I don't see why big blocks are a problem, even 10 MB blocks right now aren't.

[Jet Cash]

It's not blocksize that is the bottleneck. It's block generation time. Bitcoin needs to find a way to generate smaller blocks much faster if it wants to become a major payment system rather than a store of wealth.

[Hydrogen]

Big blocks do not prevent DDoS spam. If someone spams transactions, increasing block size does nothing to stop it.

Imagine a thief smashes the window of a car to steal something inside it. Then big blockers say, we need bigger glass windows in cars to prevent a scaling problem.

That's what the block size debate amounts to.

[franky1]

Ok yeah. That is something i did not think of. Any idea on the specs needed to process a 1mb transaction with a lot of inputs? I didnt think it was that significant for the modern day processors but i dont know that much about that side.

Lauda has already posted some data - RAM seems even more a show-stopper for big blocks on consumer hardware than CPU. If you want the details, the source, a Bitfury study, is here. (It's from 2015, though; but also 2 years later it seems everything >8MB blocks is simply too much.)

lol the stats lauda / bitfury have are out of context

firstly its stupidly silly maths to promote "sgwit linear"

without segwit, changing things like
maxtxsigops to 4k has REDUCED the ram cpu usage

in the future dropping it again to 2k can further reduce the ram/cpu usage, yet even with 'bigger blocks' you can reduce cpu/ram usage aswell as over time having an average computer that has more ram in it. EG Raspi1 vs Raspi3.. vs todays basic desktop pc

[iluvpie60]

Ok yeah. That is something i did not think of. Any idea on the specs needed to process a 1mb transaction with a lot of inputs? I didnt think it was that significant for the modern day processors but i dont know that much about that side.

Lauda has already posted some data - RAM seems even more a show-stopper for big blocks on consumer hardware than CPU. If you want the details, the source, a Bitfury study, is here. (It's from 2015, though; but also 2 years later it seems everything >8MB blocks is simply too much.)

I  probably in favor of banning any transaction under 100satoshi as it is not worth almost anything currently.[...]
Is thsir real value in doing 1sat transactions beyond spam attacks?

That would be very probably seen by the Bitcoin community as "censorship", and the block size debate shows us that it would not be easy to lift the restriction in the future even if it's needed for several use cases.

And even today there are some valid use cases for transactions that even have no value at all (OP_RETURN transactions), like a decentralized notary system like Factom, coloured coins, etc..  See the usage cases here: in this study.

Thanks for the link to that, I have been out of the loop on BTC info and tech related details for a few years. This is a great read and I will dive in to this.

From what I have found on the web for minimum requirements for current, it seems almost everyone can run a full node as current requirements are quite pitiful in terms of power/internet speed/hardware involved. I am going to look into what the cost of this will be in electricity and other factors. I already have a 4TB HDD not really being used, have a 4 physical core (8hyperthreaded) 3.1 GHZ processor and a pretty decent GPU + reliable internet + 16 GB ram.

I see that I am underestimating the current hardware used for the network. If 95% of nodes would be eliminated at 8MB blocks that means they are really scraping by with pitiful outdated computers for the most part. I would have thought in our community we were running 1,000s of nodes and that we all pretty much had gaming rigs or high powered rigs.

I hope that more people run nodes is all I have to say. Really not advocating for anything other than onchain solutions.




Minimum Requirements
Bitcoin Core full nodes have certain requirements. If you try running a node on weak hardware, it may work—but you’ll likely spend more time dealing with issues. If you can meet the following requirements, you’ll have an easy-to-use node.

Desktop or laptop hardware running recent versions of Windows, Mac OS X, or Linux.

145 gigabytes of free disk space

2 gigabytes of memory (RAM)

A broadband Internet connection with upload speeds of at least 400 kilobits (50 kilobytes) per second

An unmetered connection, a connection with high upload limits, or a connection you regularly monitor to ensure it doesn’t exceed its upload limits. It’s common for full nodes on high-speed connections to use 200 gigabytes upload or more a month. Download usage is around 20 gigabytes a month, plus around an additional 140 gigabytes the first time you start your node.

6 hours a day that your full node can be left running. (You can do other things with your computer while running a full node.) More hours would be better, and best of all would be if you can run your node continuously

[elite3000]

Not a problem for average?

For the average people of the world is too much.

Also you'll have the hardware to support the bigger blocks, like memory and processor.

[franky1]

It's not blocksize that is the bottleneck. It's block generation time. Bitcoin needs to find a way to generate smaller blocks much faster if it wants to become a major payment system rather than a store of wealth.

there can be side services to handle the niche market of 'instant spend' .. but that is no reason to try doomsday the blockchain using propaganda to stifle natural growth.

segwit cannot solve the promises it promises. and you cant segwit a segwit after segwiting it.. so the last 2 years of promoting and pushing for segwit has been nothing more then an empty gesture

itss time we move passed "segwit hope" and realise the trths about it.. and start actually getting bitcoins real and productive code running such as dynamic blocks (there are many different brands with different proposals so dont shout BU as your doom cry)

[Syke]

Absolute nonsense. The hardware used on Steam is an overestimation for the hardware used for nodes. In other words, the situation would actually be worse than what is claimed by this paper.

