't the worst option for a "nuclear option" algorithm, because there is already hardware available and there is lots of experience in the altcoin world.

       ▀
   ▄▄▄   ▄▀
   ███ ▄▄▄▄  ██
       ████
    ▄  ▀▀▀▀
▄▄
      ██    ▀▀
██▄█▄▄▄████████
▄▄▄▄▄▄▄▄▀▀███▀▀▀
██████████████████
████▄▀▄▀▄▀███▀▀▀▀▀
████▄▀▄▀▄▀███ ▀
████▄▀▄▀▄▀████████
▀█████████████████
]
,CoinPayments,
█████
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████
█████
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████
█████
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████
Carlton Banks
Legendary
*
Offline Offline

Activity: 1848



View Profile
September 21, 2017, 11:25:08 AM
 #32

Anyway - I think the algorithm must be chosen so that Bitcoin cannot be easily attacked by the big mining cartels. Maybe Scrypt isn't the worst option for a "nuclear option" algorithm, because there is already hardware available

Roll Eyes

Uh, that would be a reason against using Scrypt. Mining cartels would still be pushing their agenda if there is ASIC hardware in existence, which for Scrypt, there is.

Vires in numeris
thejaytiesto
Legendary
*
Offline Offline

Activity: 1134



View Profile
September 21, 2017, 11:40:54 AM
 #33

Could have sworn the argument used to be that it wouldn't be a valid hardfork unless there was a super-majority.  Now that it might happen, a super-majority is a bad thing because it suits you?
No, the argument is still that a valid hardfork requires a super majority of all users of Bitcoin (meaning miners, businesses, and users), in fact, it requires consensus. A super majority of miners is not enough; miners do not and cannot dictate what Bitcoin is. You need a super majority of everyone involved in Bitcoin.

I'm aware that it's not just miners who make the decision and there has to be both support from non-mining nodes and economic support from users and third party services.  The real point is, if a fork happens, we'll get to find out for certain where that super-majority lies, rather than than putting the cart before the horse and just assuming based on your own personal preference about which chain each of you happen to like best.  It's largely speculation until we see the actual usage in real time.  You can't prevent the open market from doing its thing here in Cryptoland, and there's undeniably a gap in the market attempting to fill itself if these forks are occurring.  Assign the blame for that where you will, whether it's crying foul of "miner cartels", or "benevolent dictators" developing alternative clients, but there's no stopping it.  Plus, everyone can finally stop bitching about the miners having too much influence if they're going to head off and do their own thing.  People should be happy about all this, but I suspect no one actually will be because it's all about teh dramas(sic)   Roll Eyes



Full nodes dictate what bitcoin is.

LOL NOPE.  No one group does.  That's the whole point.  No one dictates anything.  Everyone does what's best for their own selfish interests and consensus, together with the free market, sorts it all out.  Create the right alignment of incentives and everyone works together voluntarily, don't create the right alignment of incentives and people fork off.  



I believe the thread was hijacked by the Segwit2x vs. Core topic. They are still good to read though, but please answer my original question first.

IF Core decided that a Proof of Work change is needed, what kind of mining algorithm will they use in your opinion? Will they develop something new, or will they use something that is used now like Scrypt?

Technically, the developers don't even have to settle on a single algorithm.  It's possible to use more than one.  I'm pretty sure there's an altcoin out there utilising 6 at once.  This also apparently helps with ASIC resistance due to the extra math involved.  Certainly something to consider.  If the goal is ASIC resistance, maybe leave Scrypt well alone.

Cool, let them fork in november, we'll see what happens. At the end of the day, BCash forked, and it has been proven that no one wants bigger blocks. They can't fill their own blocks with more than 1MB, not anywhere close to 1MB, then why the fuck the rush to raise the blocksize? Turns out nobody is transacting there.

