Bitcoin Forum
November 03, 2024, 08:02:04 PM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Poll
Question: Bitcoin fork proposal by respected Bitcoin lead dev Gavin Andresen, to increase the block size from 1MB to 20MB.
pro
anti
agnostic
DGAF

Pages: « 1 2 3 4 5 [6] 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 ... 125 »
  Print  
Author Topic: Bitcoin 20MB Fork  (Read 154781 times)
fat buddah
Full Member
***
Offline Offline

Activity: 336
Merit: 102


Get Ready to Make money.


View Profile
February 01, 2015, 02:28:34 AM
 #101

Has china been asked what they think about it?

Only banned it 17 times so far but I heard someone on twitter heard their friend's cousin said China's going to block github.

No, i mean, the users, traders and such. After all China holds a lot of coins and controls the marketplace. You're having this little neat discussion somewhere no chinese person ever looks. They are going to be surprised by your idiocy.

inBitweTrust
Hero Member
*****
Offline Offline

Activity: 658
Merit: 501



View Profile
February 01, 2015, 02:32:59 AM
 #102

Isnt the 1k GB per year blockchain size a HUGE issue here ?

The price of storage is exponentially getting cheaper and the 1K GB would be the theoretical maximum growth.... or not going to happen for at least a few more years.  Additionally, there are attempts to work on merkle tree pruning and other solutions to reduce blockchain bloat.

The reason this is important now is because their are some blocks that already are completly full at the 1MB limit. we don't want this to become the norm and have transaction fees skyrocket and confirmation times becoming 2-3 times longer.

cozk
Hero Member
*****
Offline Offline

Activity: 812
Merit: 500



View Profile
February 01, 2015, 02:41:36 AM
 #103

Isnt the 1k GB per year blockchain size a HUGE issue here ?

The price of storage is exponentially getting cheaper and the 1K GB would be the theoretical maximum growth.... or not going to happen for at least a few more years.  Additionally, there are attempts to work on merkle tree pruning and other solutions to reduce blockchain bloat.

The reason this is important now is because their are some blocks that already are completly full at the 1MB limit. we don't want this to become the norm and have transaction fees skyrocket and confirmation times becoming 2-3 times longer.

No way regular people will buy 1-5TB drives only to run a full node out of good will. This is dreaming.
Less full nodes means a less secure network.

I don't know much about this specific problematic but put this way i'd rather have higher transaction fees than a substantially weaker network.
BillyBobZorton
Legendary
*
Offline Offline

Activity: 1204
Merit: 1028


View Profile
February 01, 2015, 02:42:45 AM
 #104

Isnt the 1k GB per year blockchain size a HUGE issue here ?

The price of storage is exponentially getting cheaper and the 1K GB would be the theoretical maximum growth.... or not going to happen for at least a few more years.  Additionally, there are attempts to work on merkle tree pruning and other solutions to reduce blockchain bloat.

The reason this is important now is because their are some blocks that already are completly full at the 1MB limit. we don't want this to become the norm and have transaction fees skyrocket and confirmation times becoming 2-3 times longer.

Dude its a shitton. I have 4 TB of total size and i dont see any reason to upgrade, runing a bitcoin node would be the only reason basically.
FeedbackLoop
Hero Member
*****
Offline Offline

Activity: 742
Merit: 500



View Profile
February 01, 2015, 02:43:54 AM
 #105

The price of storage is exponentially getting cheaper

The rate of improvement in storage costs seems to have slowed down dramatically in the last decade or so:

http://www.mkomo.com/cost-per-gigabyte-update

inBitweTrust
Hero Member
*****
Offline Offline

Activity: 658
Merit: 501



View Profile
February 01, 2015, 02:52:44 AM
 #106

The price of storage is exponentially getting cheaper

The rate of improvement in storage costs seems to have slowed down dramatically in the last decade or so:

http://www.mkomo.com/cost-per-gigabyte-update



That graph shows a pretty healthy decrease in price over the last decade. Not exponential, but linear as Gavin's projections have charted.

a 4TB hard drive costs less than 135usd today guys.... have you checked prices? I currently have over 10TB in my desktop alone, space is so cheap.

