brg444 (OP)
|
|
December 31, 2015, 07:48:17 PM |
|
As I said earlier when looking at block size you have to look at transaction fees.Do we force larger transaction fees on people or do we just let the chinese tell us what they want.I was told this was in the future but sorry the future is now
Yes! Let's socialize costs on everyone like all good collectivists shills would. TRANSACTION SUBSIDY FOR EVERYONE!!! BITCOIN FOR THE PEOPLE!! You statists make me sick
|
"I believe this will be the ultimate fate of Bitcoin, to be the "high-powered money" that serves as a reserve currency for banks that issue their own digital cash." Hal Finney, Dec. 2010
|
|
|
tl121
|
|
December 31, 2015, 07:50:32 PM |
|
Bitfury's paper here: http://bitfury.com/content/4-white-papers-research/block-size-1.1.1.pdf"The table contains an estimate of how many full nodes would no longer function without hard-ware upgrades as average block size is increased. These estimates are based on the assumption that many users run full nodes on consumer-grade hardware, whether on personal computers or in the cloud. Characteristics of node hardware are based on a survey performed by Steam [19]; we assume PC gamers and Bitcoin enthusiasts have a similar amount of resources dedicated to their hardware. The exception is RAM: we assume that a typical computer supporting a node has no less than 3 GB RAM as a node requires at least 2GB RAM to run with margin[15]. For example,if block size increases to 2 MB, a node would need to dedicate 8 GB RAM to the Bitcoin client, while more than a half of PCs in the survey have less RAM." Based on his estimation, raise the block size to 4MB will drop 75% of nodes from the network The 8 GB RAM module for the computer that runs my Bitcoin Unlimited node has 8 GB of RAM. I paid $67 for this memory module 13 months ago. I note that Amazon is selling the same module today for $35 USD. It is unreasonable to cripple bitcoin to support users running obsolete hardware.
|
|
|
|
jbreher
Legendary
Offline
Activity: 3080
Merit: 1688
lose: unfind ... loose: untight
|
|
December 31, 2015, 08:14:19 PM |
|
BU ... doesn't even have the proper code that it needs (Source: Gavin).
I see that you are trying to build in the mind the reader the impression that Gavin has declared that the BU codebase has resulted in a faulty executable program. Unless Gavin has issued further input to the BU review, you are mischaracterizing his statement. Nothing i have read in his comments would support that position. Of course, if he has issued a further statement, I'd be happy to read it. Incidentally: appeal to authority?
|
Anyone with a campaign ad in their signature -- for an organization with which they are not otherwise affiliated -- is automatically deducted credibility points.
I've been convicted of heresy. Convicted by a mere known extortionist. Read my Trust for details.
|
|
|
jbreher
Legendary
Offline
Activity: 3080
Merit: 1688
lose: unfind ... loose: untight
|
|
December 31, 2015, 08:26:20 PM |
|
Thanks for the link. I'll read it in its entirety. Yeah. Back from reading it. The conclusion you quote is completely based upon RAM demands. The RAM per node assumptions hinge completely upon a survey of machines running Steam. Other than to reiterate how absurd this appears to me, I shall refrain from further comment. Well, after adding this... I can see the motivation for using Steam figures. They are publicly available, and form a fairly broad sampling. As such, they are an easy thing to cling to. However, they are a sampling only of a completely unrelated user base. Pity I've not been able to locate absolute numbers in Steam's data set. Any guesses as to whether or not the raw number behind the 15.8% of their Win user base that has 12 GiB or more completely dwarfs the 5555 or so full Bitcoin nodes?
|
Anyone with a campaign ad in their signature -- for an organization with which they are not otherwise affiliated -- is automatically deducted credibility points.
I've been convicted of heresy. Convicted by a mere known extortionist. Read my Trust for details.
|
|
|
Soros Shorts
Donator
Legendary
Offline
Activity: 1617
Merit: 1012
|
|
December 31, 2015, 10:15:24 PM |
|
Any guesses as to whether or not the raw number behind the 15.8% of their Win user base that has 12 GiB or more completely dwarfs the 5555 or so full Bitcoin nodes?