The scam in segwit2x is that these corporations claim to represent their users. So Coinbase and Xapo and all these corporations claim that their millions of users want segwit2x because they are using it. Motherfucker nobody knows shit about any of this, only us 1% of guys that look into the details, the rest of the people are using Coinbase or whatever because that's what they've heard, not because they prefer segwit2x over the original Bitcoin. So they are taking hostage all of these clueless people and claiming they got their support.

Not to mention the agreement itself has been broken already several times: https://medium.com/@WhalePanda/segwit2x-the-broken-agreement-e9035a453c05

But it doesn't matter. Let them fork again and we'll see what's up.

DooMAD
Legendary
*
Online Online

Activity: 1456



View Profile WWW
September 21, 2017, 08:09:54 PM
 #34

The scam in segwit2x is that these corporations claim to represent their users. So Coinbase and Xapo and all these corporations claim that their millions of users want segwit2x because they are using it. Motherfucker nobody knows shit about any of this, only us 1% of guys that look into the details, the rest of the people are using Coinbase or whatever because that's what they've heard, not because they prefer segwit2x over the original Bitcoin. So they are taking hostage all of these clueless people and claiming they got their support.

Not to mention the agreement itself has been broken already several times: https://medium.com/@WhalePanda/segwit2x-the-broken-agreement-e9035a453c05

But it doesn't matter. Let them fork again and we'll see what's up.

"Let them fork"?  Oh, so they have your permission now, do they?    Roll Eyes

I suppose that's about the best I can hope for, so I won't waste too much time pointing out that they were free to do that all along, with or without your blessing.  Also, I like the latest shift in emphasis that it's not about the economic majority anymore and it's only the informed 1% who matter.  Keep moving those goalposts, you insipid fascist.

But yes, they may well be making a mistake.  There may be no real demand for 2x.  The only thing that matters is you can't stop them from trying because that isn't your decision to make.

Carlton Banks
Legendary
*
Offline Offline

Activity: 1848



View Profile
September 21, 2017, 08:32:47 PM
 #35

51% attacks against the Bitcoin network are:

  • Permissionless
  • Also possible with 92% of the hashrate

And that's what this "fork" really is, using a cartelised hashrate to perform a 51% attack to steer the Bitcoin network away from the interests of the users. It's pretty obvious, as the fork provides no discernible upgrade to the network, includes uncompromising support from every questionable corporation in Bitcoin commerce, and features a political authoritarian as the chief developer.

If there was some kind of technical merit to Segwit2x, sure, we call it a fork. But seeing as it's a hashrate-led political move, it's a 51% attack.

Vires in numeris
d5000
Legendary
*
Offline Offline

Activity: 1582



View Profile
September 21, 2017, 11:08:04 PM
 #36

Uh, that would be a reason against using Scrypt. Mining cartels would still be pushing their agenda if there is ASIC hardware in existence, which for Scrypt, there is.

In this case, I think it's the other way around ... if there is no algorithm-specific ASIC hardware available, then someone with big pockets and experience in the field (and who would be better prepared for that than Bitmain?) will develop it, as it's probably impossible to find a totally "ASIC-resistant" algorithm (like you said in a previous post).

They could use their first mover advantage to get a large portion of the network hashrate holding back a portion of the equipment for themselves, so they could dominate easily and even attack the chain.

That's why I think with an algorithm like Scrypt, where already ASICs exist, could be better suited, because there is already a market for specialized hardware, so it would become more difficult to get that first mover advantage.

       ▀
   ▄▄▄   ▄▀
   ███ ▄▄▄▄  ██
       ████
    ▄  ▀▀▀▀
▄▄
      ██    ▀▀
██▄█▄▄▄████████
▄▄▄▄▄▄▄▄▀▀███▀▀▀
██████████████████
████▄▀▄▀▄▀███▀▀▀▀▀
████▄▀▄▀▄▀███ ▀
████▄▀▄▀▄▀████████
▀█████████████████
]
,CoinPayments,
█████
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████
█████
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████
█████
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████ ██
█████
Carlton Banks
Legendary
*
Offline Offline

Activity: 1848



View Profile
September 22, 2017, 10:15:33 AM
 #37

That's why I think with an algorithm like Scrypt, where already ASICs exist, could be better suited, because there is already a market for specialized hardware, so it would become more difficult to get that first mover advantage.

