Bigger blocks coming in release 0.11

[DarkHyudrA]

Even though getting a 1TB HDD is cheap today, what gets me worried is about the network.
Downloading a 40GB+ file isn't easy in the whole world(tons of places and countries have poor quality ISP with low speeds).
Light weight clients are nice and even I use them, but the blockchain is getting more and more centralized, what kills the initial concept o Bitcoin.

[Q7]

I think this has been discussed in-depth in the other thread and has been long in the planning stage. So I would say it's about time. Anyway it is something that we would need to prepare for the future because eventually we have to do it no matter how. If we want to go mainstream, increasing the size to absorb more transactions per block is the only way.

[R2D221]

Light weight clients are nice and even I use them, but the blockchain is getting more and more centralized, what kills the initial concept o Bitcoin.

I don't that “kills” the initial concept of Bitcoin at all.

Long before the network gets anywhere near as large as that, it would be safe
for users to use Simplified Payment Verification (section 8) to check for
double spending, which only requires having the chain of block headers, or
about 12KB per day.  Only people trying to create new coins would need to run
network nodes.  At first, most users would run network nodes, but as the
network grows beyond a certain point, it would be left more and more to
specialists with server farms of specialized hardware.  A server farm would
only need to have one node on the network and the rest of the LAN connects with
that one node.

The bandwidth might not be as prohibitive as you think.  A typical transaction
would be about 400 bytes (ECC is nicely compact).  Each transaction has to be
broadcast twice, so lets say 1KB per transaction.  Visa processed 37 billion
transactions in FY2008, or an average of 100 million transactions per day. 
That many transactions would take 100GB of bandwidth, or the size of 12 DVD or
2 HD quality movies, or about $18 worth of bandwidth at current prices.

If the network were to get that big, it would take several years, and by then,
sending 2 HD movies over the Internet would probably not seem like a big deal.

Satoshi Nakamoto

[R2D221]

one small step for future upscaling one Giant Leap towards complete centralisation 

How is it a "Giant Leap" to complete centralization? The blocks won't instantly become 20mb and by the time they are it won't be a big difference that storage size will be larger than it is today. We are talking a full Terabyte a year at every block being 20mb, this would have to be one hell of a busy network and at the standard. For 5 years of block chain it would only cost 120$ http://www.newegg.com/Product/Product.aspx?Item=9SIA5AD2RS3903, if you don't think in 5 years the price of 5tb will be A LOT cheaper just look at my previous post and again this would have to be at full block capacity which it won't be for quite some time. The idea of nodes being centralized is only theory and it doesn't hold much water.

But no! The problem is not storage, but bandwidth. At least that's what many claim.

[Klestin]

But no! The problem is not storage, but bandwidth. At least that's what many claim.

Bandwidth becomes a problem at over 1 MB every 10 minutes?  My moderate Internet connection can download 1 MB every 1/8th of a second.

[R2D221]

But no! The problem is not storage, but bandwidth. At least that's what many claim.

Bandwidth becomes a problem at over 1 MB every 10 minutes?  My moderate Internet connection can download 1 MB every 1/8th of a second.

Yes, but apparently there are people with 56k connection that want to run full nodes...

And now that I think about it, it's still enough to handle 20 MB every 10 minutes, isn't it?

[Buffer Overflow]

But no! The problem is not storage, but bandwidth. At least that's what many claim.

Bandwidth becomes a problem at over 1 MB every 10 minutes?  My moderate Internet connection can download 1 MB every 1/8th of a second.

Yes, but apparently there are people with 56k connection that want to run full nodes...

And now that I think about it, it's still enough to handle 20 MB every 10 minutes, isn't it?

Yeah I remember those 56k modem days. That's 56000 bits per second, not bytes. It wouldn't cope with 20MB blocks.

[Foxpup]

Yes, but apparently there are people with 56k connection that want to run full nodes...

And now that I think about it, it's still enough to handle 20 MB every 10 minutes, isn't it?

It isn't. It's only 2.5 MB (V.90) or 3 MB (V.92) every 10 minutes, assuming the node uploads as much as it downloads.

[cellard]

But didn't satoshi predict that this wouldn't be a problem? I remember reading emails of satoshi where he said it would never be a problem because technology would catch up as transaction volume grows.

[DarkHyudrA]

But no! The problem is not storage, but bandwidth. At least that's what many claim.