I suspect there are many nodes running on hosted VMs in the sub-12GB RAM range. This category of hosted servers is not even represented in the Steam sample. I actually run a few 3.5GB nodes on Azure and they run real well because of the good connectivity. If you price 12GB VM images on Amazon or Azure you'll find that they run for a couple of hundred dollars a month or more, which is kind of expensive. So I would imagine that hobbyist-run hosted VMs would be the first drop off as RAM requirements went up.
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
|
January 01, 2016, 02:44:43 AM Last edit: January 01, 2016, 03:27:52 AM by johnyj |
|
Bitfury's paper here: http://bitfury.com/content/4-white-papers-research/block-size-1.1.1.pdf"The table contains an estimate of how many full nodes would no longer function without hard-ware upgrades as average block size is increased. These estimates are based on the assumption that many users run full nodes on consumer-grade hardware, whether on personal computers or in the cloud. Characteristics of node hardware are based on a survey performed by Steam [19]; we assume PC gamers and Bitcoin enthusiasts have a similar amount of resources dedicated to their hardware. The exception is RAM: we assume that a typical computer supporting a node has no less than 3 GB RAM as a node requires at least 2GB RAM to run with margin[15]. For example,if block size increases to 2 MB, a node would need to dedicate 8 GB RAM to the Bitcoin client, while more than a half of PCs in the survey have less RAM." Based on his estimation, raise the block size to 4MB will drop 75% of nodes from the network The 8 GB RAM module for the computer that runs my Bitcoin Unlimited node has 8 GB of RAM. I paid $67 for this memory module 13 months ago. I note that Amazon is selling the same module today for $35 USD. It is unreasonable to cripple bitcoin to support users running obsolete hardware. Let's compare the cost for 4MB blocks: 1. You spend several hundred dollars on hardware and make a new node dedicated to bitcoin (many gaming machines do not support 16GB memory, thus you need to upgrade pretty much the whole machine), in hope of maintaining the $0.05 fee for bitcoin transactions (and it requires thousands of other nodes also do the same as you) 2. You use those several hundred dollars to pay the transaction fee (should be enough for at least one hundred transactions even the fee rose 100x to $5 per transaction) Notice that setting up a full node does not benefit the node operator in anyway, and raise the block size will require thousands of such voluntarily setup nodes to upgrade. So I guess any rational human would refuse the node upgrade and pay the fee instead. I guess even the fee has risen to a prohibitive level, average user would still pay fee instead of setting up full nodes using dedicated hardware
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
|
January 01, 2016, 03:20:12 AM Last edit: January 01, 2016, 03:36:22 AM by johnyj |
|
Miners do not even run full nodes themselves for the purpose of mining, which is why the increased difficulty of running a full node does not increase the difficulty or barrier to entry of mining.
I am a miner myself, solo mining is only feasible if you are a huge industrial operation, not exactly contributing to decentralization. P2P mining unfortunately is not good enough both due to the increased complexity and incompatibility with certain ASIC miners, this is reflected in the hashrate. More then seventy percent of the mining power is presently inside of public pools, this is a good thing. Pools promote decentralization compared to the alternatives, pools are like a form of representative democracy for the miners.