You don't get it, you're thinking in snapshots. The moment when some group of upstart ASIC producers make a Scrypt miner is not somehow frozen in time forever.

ASIC makers, irrelevant of their size when they begin their business, have an incentive to gouge their prices to bestow their own personal mining operations with an unassailable competitive advantage.

The actual solution is to choose a PoW hashing algorithm for which no-one has an ASIC design, thereby allowing regular users to use standard equipment to compete more effectively against warehouse scale mining operations. Algorithms that allow regular people to use standard, non-specialised equipment are the key to decentralising the hashrate.


(you should get a job doing propaganda for established Bitcoin mining ASIC producers, you accidentally say all the things they really want the public to believe)

Vires in numeris
BitUsher
Hero Member
*****
Offline Offline

Activity: 840


View Profile
September 22, 2017, 09:59:44 PM
 #38

I read a comment on Reddit that if 90% of the miners will still follow Segwit2x on November

Signalling is akin to people wearing hats on twitter and doesn't hold much weight. Miners ultimately follow profits and their signalling is often contradictory but this is the most recent tally -

Bitcoin Mining Pool 6 month hashrate share % for miners who agreed to NYA

34 1 Hash (China) 2.8%
35 BitClub Network (Hong Kong) 3.5%
36 Bitcoin.com (St. Kitts & Nevis) 2.3%
37 Bitfury (United States) 7.1%
38 Bitmain (China) (Antpool) 17.5%
39 btc.com (China) 10.7%
10 BTCC (Also a DGC Company) 7.8%
40 BTC.TOP (China) 10.7%
41 F2Pool (China) 9.5% Abandoned NYA

42 ViaBTC (China) 5.3%

Share of hashrate 67.7%


then Core will have no choice but to fork off and upgrade to a new mining algorithm.

No. A PoW change is generally regarded as a worst case scenario and unlikely to happen . If segwit2x forks people like me and many others will simply split our coins and begin to dump our barrycoin alt on exchanges to rebuy back bitcoin. This should drive back most the hashrate . If ~15% of the hashrate doesn't return in a few weeks than users like myself might consider a PoW change HF.


but what are the proposals made for the new kind of mining algorithm?


Various plans have already been made , but it isn't wise to announce before necessary. The algo will likely memory hard , thus mined with GPUs. Candidates like Keccak SHA512 or cuckoo are example candidates
ChromaticStar
Full Member
***
Offline Offline

Activity: 196


View Profile
September 24, 2017, 03:09:06 PM
 #39

I don't have much technical experience and I've never mined before. From everything I've been reading on the technical forums, the 2x chain is not in any way "Bitcoin." At this point, with a fork almost inevitable, would it be worth it to buy a GPU miner, learn to use it and support the Core backed chain when the network splits in November?
DooMAD
Legendary
*
Online Online

Activity: 1456



View Profile WWW
September 24, 2017, 04:17:40 PM
 #40

I don't have much technical experience and I've never mined before. From everything I've been reading on the technical forums, the 2x chain is not in any way "Bitcoin." At this point, with a fork almost inevitable, would it be worth it to buy a GPU miner, learn to use it and support the Core backed chain when the network splits in November?

It's a bit of a financial gamble at this stage, as the change of algorithm isn't set in stone.  If there are any GPU-mineable altcoins you wouldn't mind accumulating as a fallback, then go for it.  But if you have your mind set on Bitcoin, wait until the change of algo is confirmed.  GPU miners don't really touch the sides as it were while the ASICs are still in play.

Pages: « 1 [2] 3 4 »  All
  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!