You may be right, you may be wrong, I'm just asking for actual evidence of running nodes. The paper presented no evidence about the hardware nodes are running on. What's your evidence?

[FandangledGizmo]

OP, you're clearly uninformed and unqualified to make these kinds of decisions (same as everyone else promoting this stupid idea).


https://twitter.com/TuurDemeester/status/881851053913899009
Do you want a peer-to-peer network, or a proxy-to-proxy network?

That is a compelling study and I'm definitely concerned about centralisation.

Why is 'centralization of mining' not the no.1 topic that the community is trying to address everyday?

Over the years we've had situations where it would only take control of just two mining pools to get over 51% hashrate, (even just one once). Which meant regardless of nodes you'd only need to get to two people to 'attack' Bitcoin, crash it's value to near zero & destroy confidence in the system for a couple of years even if hashers left the attacking pools.

Is node centralisation really a more dangerous/centralising attack/control vector than the problem Bitcoin already has with mining pools and large miners?

[Jet Cash]

Hardware isn't a problem. I run a full node on a cheap netbook with an external SSD, and I do it over McDonalds free WiFi. I've got anothe node on a Windows 10 notebook, and that is also kept up to date over free WiFi.

However much you mess about with blocksize, with an average 10 minute generation time, and a couple of confirmations to verify the transaction, you are looking at average times of 30 minutes. That isn't competitive for a modern money transfer system.

[franky1]

Internet connection speed isn't biggest problem here, the biggest problem in this case are metered bandwitch and low-end/cheap computer won't able run bitcoin full nodes anymore. Mini PC such as Raspberry Pi 3B will struggle since they have small RAM.
It's no problem if most of people who run full nodes can afford better computer to run full nodes.

solution for users independantly
(same solution for torrent seeds)
reduce number of leachers(connections) = less bandwidth draw*

solution for dev implementations
reduce sigops and max txbytes = less ram/cpu demand

afterall with todays 4ktxsigops with 20k max block sigops
and segwits 16k txsigops 80k blocksigops.. that means a base block can be filled with just 5 tx's

and the max tx bytes of 100kb = 10 tx fills base block..

WHO THE F*CK deserves to have 10%-20% of the blocksize for just 1 tx..!!!!
reduce the greedy stupid, artificially high numbers, meaning transactions become leaner. and you will find that bloat, bottlenecking and processing time REDUCE all while allowing the block size to increase.

*i see many people with 'capped' internet that have 50+ connections. i facepalm them and tell them to learn "6 degree's of separation" and reduce their connection count.
not everyone needs to be a 'supernode' (multiple dozens of connections)

[jbreher]

Mini PC such as Raspberry Pi 3B will struggle since they have small RAM.
It's no problem if most of people who run full nodes can afford better computer to run full nodes.

RaspPi users can fuck right off.

If you want to be an integral part of the most significant advance in financial technology since the 1400's, use a real machine. Don't cripple the network just because your cheap ass doesn't want to meet the required expenditure.

That is a compelling study and I'm definitely concerned about centralisation.

That is an ancient study built upon faulty premise.

[Milan944]

This is gonna be a big problem in the future,if they don't upgrade it.

[Last of the V8s]

..study built upon faulty premise.

what premise is that?

[jbreher]

..study built upon faulty premise.

what premise is that?

That Steam gamers and their machines are equivalent to Bitcoin node operators and their machines.

aka ...I coulda had a V8, but all I got is this lousy rice burner - but at least it's got a fart pipe

[d5000]

Hardware isn't a problem. I run a full node on a cheap netbook with an external SSD, and I do it over McDonalds free WiFi. I've got anothe node on a Windows 10 notebook, and that is also kept up to date over free WiFi.

That is the current situation. But there are two things to consider:

1) Blocks are about ~1MB only in the last 1-2 years. The years before they were much smaller. So in one year the blockchain - only with 1MB blocks - will already be much heavier than now - and also much more difficult to syncronize.

2) Segwit alone will already give us blocks with a size from 1,7 to 2,x MB (4MB max, but I don't believe we'll see that). If Segwit2x comes, then we'll potentially have 3,4 to 4,x MB blocks with a maximum of 8 MB. If Bitcoin experiences a yearly 40-50% transaction growth like until now, then we will have these blocks full in about 2-3 years, and the blockchain then will be at least 3 times as heavy than now and will keep growing. Again, storage isn't the problem - the problem is the initial block download/verification (first sync) and the propagation and verification of new blocks. Now do the math for 10MB+ blocks.