So you are mining in a BU mining pool? Which one is that? Today's way of mining, as a core dev pointed out, is a historical mistake that became popular. If they had advertised p2pool from the beginning, then pooled mining might not look like it is today In earlier version of bitcoin, every miner is running a full node and use CPU to mine, that is the ideal configuration for a decentralized network, because there is no concentration of hash power However with the rise of dedicated mining software and hardware, the mining start to concentrate at a handful of large pools, this dramatically increased the degree of centralization, because a few large pool can use miner's hash power to change the network without their awareness (2013 hard fork demonstrated god-like power from pools: BTCGUILD and Slush pool rolled back the fork with their hash power, most of their miners did not even know what happened) P2Pool is painful to setup and run, but if you think long term wise, the future bitcoin mining should ideally all be done through P2POOL or similar way, thus totally eliminate the risk of large mining pools colluding to threaten the network. More importantly, if every miner need to run a full node, then the miner will be very aware of the impact of all the protocol change, thus become more involved in the protocol related discussion
|
|
|
|
VeritasSapere
|
|
January 01, 2016, 07:33:47 AM Last edit: January 01, 2016, 08:13:32 AM by VeritasSapere |
|
Bitfury's paper here: http://bitfury.com/content/4-white-papers-research/block-size-1.1.1.pdf"The table contains an estimate of how many full nodes would no longer function without hard-ware upgrades as average block size is increased. These estimates are based on the assumption that many users run full nodes on consumer-grade hardware, whether on personal computers or in the cloud. Characteristics of node hardware are based on a survey performed by Steam [19]; we assume PC gamers and Bitcoin enthusiasts have a similar amount of resources dedicated to their hardware. The exception is RAM: we assume that a typical computer supporting a node has no less than 3 GB RAM as a node requires at least 2GB RAM to run with margin[15]. For example,if block size increases to 2 MB, a node would need to dedicate 8 GB RAM to the Bitcoin client, while more than a half of PCs in the survey have less RAM." Based on his estimation, raise the block size to 4MB will drop 75% of nodes from the network The 8 GB RAM module for the computer that runs my Bitcoin Unlimited node has 8 GB of RAM. I paid $67 for this memory module 13 months ago. I note that Amazon is selling the same module today for $35 USD. It is unreasonable to cripple bitcoin to support users running obsolete hardware. Notice that setting up a full node does not benefit the node operator in anyway, and raise the block size will require thousands of such voluntarily setup nodes to upgrade. So I guess any rational human would refuse the node upgrade and pay the fee instead. I guess even the fee has risen to a prohibitive level, average user would still pay fee instead of setting up full nodes using dedicated hardware It is not just that the fees will go up, there will not be enough space for everyone to transact regardless of the fee that they pay. It would also render transactions on the Bitcoin network much more unreliable. It is not a rational decision for the network to prioritize full nodes over keeping the fees low. After all, the majority of users do not even run full nodes at all, why should they even care so much about servers in a datacentre. The benefits of low fees far outweigh the increased cost of running full nodes. The current system where every user is a network node is not the intended configuration for large scale. That would be like every Usenet user runs their own NNTP server. The design supports letting users just be users.
|
|
|
|
VeritasSapere
|
|
January 01, 2016, 07:51:29 AM Last edit: January 01, 2016, 08:10:12 AM by VeritasSapere |
|
Miners do not even run full nodes themselves for the purpose of mining, which is why the increased difficulty of running a full node does not increase the difficulty or barrier to entry of mining.
I am a miner myself, solo mining is only feasible if you are a huge industrial operation, not exactly contributing to decentralization. P2P mining unfortunately is not good enough both due to the increased complexity and incompatibility with certain ASIC miners, this is reflected in the hashrate. More then seventy percent of the mining power is presently inside of public pools, this is a good thing. Pools promote decentralization compared to the alternatives, pools are like a form of representative democracy for the miners.