Bandwidth becomes a problem at over 1 MB every 10 minutes?  My moderate Internet connection can download 1 MB every 1/8th of a second.

Yes, but apparently there are people with 56k connection that want to run full nodes...

And now that I think about it, it's still enough to handle 20 MB every 10 minutes, isn't it?

The problem is, even a 1Mb needs 3 minutes to download one 20MB block.
Uploading a block with such connection takes more than 10 minutes, and downloading even only the headers will be frustrating.

Although today I have a 10Mb connection(its decent, could be better), I do worry about other people where internet connection is expensive.

[Klestin]

The problem is, even a 1Mb needs 3 minutes to download one 20MB block.
Uploading a block with such connection takes more than 10 minutes, and downloading even only the headers will be frustrating.

Although today I have a 10Mb connection(its decent, could be better), I do worry about other people where internet connection is expensive.

Why worry?  Nobody is forced to run a full node.  The number of users who have bandwidth and storage to spare is nontrivial.  We are in no danger of a lack of users willing to run full nodes due to those concerns.

[unamis76]

I think it's about time we change. we need to be future proof, and this is one big step towards that.

As for storage, the GB per $ price is quite cheap and is declining in the long run, for both mechanical and flash disks. I think the blockchain size is a bit of a non-issue. Also, the fact that the new limit is set at 20mb won't suddenly make the blockchain triple in size. The block sizes will remain the same. Our usage and the amount of transactions per block are the ones who will make the blockchain bigger, and I think more transactions is a good thing...

[inBitweTrust]

Reminder-

https://bitcointalk.org/index.php?topic=532.msg6306#msg6306

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.  The more burden it is to run a node, the fewer nodes there will be.  Those few nodes will be big server farms.  The rest will be client nodes that only do transactions and don't generate.

Quote from: bytemaster on July 28, 2010, 08:59:42 PM
Besides, 10 minutes is too long to verify that payment is good.  It needs to be as fast as swiping a credit card is today.
See the snack machine thread, I outline how a payment processor could verify payments well enough, actually really well (much lower fraud rate than credit cards), in something like 10 seconds or less.  If you don't believe me or don't get it, I don't have time to try to convince you, sorry.
http://bitcointalk.org/index.php?topic=423.msg3819#msg3819

I don't predict that we will need to become this centralized but this was the initial plan.
One reason-Merkle tree Pruning scheduled for next bitcoin release(Core 0.11) allowing full nodes that are 1GB.

[HI-TEC99]

If 20MB block size comes with pruning feature, i'm OK for this. 
i want a counterparty ... 

Satoshi said it would make sense to introduce pruning when the blockchain got too big. If a 20MB block size is going to be introduced it would be stupid to delay introducing pruning too.

[Mikestang]

but the blockchain is getting more and more centralized, what kills the initial concept o Bitcoin.

Uh, what?  Substantiate your claim, provide proof.  There are over 6100 full nodes, and more added every day.  Making blocks bigger has no correlation with "centralization".

[Bitcoin_BOy$]

Bigger blocks I think the block reward decrasing every year , can you please explain more what's coming

Kind Of Respect ,
Bitcoin Boy .

[jl2012]

but the blockchain is getting more and more centralized, what kills the initial concept o Bitcoin.

Uh, what?  Substantiate your claim, provide proof.  There are over 6100 full nodes, and more added every day.  Making blocks bigger has no correlation with "centralization".

My home PC is one of those 6100 and I will have absolutely no problem to maintain it even the block size becomes 100MB

[jl2012]

Bigger blocks I think the block reward decrasing every year , can you please explain more what's coming

Kind Of Respect ,
Bitcoin Boy .

It's about having more data stored in the block. It has noting to do with the block reward

[inBitweTrust]

Bigger blocks I think the block reward decrasing every year , can you please explain more what's coming

Kind Of Respect ,
Bitcoin Boy .

Block reward decreasing approx every 4 years...this proposed fork is intended to allow more transactions to be processed per second to help bitcoin grow and pay for the security.

[redsn0w]

For me it would be a nice feature, and it is very needed if bitcoin 'wants' to be mainstream. Because if we compare it with the credit card system we will see that 7 transactions per second - against - 2'000 transactions  per second (visa). In this way we will have a lot of transaction in one block and it will speed up the send of bitcoin.


I'm waiting to see this feature in the next releases and I am sure the majority will choose the right thing .