Additionally, the average user with a 500GB hard-drive in a laptop isn't going to be running a node regardless as they don't even know how to call up their ISP and open port 8333 to begin with . There are only between 6-7k active nodes and most are miners at the moment.

I agree we should increase this amount for better decentralization , but we can do so by buying 5 dollar vps's which have much better bandwidth in the first place so, thus are preferred nodes to be distributing.

People already need a VPS to host their site or ownclowd, or TOR relay ... why not make it a bitcoin node at the same time?

cozk
Hero Member
*****
Offline Offline

Activity: 812
Merit: 500



View Profile
February 01, 2015, 02:55:06 AM
 #107

The price of storage is exponentially getting cheaper

The rate of improvement in storage costs seems to have slowed down dramatically in the last decade or so:

http://www.mkomo.com/cost-per-gigabyte-update



That graph shows a pretty healthy decrease in price over the last decade. Not exponential, but linear as Gavin's projections have charted.

a 4TB hard drive costs less than 135usd today guys.... have you checked prices? I currently have over 10TB in my desktop alone, space is so cheap.

Additionally, the average user with a 500GB hard-drive in a laptop isn't going to be running a node regardless as they don't even know how to call up their ISP and open port 8333 to begin with . There are only between 6-7k active nodes and most are miners at the moment.

I agree we should increase this amount for better decentralization , but we can do so by buying 5 dollar vps's which have much better bandwidth in the first place so, thus are preferred nodes to be distributing.

1k GB per year is a bigger problem than the 1MB vs 20MB thing imho or at the very least equivalent...
MakingMoneyHoney
Hero Member
*****
Offline Offline

Activity: 504
Merit: 500



View Profile
February 01, 2015, 02:58:45 AM
 #108

I've voted "anti" because it could increase the HDD usage. At this moment the Blockchain is very large, and this fork could made it even bigger. Maybe it could grow faster than HDD storage. Undecided

and bandwith, centralizing the whole thing.

for all i know, bitcoin is no cash, bitcoin is no gold, we dont know what it is yet (or i honestly dont), and mass adoption should not matter in the sense that they will get what usually get: what they are told to.


This reason, is the reason it isn't a clear cut "sure, why not" from me.....

And yes, altcoins can be used if you need larger transactions in the future, why not?
inBitweTrust
Hero Member
*****
Offline Offline

Activity: 658
Merit: 501



View Profile
February 01, 2015, 03:20:47 AM
Last edit: February 01, 2015, 03:48:46 AM by inBitweTrust
 #109

1k GB per year is a bigger problem than the 1MB vs 20MB thing imho or at the very least equivalent...

This is misinformation and you are confusing people.

That number isn't true as it assumes we will immediately be processing 84k transactions per block which isn't going to happen.
20GB is the future limit and not the size of most blocks in the future. We are only 600-700tpb at the moment
https://blockchain.info/charts/n-transactions-per-block which means we are typically 15-25% full blocks for 1MB blocks.... do you really expect this to shoot up to 100% full 20MB blocks overnight?.... it will take years to start getting close to those levels.

Additionally, you are completely ignoring the work that is being completed on merkle tree pruning.

P.S... You may ask yourself why bother if we aren't even fully utilizing 1MB blocks now ... the answer is that we are having more and more examples where the 1 MB block limit is preventing more transactions from confirming and this is only going to get worse with each passing day.

Analogy to think about --- Why do gamers look for gfx cards that can support incredibly high fps rendering their games avg's of 120fps when their monitors typically won't reflect more than 60 fps and they would be hard pressed to tell much of a difference between 30 to 60 fps with their naked eye? Answer- because avg fps doesn't matter but minimum fps matters and if you buy a gfx card that benchmarks 1080p @ 50fps , it may occasionally dip to 10-18fps during certain moments of the game and create major moments of lag.

cozk
Hero Member
*****
Offline Offline

Activity: 812
Merit: 500



View Profile
February 01, 2015, 03:41:49 AM
 #110

1k GB per year is a bigger problem than the 1MB vs 20MB thing imho or at the very least equivalent...