Big blockers are hoping for hardware improvements, that's true - but I would say, better let the hardware improvements come and adjust the block size _after_ we know we can handle these sizes. This proposal, where miners can vote for 10kB increases every 2 weeks, is just conservative enough to give us time to do that. Segwit2x, for me, is the most "big blockerish" proposal I would accept to avoid a chain split - nothing more.

[iluvpie60]

..study built upon faulty premise.

what premise is that?

That Steam gamers and their machines are equivalent to Bitcoin node operators and their machines.

aka ...I coulda had a V8, but all I got is this lousy rice burner - but at least it's got a fart pipe

I don't understand how people don't have a good or decent gaming computer if they are playing Steam games. You almost have to go out of your way to not get something that has at least 5 or 6 GB ram these days. Most built in Intel HD Graphics cards(built into processor) are pretty decent or better compared to dedicated graphics cards from 6 or 7 years ago too, or just buy a dedicated one for like 50 bucks if it is really necessary for the blockchain.

I built my computer and got everything I needed for less than $1,000 like 4 years ago or so, and it has a water cooled processor too.

32GB ram
I7-3770kK @ 3.5 GHZ4 physical cores (8 hyper threaded in total)
Radeon HD6870 Sapphire
750W power supply
Large case
Larger motherboard(for adding in upgrades later)
4TB External Hard Drive for the blockchain and other files

On Windows Experience Index I have the following scores.

Processor 7.7 out of 7.9
Memory(Ram) 7.7 out of 7.9
Graphics 7.8 out of 7.9
Gaming Graphics 7.8 out of 7.9
Primary Hard Disk 7.3 out of 7.9 (this is an older SSD that is still really fast too).

If people are struggling to have what I have 4 or 5 years later, I really don't know what to say to them I guess. You can almost build what I currently have now for half the price or less, and what I have is overkill by a lot for running a node.

So in reality, it would only cost a few hundred dollars to build a computer(buying a case/power supply/processor/ram/motherboard/HDD) that can handle current blocks and have enough power for larger blocks later.

What are these guys playing their Steam games with lol.

[pereira4]

..study built upon faulty premise.

what premise is that?

It looks like there's no way to convince big blockers of their mistake. What can be done at this point? Will they actually hardfork and find out for themselves?

Looks like no amount of arguments will ever change their mind. They want to live in 2010 when bitcoin was cheap as fuck because no one used it. They have lost their mind, they don't get it. We are witnessing something ridiculous unfolding. If the JarzikCoin code actually forks backed by the mining cartel it's time to get some popcorn.

[franky1]

sounds like some people are stuck in 1999 trying to scream at twitch/skype that their business model of live streaming just wont work

anyway onto other points

the problem is the initial block download/verification (first sync) and the propagation and verification of new blocks.

the synchronisation is not a problem. the actual headache is that cores UI is not set up to function until synchronisation is complete.
as thats the real headache people gripe and moan about in regards to syncing
EG
no user balance or being able to spend until node has downloaded certain blocks that contain the users UTXO.
this can be fixed by having the UI function in Lite mode where it grabs a UTXO set first. or even just requests UTXO from peers of the keys the user owns. and then synchronisation becomes a background issue that can be done as and whenever the user pleases.

thus making a node spendably functional from the first hour

again the mindset is not about hinder bitcoin growth for the 99% purely for the 1% with limitations,
if it was the case, then core should not have removed fee controls that have wiped out utility for about 33% of the world*

(analogy time of dumb mindsets)
whats next. scream that employment wont work because most jobs are more than a mile from people houses and not everyone has decent transport
whats next. scream that retail shops wont work because most shops are more than a mile from people houses and not everyone has decent transport

*i do love the snobbery
"bitcoin should not grow because americans might need to pay 3 hours of minimum wage labour more PER MONTH to use it
but dont worry about tx fee, just pay more, it doesnt matter if cuba, india, africa have to pay 40 hours of labour PER TX!"

[Lauda]

I don't understand how people don't have a good or decent gaming computer if they are playing Steam games.

The data that I've posted is based off of Steam hardware in 2015.

So in reality, it would only cost a few hundred dollars to build a computer(buying a case/power supply/processor/ram/motherboard/HDD) that can handle current blocks and have enough power for larger blocks later.

Which is absolute nonsense. I have no idea why you are trying to push your, clearly uneducated, opinion on the matter? Do you really expect me to run a node, on the same machine that I'd play games on? Do you want the entry for nodes to be >$1000 worth of hardware? If you do, then you may as well start building Paypal 2.0.

[franky1]

If you do, then you may as well start building Paypal 2.0.

LN will become paypal2.0
go check out the idea of hubs.. then check out the meaning of multisig and then check out CLTV (3-5business day maturing after confirm) CSV revoke (chargeback)

lauda please research the finer details. stop relying on reddit scripts