Block chain size/storage and slow downloads for new users

[fryarminer]

So I just built my new Hack Pro - it's a beast. And for the past two days now I've been trying to synchronize Bitcoin QT and I'm at about 10%.
I figured I'd see what Bitcointalk would say about it, and I wondered if anyone ever thought that there might be a time when the blockchain might get too ridiculously large, and what was my surprise to see this thread as the first thread listed when I logged in!

K, I'm going to go back and read it now.

[tvbcof]

Bitcoin is decentralized... the antithesis of bitcoin is really centralized websites and authorities.

As rapidly increasing Bitcoin utilization causes bloat and complexity, centralization is a natural consequence.  Already we see centralized constructs reached for as a problem solving means since it is the quickest and most obvious way to solve many problems.  (Leveraging state issued passports and SSL cert authorities for instance.)

I posit that a more healthy way to go is to jealously and rigorously guard against centralization at a 'first tier' or 'core' or 'native' level.  And to do this by actively embracing and facilitating centralization at a 2nd tier level.  This to take the load of growth as well as the risk of additional bells and whistles at this second tier level where failure does not result in a collapse of the entire solution.

Ultimately what I want is an unassailable 'native' Bitcoin which looks much like we see today.  I'll keep the bulk of my value in deep off-line storage relying on nothing more than the 'first tier' which is light weight and can realistically be operated by independent enthusiast who risk little capital by operating the system.

For general day-in/day-out use I'll choose a handful of promising 'second tier' organizations who seem to be trustworthy and risk a certain fraction of my wealth with them.  Currently I do this with Coinbase and Blockchain.info for instance.  If they fold or get shut down I may lose, but I lose no more than I choose to risk and my nest-egg is safe with my plain old no-frills deep storage wallet which I rarely dig into.

The alternate of trying to support an economy which could basically need to grow without limits in a native 'first tier' Bitcoin layer is very dangerous and I believe almost certain to fail.  And fail because it will almost certainly centralize out of necessity and thus become more prone to successful attack.

A multi-tiered system still relies on Bitcoin as the foundation upon which it rests, and is still available in native form when really needed so evolving in that direction is not 'selling out'.  It's simply a logical way to scale and grow while remaining hardened against centralization and the brittleness and risks that that condition brings.

[roslinpl]

This, combined with off-chain transactions like https://inputs.io, gives me a lot of hope for Bitcoin.

Yes, seems like problems are getting solved. I am hopeful too...

there will be always war between hackers and programmers

Bitcoin will uprgade, and those who use boitcoin network will learn how to do it better.

[chekhov]

Thanks everyone. Learning a lot from this thread.

Just for the reference - finally downloaded the whole blockchain tonight.

So, for Feb. 26, 2014 we have:

# of blocks: 287930
19.267088 GB in .blocks folder.

Soon approaching 20Gb.

[E.exchanger]

Thanks for the information about SPV mike. Great effort 

[awesomeami]

slow downloads for new users

use this: [ANN] Bitcoin blockchain data torrent

This is a torrent of the publicly available bitcoin blockchain data.

Bitcoin-Qt/bitcoind version 0.7.1 and later supports a special import feature:  If the file "bootstrap.dat" is found in the bitcoin data directory, it will validate and import all blockchain data found in that file.  The following torrent presents a bootstrap.dat file for that feature.

[Meuh6879]

at this day, i have re-installed a bitcoin-QT on a machine : 17,5Go take by the blochchain folder (that i have separated from the bitcoin folder).
well, with 500Go of hard drive (portable PC), it's not a big deal... and 24h to retrieve the blockchain, you must have a 10MBit/s to do that.

[awesomeami]

and 24h to retrieve the blockchain, you must have a 10MBit/s to do that.

20GB @ 1Mbit = 45 hours - in theory - in real it could be 5 hours on 10mbit - the only problem is fast CPU&enough RAM, no other things doing and fast (not busy HDD) - so lets say 6-10 hours on slow 3yo PC

So we just need much more 1Gbit nodes - I plan to get 3 in EU + 2 in US - June??
And then it only depends on CPU ...

20*8*1024/1/3600=45

You can replace "1" with other number of megabits
20*8*1024/1/3600=45

explanation:
20GB*8(byte to bits)*1024(GB to MB)/1(conn speed)/3600(seconds to hours)=45

So IMHO - less full nodes (not every user needs full blockchain - webwallits, light clients, other projects, offline transactions ...) and 100/100 Mbit is standard for servers - so again - i c no problem.
Like 1hour to sync - some mirrors i am syncing 2 days on 1 Gbit ... - but I do it once a 4 years?

[makiyo_love]

what about electrum how does this compare?

[freddyfarnsworth]

Thanks everyone. Learning a lot from this thread.

Just for the reference - finally downloaded the whole blockchain tonight.

So, for Feb. 26, 2014 we have:

# of blocks: 287930
19.267088 GB in .blocks folder.

Soon approaching 20Gb.

My two bits on blockchain size, 20gb at a time when you will have trouble buying a 1tb drive.
Seems to be scaled well enuf. 2tb and up are all the rage now. 20gb is nothing.

Network speed and bandwidth is another story. We are way behind on that. By now home users should be at OC3 capability.
We are far far behind that. Wireless ?? speeds/bandwidth are a joke.