This is misinformation and you are confusing people.

That number isn't true as it assumes we will immediately be processing 84k transactions per block which isn't going to happen.
20GB is the future limit and not the size of most blocks in the future. We are only 600-700tpb at the moment
https://blockchain.info/charts/n-transactions-per-block which means we are typically 15-25% full blocks for 1MB blocks.... do you really expect this to shoot up to 100% full 20MB blocks overnight?.... it will take years to start getting close to those levels.

Additionally, you are completed ignoring the work that is being completed on merkle tree pruning.

P.S... You may ask yourself why bother if we aren't even fully utilizing 1MB blocks now ... the answer is that we are having more and more examples where the 1 MB block limit is preventing more transactions from confirming and this is only going to get worse with each passing day.

Analogy to think about --- Why do gamers look for gfx cards that can support incredibly high fps rendering their games avg's of 120fps when their monitors typically won't reflect more than 60 fps and they would be hard pressed to tell much of a difference between 30 to 60 fps with their naked eye? Answer- because avg fps doesn't matter but minimum fps matters and if you buy a gfx card that benchmarks 1080p @ 50fps , it may occasionally dip to 10-18fps during certain moments of the game and create major moments of lag.

Thanks for the infos.
inBitweTrust
Hero Member
*****
Offline Offline

Activity: 658
Merit: 501



View Profile
February 01, 2015, 03:44:40 AM
 #111

Thanks for the infos.

  Smiley No problem sir... we do need better communication between the devs and general bitcoin users. Bitcoin can be so complicated at times that even I have problems getting all the facts.

Perhaps, I should start creating more visual development roadmaps to distill the changelogs and future development roadmaps..

benjamindees
Legendary
*
Offline Offline

Activity: 1330
Merit: 1000


View Profile
February 01, 2015, 06:59:10 AM
 #112


IX. "Hard fork block size politics: do we want decentralized digital gold, or just another Visa? We want both. And there's no reason we can't have both."

Actually, there's a damned good reason "we" can't have both : one of them is nonsense. To grok this, tell me, whom do you know in Mali ?

No, not Bali, that island where preppy sluts go for spring break. Mali, the funny-shaped, landlocked West African nation.

Nobody ? Never even heard of anyone from there, never ate in a Mali restaurant, never bought a Mali anything over the mail, as far as you're concerned Mali could just as well not exist at all ? Well, bless you, the sentiment's exactly mutual! As far as they're concerned, the entire population of Mali couldn't give less of a fuck, and you might as well not exist.

Yet your proposal here is that we take everything of everyone : each pair of underwear, each box of notebooks left over from highschool, each car and set of car keys, each coffee mug, each old printer cartridge, each and every single item of everyone, worldwide, and dump them all into a huge warehouse in San Francisco. And then, you say, you'll give everyone receipts to keep track of their stuff. Obviously the receipts will be rather lengthy and complicated an affair, but all the inconvenience, confusion and sheer effort is worth it, you say. Because why ? "Because there's no reason we can't have both". Orly.

I get it, you're a centralist at heart

I am anything but.  On the contrary, I think it's great that Bitcoin is as decentralized as it is.  Believe me, I don't want to see Bitcoin centralized.  But to the extent that technical, or even political, reality necessitates some degree of voluntary centralization, people should have that option as well.  You shouldn't be allowed to hobble usage or force reliance on risky off-chain services in the name of decentralization any more than you should be forced to store transactions that don't interest you.  It works both ways.

Quote
, you want this globalisation thing where everyone's stuff is locked up in Fort Knox so the misfortunate indigents of Mali get to curse the day your pasty ass showed up forcing "democracy" on them.

I'm not a fan of democracy.  Perhaps Gavin is.  And perhaps you have a point there.  It's unfortunate that a valid criticism of Gavin's proposal has been drowned out by a lot of misdirected bloviating.

Quote
I don't particularly like this outlook, but who am I to tell you how to live your life.

