FACT CHECK: Bitcoin Blockchain will be 700GB in 4 Years

[coins101]

Do to continuous usage of bitcoin transaction it's not surprising to hear that blockchain's size will be 700GB
 It's of normal size. Even a medium e commerce site would generate 1TB data in 1 year. As I had worked in a live project so it's of normal size.

We are talking about Bitcoin generating 1TB every month, at current levels. So at 700GB it would generate 10TB per month.

[Lauda]

I have no idea how this thread didn't catch my eye yet. I guess there's too much crap in the section.

in some ways this is very technical so it really requires a lot of analytical thinking.

No, all that this graph and 'calculation' requires is a brain.

But as a not so technical type of person like me i also know that this thing would really happens but as we can see processor's of the computers are also innovating right? so i think their system could still afford to do so.

Speculating on an ever dying Moore's "law" (which isn't even a law in the traditional sense) can and will lead to trouble.
@OP: There's too much crap in this thread that isn't worth reading; please start using self-moderated threads in the future. That said, you're assuming a growth rate in % or have you done your calculations based on X average block size per day/week/month? Did you factor in Segwit? If not, you will have to update that chart.

[coins101]

I have no idea how this thread didn't catch my eye yet. I guess there's too much crap in the section.

in some ways this is very technical so it really requires a lot of analytical thinking.

No, all that this graph and 'calculation' requires is a brain.

But as a not so technical type of person like me i also know that this thing would really happens but as we can see processor's of the computers are also innovating right? so i think their system could still afford to do so.

Speculating on an ever dying Moore's "law" (which isn't even a law in the traditional sense) can and will lead to trouble.
@OP: There's too much crap in this thread that isn't worth reading; please start using self-moderated threads in the future. That said, you're assuming a growth rate in % or have you done your calculations based on X average block size per day/week/month? Did you factor in Segwit? If not, you will have to update that chart.

Hey, thanks for chipping in. I'm concerned that if I apply and show good moderator skills, I might get asked to join the gang  

Anyway, to your substantive point. SegWit is not yet implemented, so I want to see the evidence before making a change to my assumptions on the ~4% monthly growth rate.

Another, for instance to counter your point. If multi-sig keeps growing in popularity, does that wipe out the benefits of SegWit, making the need for a block size increase more likely?

However, I have put out a challenge on the assumptions I have made...and so far you have made the best challenge. Thanks.

[Lauda]

Hey, thanks for chipping in. I'm concerned that if I apply and show good moderator skills, I might get asked to join the gang  

I wouldn't worry about it, that isn't likely to happen just because you run self-moderated threads.

Anyway, to your substantive point. SegWit is not yet implemented, so I want to see the evidence before making a change to my assumptions on the ~4% monthly growth rate.

False. Segwit is implemented in 0.13.0, there are just no activation parameters (i.e. it is not active). They're making some final changes for 0.13.1 which should be released soon.

Another, for instance to counter your point. If multi-sig keeps growing in popularity, does that wipe out the benefits of SegWit, making the need for a block size increase more likely?

Well, the same could be asked for a block size increase, could it not (e.g. what if bigger average TX size becomes more popular)? Multisig with P2SH should be safer with Segwit, so I expect even more usage as well.Actually, if more people used multi-sig Segwit should be able to provide a bit more headroom IIRC.
It is expected to see a capacity of anywhere between 1.6 MB to 2 MB on average. The exact numbers are debatable, and yet to be seen in practice (you can/could check testnet blocks).  Someone did the math a few months back on the mailing list; I used to quote that, but can't find it at this time. If I do, I'll post it here.

[Meuh6879]

segwit will rebuild the blocks ... after 1 year.
and then, the whole local blockchain will be reduce.
and then, olds clients will not be able to connect to eradicate 0-confirmation trap.

it's a revolution.

Local blockchain will be at 30 Go at the end of the 2017.

it's really a revolution.

[Lauda]

segwit will rebuild the blocks ... after 1 year.
and then, the whole local blockchain will be reduce.
and then, olds clients will not be able to connect to eradicate 0-confirmation trap.

it's a revolution.

Local blockchain will be at 30 Go at the end of the 2017.

What are you talking about? I'm pretty sure that nothing like that is going to happen.

[Vhern]

And what about 10, 20 years?

1000TB?? Anyone have a bigger projection about this info?

[Lauda]

And what about 10, 20 years?

1000TB?? Anyone have a bigger projection about this info?

This question is pointless and so would the follow up questions be (30, 40, 50 years, etc.). Just do your own math, it's really simple: average expected block size x number of blocks a day x number of days a year x number of years that you want. You won't get the most accurate projection, but you will get a rough picture and will avoid redundant questions.

[rizkyhiw]

And what about 10, 20 years?

1000TB?? Anyone have a bigger projection about this info?

Imagine 100 gigabytes 20 years ago, terrifyingly large, not such a big deal of a file today, in 20 years I imagine a similar issue.

yeah 700gb in 20 years later maybe like 20gb now?
as technology growing so faster , you will not to think 700gb as a large size then
people will find another technology to make the 700gb as cheap as possible 
really not a big deal .

[Yakamoto]

And what about 10, 20 years?