Blockchain is designed well. Well thought out.

[irrational]

Thanks everyone. Learning a lot from this thread.

Just for the reference - finally downloaded the whole blockchain tonight.

So, for Feb. 26, 2014 we have:

# of blocks: 287930
19.267088 GB in .blocks folder.

Soon approaching 20Gb.

My two bits on blockchain size, 20gb at a time when you will have trouble buying a 1tb drive.
Seems to be scaled well enuf. 2tb and up are all the rage now. 20gb is nothing.

Network speed and bandwidth is another story. We are way behind on that. By now home users should be at OC3 capability.
We are far far behind that. Wireless ?? speeds/bandwidth are a joke.

Blockchain is designed well. Well thought out.

1 and 2 TB drives IF (and big IF) you're assuming people prefer archaic rotating hard disks. For a LOT of applications people have moved to solid state drives. Having a 20GB mess of junk on my 400GB SSD is EXPENSIVE.

If Bitcoin is to attract the masses, we (collectively) cannot assume that everyone uses a computer the same way.

For me, the blockchain has gotten STUPID the past year and a half.

[roslinpl]

what about electrum how does this compare?

You can use electrum without full chain, but if you want to download full chain it is always the same size for any client
(size growing constantly)

[freddyfarnsworth]

Thanks everyone. Learning a lot from this thread.

Just for the reference - finally downloaded the whole blockchain tonight.

So, for Feb. 26, 2014 we have:

# of blocks: 287930
19.267088 GB in .blocks folder.

Soon approaching 20Gb.

My two bits on blockchain size, 20gb at a time when you will have trouble buying a 1tb drive.
Seems to be scaled well enuf. 2tb and up are all the rage now. 20gb is nothing.

Network speed and bandwidth is another story. We are way behind on that. By now home users should be at OC3 capability.
We are far far behind that. Wireless ?? speeds/bandwidth are a joke.

Blockchain is designed well. Well thought out.

1 and 2 TB drives IF (and big IF) you're assuming people prefer archaic rotating hard disks. For a LOT of applications people have moved to solid state drives. Having a 20GB mess of junk on my 400GB SSD is EXPENSIVE.

If Bitcoin is to attract the masses, we (collectively) cannot assume that everyone uses a computer the same way.

For me, the blockchain has gotten STUPID the past year and a half.

There are ways to get around the stupid small size of the SSD's. Blockchain for some reason with the OLD Stupid stuff, seems not a issue at all.

[jsgayo]

Thanks! Much appreciated!

[irrational]

There are ways to get around the stupid small size of the SSD's. Blockchain for some reason with the OLD Stupid stuff, seems not a issue at all.

I think most people here are either in IT or know there way around a computer. So, I know you know how to deal with this, as do I.

I guess my point really was that my father/mother/uncle/aunt wouldn't, and if we can't make bitcoin approachable for the non-techie then bitcoin has failed.

I'm hopeful since alternative clients exist, and hopefully they take off. But, the point still remains that Bitcoin-Qt is "marketed" as the "official" client and is the one most people start with. First impressions are very important.

[freddyfarnsworth]

There are ways to get around the stupid small size of the SSD's. Blockchain for some reason with the OLD Stupid stuff, seems not a issue at all.

I think most people here are either in IT or know there way around a computer. So, I know you know how to deal with this, as do I.

I guess my point really was that my father/mother/uncle/aunt wouldn't, and if we can't make bitcoin approachable for the non-techie then bitcoin has failed.

I'm hopeful since alternative clients exist, and hopefully they take off. But, the point still remains that Bitcoin-Qt is "marketed" as the "official" client and is the one most people start with. First impressions are very important.

Sell em on multibit, it rips thru the blockchain sync, in seconds.
Selling a faster wallet type is your job.

[pandalion98]

So much info.

[roslinpl]

"Block chain size/storage and slow downloads for new users"

IMO - any new game like example Thief need about 25gb... so you are telling me blockchain size is huge and scary?
in next 5 years games will weight 100gb perhaps. Till 1995 biggest game I had was recorded on two 1.4mb disks...

so do not tell me anything about blockchain size.
It is SMALL. and maybe one day it will be big.

You can buy 1TB HDD for less than 100$ ...
Don't tell me Blockchain is big.

Anyway. I haven't got blockchain on my pc and I am still bitcoin user. How come? ( don't anwer, we all know )

You do not need qt to be a bitcoiner.

[iamibo]

Anyway. I haven't got blockchain on my pc and I am still bitcoin user. How come? ( don't anwer, we all know Tongue )

You do not need qt to be a bitcoiner.

[tvbcof]

Anyway. I haven't got blockchain on my pc and I am still bitcoin user. How come? ( don't anwer, we all know Tongue )

You do not need qt to be a bitcoiner.

I don't have PayPal's database on my computer, but I'm a PayPal users.  Nor do I have a printing press, but I'm a $USD user.

One of the most interesting things about Bitcoin in the early days was that my router served as a perfectly adequate first-class peer in what was then a Peer-2-Peer solution.  I had to rely on nobody but myself to fully validate transactions.

Everyone (or at least a fraction of us who have some comprehension of computer science) knew that there would be a shift here.  It remains an open question if the end result looks more like early Bitcoin or more like a significantly centralized construct like our mainstream banking solution.  Time will tell.