Nevertheless, there's a damned good reason why you can't have both at my expense. You can't ask demand me and my friends and my business partners run complicated, expensive and ultimately pointless computer systems that are required to distinguish our transactions from Mali bound transactions, avoid double spends and all of that simply because you want the world to be a centrally-planned, single-core thing because you're too intellectually lazy and too mentally simple to accomodate the actual variety of the world in your barren skull. I get it, it'd be much simpler for you to think of "everything in Bitcoin".

This simplicity for you has actual costs in the world. If large classes of transactions among which there is no possible cross-ambiguity remain limited to their own context, there's less hassle for everyone involved. Imagine the common occurence of someone sitting in your seat on the airplane. Fortunately, the tickets carry a seat number, which can be compared, and there you go. Imagine instead the Gavinairport where Gavintickets require everyone in the whole airport get out of their planes, single file to the tickets checking office, and have their tickets checked. Every single time someone sits in someone else's seat. And what'd the TSA say to this ? I can almost see them, "there's no reason we can't have it". Of course there isn't, if nobody gives a shit about people or their legitimate interests.

There's no benefit to making everything wait on everything else if large swaths can be readily isolated that'd absolutely never meet. If my blockchain doesn't have to wait for Mali blocks to propagate, and if it doesn't have to to check against Mali doublespends of transactions nobody in Mali could ever be conceivably involved in under any circumstances, then my blockchain is easier to run, to maintain, to debug and so it can provide for the citizens of Mali exactly the only thing they actually want me to provide : a backup value.

They don't perceivably want nor do they conceivably need main chain transactions for every single quarter of cent / West African CFA franc transaction they undertake. What they clearly need and possibly even want is the ability to turn a pile of however many of these they've saved into a few Satoshi. So they can save that

Seven billion people accessing their savings twice a year requires a blocksize of 130 megabytes.

Quote
, so they can buy Trilema credits, so they can do the few and precious things where they actually interact with the world at large. There's absolutely no need to make every single move they take dependent on the actions of far removed parties that couldn't care less about their interests, needs or proclivities.

Yeah, yeah, you're thinking "but Bitcoin is decentralized". Sure, it's a decentralized implementation. But it is an implementation of a centralized concept. Bitcoin is universal money, and that's quite by definition central. Even if implemented in a decentralized manner, as all usable money always has to be implemented, it nevertheless is a centralized thing, by the very nature of what money has to be in order to be money. Now why marry to this an obligation that's burdensome on everyone and not really useful to anyone ?

It's important to realize that Bitcoin is not, and really cannot be, simply a replacement for gold.  Gold is more decentralized.  It is arguably a more efficient store of value.  If Bitcoin is only as usable as gold, it will have failed to realize its potential, in my opinion.

And that was my point.  Bitcoin can be an extremely flexible system, because it is built on sound ideas and moral principles.  There are very few actual technical limitations, only limits to our imagination.

Civil Liberty Through Complex Mathematics
altcoin hitler
Member
**
Offline Offline

Activity: 84
Merit: 10


View Profile
February 01, 2015, 07:09:40 AM
 #113

proposal to rename the forum to Gavincointalk:
https://bitcointalk.org/index.php?topic=942454.0

King of the real Bitcoin Foundation https://bitcointalk.org/index.php?topic=934517.0
peligro
Hero Member
*****
Offline Offline

Activity: 593
Merit: 500


1NoBanksLuJPXf8Sc831fPqjrRpkQPKkEA


View Profile
February 01, 2015, 08:42:08 AM
 #114

It takes me a day from downloading the core wallet and getting it synced. If this proposal increases that time then I am against it.
Chicago
Sr. Member
****
Offline Offline

Activity: 592
Merit: 259


View Profile
February 01, 2015, 08:51:53 AM
 #115

It takes me a day from downloading the core wallet and getting it synced. If this proposal increases that time then I am against it.

The proposed bs changes will lead to increased transaction volume as a convenient artifact of more of them (transactions) being able to occur in a shorter period of time and will definitely add to your initial core wallet sync -- but the greater issue is how much time you will spend flooding your upstream link every 10 minutes blasting the block out to your peers.