1000TB?? Anyone have a bigger projection about this info?

Imagine 100 gigabytes 20 years ago, terrifyingly large, not such a big deal of a file today, in 20 years I imagine a similar issue.

To be honest I have found 100 gigs to be fairly large to download, however it isn't all that big of a file once it is downloaded and ready to go. It might be something like a 1/10th of a terabyte, but that isn't that bad is you do enough for data management and so.

In 20 years, chances are 20 terabytes will just be moderate amounts of data.

[MONKEYJUNK]

And what about 10, 20 years?

1000TB?? Anyone have a bigger projection about this info?

Imagine 100 gigabytes 20 years ago, terrifyingly large, not such a big deal of a file today, in 20 years I imagine a similar issue.

To be honest I have found 100 gigs to be fairly large to download, however it isn't all that big of a file once it is downloaded and ready to go. It might be something like a 1/10th of a terabyte, but that isn't that bad is you do enough for data management and so.

In 20 years, chances are 20 terabytes will just be moderate amounts of data.

Hope the technology can raise exponentially too however it's 2016 and I have only 300gb hd... with a 5mb internet lol

Well, in the future only big miners will mining bitcoin...

[Lauda]

Imagine 100 gigabytes 20 years ago, terrifyingly large, not such a big deal of a file today, in 20 years I imagine a similar issue.

We can't predict technological growth that far out into the future.

To be honest I have found 100 gigs to be fairly large to download, however it isn't all that big of a file once it is downloaded and ready to go. It might be something like a 1/10th of a terabyte, but that isn't that bad is you do enough for data management and so.

It isn't just the download, it's also the validation time. A full client may occasionally get broken for whatever reason, and that usually requires a reindex which takes a painful amount of time on non high-end hardware.

Well, in the future only big miners will mining bitcoin...

This is also known as ignoratio elenchi (irrelevant conclusion).

[Wendigo]

Well in 5 years or more the production process will be much more refined than it is right now and the price of electronics will be considerably cheaper - I would say more than 50% cheaper than it is now. For example the price of SSDs almost halved in recent years so I would assume the price of HDDs in 5 or 10 years from now will be more or less dirt cheap. Plus not everyone is running or planning to run a Bitcoin node on their home computer. Heck you could buy a 5TB home server right now and not break the bank.

[coins101]

Well in 5 years or more the production process will be much more refined than it is right now and the price of electronics will be considerably cheaper - I would say more than 50% cheaper than it is now. For example the price of SSDs almost halved in recent years so I would assume the price of HDDs in 5 or 10 years from now will be more or less dirt cheap. Plus not everyone is running or planning to run a Bitcoin node on their home computer. Heck you could buy a 5TB home server right now and not break the bank.

It's not Moore's Law that's the issue.

It's Nielsen's Law that's the issue.

Crossing the Rubicon:


https://www.reddit.com/r/Bitcoin/comments/56bths/to_all_the_people_running_full_nodes_we_thank_you/

[puremage111]

wow it will take me almost 1 week to finish the download i suppose rofl or even more than that

[ObscureBean]

wow it will take me almost 1 week to finish the download i suppose rofl or even more than that

And that's the problem, not everyone will be able or willing to download the full blockchain. True decentralization can only be achieved if everyone has the same opportunity to download the full blockchain, if only rich countries can afford to run a full node, already it's the beginnings of centralization. Basically back to square one.

[yayayo]

wow it will take me almost 1 week to finish the download i suppose rofl or even more than that

And that's the problem, not everyone will be able or willing to download the full blockchain. True decentralization can only be achieved if everyone has the same opportunity to download the full blockchain, if only rich countries can afford to run a full node, already it's the beginnings of centralization. Basically back to square one.

Yes it may become a problem. And the problem of bandwidth is much more pressing than that of storage, because most people in the industrialized countries are greatly overestimating the real upload speeds available worldwide. Many providers - even in countries with very good infrastructure - have explicit or implicit bandwidth limits and in many developing countries, sufficient bandwidth is either not reliably available or extremely expensive (sometimes even at Western standards - without taking into account the significantly lower buying power of the population in these countries).

So to preserve decentralization, Bitcoin must be scaled in the least resource expensive way possible. In my opinion, segregated witness and lightning networks are a step in the right direction, because they address the issue of microtransactions in a responsible way. Trying to store all transactions into blocks in the current format by allowing unlimited block sizes would be the end of a decentralized Bitcoin network.

ya.ya.yo!

[Meuh6879]

wow it will take me almost 1 week to finish the download i suppose rofl or even more than that

On 10MBit/s line ... with an (very) old simple computer.
https://bitcointalk.org/index.php?topic=1594650.msg16039808#msg16039808

[Meuh6879]

Crossing the Rubicon:


https://www.reddit.com/r/Bitcoin/comments/56bths/to_all_the_people_running_full_nodes_we_thank_you/

  and the number of connexion is ... ?
full nodes MUST be set at a limited number of connexion.

it's a basic.

[Shiroslullaby]

I've done a bit of reading on segwit and how it would reduce transaction times, but I never read anything about it reducing the total size of the blockchain.
If someone could give a quick explanation on how this works it would be greatly appreciated.