So you are mining in a BU mining pool? Which one is that? I am mining with slush at the moment. Today's way of mining, as a core dev pointed out, is a historical mistake that became popular. I find this a ridiculous thing to say, pooled mining is how mining works today and it is a good thing. If they had advertised p2pool from the beginning, then pooled mining might not look like it is today. We should have done what Core told us to do instead of the market deciding for itself? In earlier version of bitcoin, every miner is running a full node and use CPU to mine, that is the ideal configuration for a decentralized network, because there is no concentration of hash power. That is not how it works today, furthermore it can not work that way today because of the size of the network, pools are essential in countering variance so that small miners can still operate profitably, without the pools we would have much more centralization in mining then we do today. However with the rise of dedicated mining software and hardware, the mining start to concentrate at a handful of large pools, this dramatically increased the degree of centralization, because a few large pool can use miner's hash power to change the network without their awareness (2013 hard fork demonstrated god-like power from pools: BTCGUILD and Slush pool rolled back the fork with their hash power, most of their miners did not even know what happened) I consider this a positive development, I think we would have gravitated towards this position naturally. Pools are like a form of representative democracy for the miners. Personally I am fond of the idea of having these large organizations like pools watching the blockchain on our behalf night and day, since they are in that position, I can go out and party and have a good time like I did tonight, without having to worry about what my hashpower was doing in case of a fork or any other complications. Another advantage of pooled mining is that you do not require a good internet connection to participate, it in fact lowers the barrier to entry. I can also mine anonymously since latency is also not a significant issue when doing pooled mining. In regards to this blocksize debate it is also evident that pooled mining allows us to increase the blocksize without effecting mining centralization whatsoever, since pooled miners do not run full nodes for the purpose of mining, therefore they are not effected by the increased difficulty of running a full node. For the larger miners and pools, the cost of running a full node is relatively insignificant considering the size of their operations, even with much larger blocks. P2Pool is painful to setup and run, but if you think long term wise, the future bitcoin mining should ideally all be done through P2POOL or similar way, thus totally eliminate the risk of large mining pools colluding to threaten the network. More importantly, if every miner need to run a full node, then the miner will be very aware of the impact of all the protocol change, thus become more involved in the protocol related discussion. These are good points, and in principle I do like P2Pool. I even looked into doing P2P mining for my own operation, there where two problems however, one was that some of my equipment seemed to be incompatible or at the very least very complex to setup, secondly the hashrate for P2Pool is still to low, and the variance was to high. The reality therefore is that mining today is mostly done through public pools, I do not see this as a problem. I think that Bitcoin is working as it was intended and it should be allowed to continue to operate as intended by increasing the blocksize thereby not blocking the stream of transactions and allowing Bitcoin to continue to grow.
|
|
|
|
sidhujag
Legendary
Offline
Activity: 2044
Merit: 1005
|
|
January 01, 2016, 08:53:07 AM |
|
Very hard to believe the guy with major in humanities is a miner of bitcoin and has a say for bigger blocks.. seems like there is a hidden agenda with this user.. whoever it is.. he or she is a small minory left to support this altcoin
|
|
|
|
Cconvert2G36
|
|
January 01, 2016, 09:07:07 AM |
|
Very hard to believe the guy with major in humanities is a miner of bitcoin and has a say for bigger blocks.. seems like there is a hidden agenda with this user.. whoever it is.. he or she is a small minory left to support this altcoin
Better start digging... I'm sure there's some nefarious motive here.
|
|
|
|
Lauda
Legendary
Offline
Activity: 2674
Merit: 3000
Terminated.
|
|
January 01, 2016, 10:06:52 AM |
|
Of course, if he has issued a further statement, I'd be happy to read it.
Incidentally: appeal to authority?
Your post is wrong. People in this discussion have become very deluded throwing fallacies around like they're candy. Naming the source of something means that I'm appealing to authority? I see that you are trying to build in the mind the reader the impression that Gavin has declared that the BU codebase has resulted in a faulty executable program. Unless Gavin has issued further input to the BU review, you are mischaracterizing his statement. Nothing i have read in his comments would support that position.
You can easily see what Gavin wrote a few posts back in this thread.
|
"The Times 03/Jan/2009 Chancellor on brink of second bailout for banks" 😼 Bitcoin Core ( onion)
|
|
|
VeritasSapere
|
|
January 01, 2016, 03:02:13 PM |
|
Of course, if he has issued a further statement, I'd be happy to read it.
Incidentally: appeal to authority?
Your post is wrong. People in this discussion have become very deluded throwing fallacies around like they're candy. Naming the source of something means that I'm appealing to authority? I see that you are trying to build in the mind the reader the impression that Gavin has declared that the BU codebase has resulted in a faulty executable program. Unless Gavin has issued further input to the BU review, you are mischaracterizing his statement. Nothing i have read in his comments would support that position.