You definitely want to read
"How A Bigger Blockchain Is Less Secure And Why Block Size Ain't Gonna Increase Any Time Soon"
http://www.contravex.com/2014/10/07/how-a-bigger-blockchain-is-less-secure-and-why-block-size-aint-gonna-increase-any-time-soon/

Kind Regards,
-Chicago
ujka
Hero Member
*****
Offline Offline

Activity: 630
Merit: 500


View Profile
February 01, 2015, 09:05:21 AM
 #116

1k GB per year is a bigger problem than the 1MB vs 20MB thing imho or at the very least equivalent...

This is misinformation and you are confusing people.

That number isn't true as it assumes we will immediately be processing 84k transactions per block which isn't going to happen.
20GB is the future limit and not the size of most blocks in the future. We are only 600-700tpb at the moment
https://blockchain.info/charts/n-transactions-per-block which means we are typically 15-25% full blocks for 1MB blocks.... do you really expect this to shoot up to 100% full 20MB blocks overnight?.... it will take years to start getting close to those levels.

Additionally, you are completely ignoring the work that is being completed on merkle tree pruning.


This needs repeating again, and again, because many here are thinking that we will have 20MB blocks added to blockchain the very moment the block limit is raised. Calm down. Average block size is now 0.35MB, https://blockchain.info/charts/avg-block-size
Chicago
Sr. Member
****
Offline Offline

Activity: 592
Merit: 259


View Profile
February 01, 2015, 09:09:52 AM
 #117


Additionally, you are completely ignoring the work that is being completed on merkle tree pruning.


@inBitweTrust,

    Nobody wants to prune their local copy of the blockchain except for the people selling a wholesale change in Bitcoin to the users who will 'auto-update' unconditionally.

Kind Regards,
-Chicago
cbeast
Donator
Legendary
*
Offline Offline

Activity: 1736
Merit: 1014

Let's talk governance, lipstick, and pigs.


View Profile
February 01, 2015, 09:20:17 AM
 #118


Additionally, you are completely ignoring the work that is being completed on merkle tree pruning.


@inBitweTrust,

    Nobody wants to prune their local copy of the blockchain except for the people selling a wholesale change in Bitcoin to the users who will 'auto-update' unconditionally.

Kind Regards,
-Chicago
Dear Chicago,

    Please post the election results that qualifies you to speak for anyone but yourself. Otherwise STFU.

With Deepest Sincerity,

cbeast

Any significantly advanced cryptocurrency is indistinguishable from Ponzi Tulips.
Chicago
Sr. Member
****
Offline Offline

Activity: 592
Merit: 259


View Profile
February 01, 2015, 09:27:58 AM
 #119

Dear Chicago,

    Please post the election results that qualifies you to speak for anyone but yourself. Otherwise STFU.

With Deepest Sincerity,

cbeast

    Hard to tempt fate with the 'election' when the risk is a real hard fork of the blockchain due to no-confidence votes from the Miners.

    I suppose the first piece of bloat which will be seen will be in the unspent transaction outputs as everyone begins to spam the blockchain with pennies to include their junk data in the feed.
    How do you propose to prune 'protest' from the blockchain?

Kind Regards,
-Chicago
cbeast
Donator
Legendary
*
Offline Offline

Activity: 1736
Merit: 1014

Let's talk governance, lipstick, and pigs.


View Profile
February 01, 2015, 09:57:04 AM
 #120

Dear Chicago,

    Please post the election results that qualifies you to speak for anyone but yourself. Otherwise STFU.

With Deepest Sincerity,

cbeast

    Hard to tempt fate with the 'election' when the risk is a real hard fork of the blockchain due to no-confidence votes from the Miners.

    I suppose the first piece of bloat which will be seen will be in the unspent transaction outputs as everyone begins to spam the blockchain with pennies to include their junk data in the feed.
    How do you propose to prune 'protest' from the blockchain?

Kind Regards,
-Chicago
As yet there isn't enough data either way to make an informed decision. Perhaps you can offer empirical evidence that pruning a blockchain weakens blockchain security. Until then, your opinion is only that.

Any significantly advanced cryptocurrency is indistinguishable from Ponzi Tulips.
Pages: « 1 2 3 4 5 [6] 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 ... 125 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!