You can easily see what Gavin wrote a few posts back in this thread. You are falsely smearing Bitcoin Unlimited. The statement you where referring to was talking about the formatting and lazy blocking. Pretty standard for a code review. The lead developer of Bitcoin Unlimited even responded to Gavin in the same thread explaining the reasoning behind this formatting, this had nothing to with there being a faulty executable.
|
|
|
|
|
sAt0sHiFanClub
|
|
January 01, 2016, 04:09:29 PM |
|
It isn't a good idea to create a cult of personality [....] he also made multiple mistakes and from his programming history wasn't the most competent developer either
There's a difference between coding skillz and vision. Code can be reviewed and improved by committee at any stage. Vision is something you either have or you haven't. Take hashcash for example - A great coder could invent hashcash and turn it into a failed email anti spam tool. However, Satoshi could use his vision to use it as a bedrock in a project that goes on to have a $6.5 billion market cap that we are still talking about 6 years later. But I agree, cult of personality is stupid, But you still have to assign credit where credit is due.
|
We must make money worse as a commodity if we wish to make it better as a medium of exchange
|
|
|
Bitfirm
Newbie
Offline
Activity: 43
Merit: 0
|
|
January 01, 2016, 04:13:21 PM |
|
XT is a temporary solution, at best. Everyone understands that. Segregated witness is a better solution in the short run, and in the long run we'll have something else (lightning network?) So XT does not make much sense, don't think they'll be able to pull it through
|
|
|
|
coinmaster222
|
|
January 01, 2016, 04:27:20 PM |
|
Anybody point me to a link in helping me build a Bitcoin node,as this is off topic will remove it in a couple of hours
|
|
|
|
johnyj
Legendary
Offline
Activity: 1988
Merit: 1012
Beyond Imagination
|
|
January 01, 2016, 04:33:21 PM Last edit: January 01, 2016, 04:43:23 PM by johnyj |
|
We should have done what Core told us to do instead of the market deciding for itself?
So you are mining on a non-BU pool and promoting BU, and constantly quote Satoshi's words while at the same time say to let market decide for itself ... As explained many times, the decision making mechanism in bitcoin community is not democracy, it is consensus. Only the super majority consensus get passed, other attempts to fork will always fade away. When no consensus can be reached, the system keeps going on at its current status
|
|
|
|
coinmaster222
|
|
January 01, 2016, 04:37:01 PM |
|
Does anybody really not believe even if he denies it,that Gavin is still in discussion with Satoshi and this must be his idea as well
|
|
|
|
VeritasSapere
|
|
January 01, 2016, 04:50:13 PM |
|
We should have done what Core told us to do instead of the market deciding for itself?
So you are mining on a non-BU pool and promoting BU, and constantly quote Satoshi's words while at the same time say to let market decide for itself ... As explained many times, the decision making mechanism in bitcoin community is not democracy, it is consensus. Only the super majority consensus get passed, other attempts to fork will always fade away. When no consensus can be reached, the system keeps going on at its current status There is no BU pool yet, there is no contradiction in my position. I disagree with your theory on Bitcoin governance, consensus in the literal sense of the word is a terrible way to resolve conflicts, it does not work, especially among larger groups of people. The consensus mechanism of Bitcoin is what we call the governance mechanism of Bitcoin, it would be a mistake however to equate this with the literal meaning of the word. I would define the governance mechanism of Bitcoin in short form like this: Consensus is an emergent property which flows from the will of the economic majority. Proof of work is the best way to measure this consensus. The pools act as proxy for the miners, pools behave in a similar way to representatives within a representative democracy. Then in turn the miners act as a proxy for the economic majority. Since the miners are incentivized to follow the economic majority. In effect the economic majority rules Bitcoin, in other words the market rules Bitcoin. Bitcoin relies on the economic self-interest of the masses to govern consensus.
|
|
|
|
|