Topic: I don't see why big blocks are a problem, even 10 MB blocks right now aren't. - page 6. (Read 3773 times)

the real reason people dont run full nodes is not about restrictions of technology.. its about their crap GUI (user interface) and the social use of not needing to have something running 24/7 if they only personally use it once a day.

its about UTILITY that makes being a full node not attractive.
if bitcoin gained utility they would use it more often

current tech limitations is not the problem. its the fact that core dont seem to care about utility or care about users. all they care about is leadership and dominance

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.)


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.

They AREN'T a problem. That's just Blockstream/Core LYING to your face because they have an agenda to turn bitcoin into a profitable settlement layer! Boycott any implementation from Core and support bigger blocks by supporting Bitcoin ABC https://www.bitcoinabc.org/

So is everyone here running a node then? Shouldn't we be? I am not saying anything about centralizing anything... I just showed some quick math that shows any average person in any western country can run a node and there is basically no cost. If you already have a gaming rig you can run a node for awhile and there are tens of millions of gaming rigs. If there is only 5,000 nodes right now doesn't some blame belong on all of us for not promoting node usage more?

I am questioning why we all aren't running nodes and showing the math is there that many people can afford it, even if it was 10MB or 20 MB or 30MB right now and was always full, we can still afford it. I am not saying it is the best way to go.

10mb per ~10 minutes

lol
i remember 1999 where two mp3 songs (10mb) took 10 minutes... now its seconds..

does anyone see MILLIONS of people crying that Call of Duty cannot function bcause online gameplay just cannot work due to hardware/internet
does anyone see MILLIONS of people crying that skype ....
does anyone see MILLIONS of people crying that youtube Livestream ....
does anyone see MILLIONS of people crying that twitch ....

here we go again..
nonsense propoganda from the bscartel* that the only future is hubs..

bscartel want LN hubs and cry that blocksize increases = centralised upstream nodes (hubs). in their mind all they can see is centralisation
all because they want people to believe that everyone wants to run bitcoin on a raspberryPi 1a board(about a decade out of date nearly) and a memorystick


boring fools like that are stuck in the past. the internet has got faster since 2009, hard drives have got bigger and the average ram on basic home computers is more then the past.

real funny part is data for data of a fully validating full archival usernode is the same.
4200 tx's is still ~2mb whether using segwit or legacyx2

what will actually kill the network is the pruning and -nowitness options that then kill off the amount of full nodes (analogy: torrent seeders) and just leaves a cludge of a network of SPV/prunned/no-witness nodes (analogy: torrent leachers)
meaning what Gmaxwell called downstream nodes and what luke JR calls stripped nodes.. become the lower tier (cludgy leacher nodes)

meaning what Gmaxwell called upstream filters and what luke JR calls bridging nodes.. become the higher tier (full seeder nodes) hubs

maybe its time those that adore gmax and core actually do the research, yes some of you i have actually told to do this research atleast a year ago, but yet you still cant get your head passed the buzzwords, gmax and co have spoonfed.. and that leaves you unable to understand it.

segwit causes more breakdown of the full node count.
wake up

*bscartel(barry silbert, blockstream, core, btcc,litecoin, and DCG portfolio, (all loving segwit))

---

as for the mempool bloat graphs.. all the spam spikes relate to times the bscartel want their new bips activated by drumming up how change needs to be made ASAP
check out june 2016, october/november 2016 and then more recently when the other segwit activating bips popped up..

its all just drama to try getting segwit activated

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.

Decent analysis, until they get to *guessing* what hardware nodes are running on. "Characteristics of node hardware are based on a survey performed by Steam." They might as well estimate that everyone is using Raspberry Pi's.

Probably the whole reason the greed of the miners. The bigger the block the easier, faster and cheaper to conduct transactions. The miners do not want to lose the money that they have invested in expensive equipment. The unit will continue to grow as if they did not resist or there will be another currency that will give the best conditions and people will begin to get rid of bitcoins, and no one is interested.

Yeah, if you have blocks that huge it can add up really quickly. Most people won't have the resources to operate such a beefy node either.

Thats a really great chart there by bitfury.

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?

https://bitcointalksearch.org/topic/segregated-witness-levolution-prochaine-du-reseau-bitcoin-1336481
https://bitcointalksearch.org/topic/m.13664554



( backup image : http://imagizer.imageshack.us/a/img910/6880/x4YPj9.jpg )

P2P network, it's node ... it's people ... it's a fraction of 500GB hard drive (300 USD laptop).

Parallel?  Multi threading?

Bandwidth?

Downloading is fuck all for most people.

Uploading is where it gets sticky. And then there's general turnover too. I've seen people mention their node getting through over 1 TB a month. Quite a few internet plans would shut you down long before it reaches that.

If we assume 10mb blocks would be full, which is an awful long way away, that's over 10tb a month.

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.


I dont see why we need to let anyone even send 1 or 10 or 100 satoshis. Do that on ETH or XRP would be better with how their tech works. I  probably in favor of banning any transaction under 100satoshi as it is not worth almost anything currently. In the future i can see how it might need to be changed and agree changes need to happen.

Or maybe i am not taking everything into account for thay either. Is thsir real value in doing 1sat transactions beyond spam attacks?

Oh I agree hardforks cause chaos and are risky.

My point was more I dont see how segwit makes sense to do. Offchain stuff is more centralized and vunerable vs onchain.

I am aware of the above also, the graph and spam attacks.

Though censorship is not ideal, why not include in the code to disable sending of 1 satoshi transactions. Maybe in 20 years if there is a need to send a 1 sat transaction it can be solved a different way. Btc would need to be worth a lot for 1 sat transaction to be important in any way.

I think people will just use ETH for stuff like that.

I dont want segwit.

I think the fear of hardforks to increase blocksize correct. But only if the code is a secret. If the code is open to all for months i dont see why raising it would be a problem in the future.

Yes mempool isnt full any more I  am also aware of that. Just the things people said about blocksize being too big arent justified. Having a 20MB block right now still wont see it filled. And people could still run their nodes and what not.

There is a third factor besides storage and bandwidth: Validation of blocks and transactions costs a certain amount of CPU power and RAM. Large transactions particularly are resource-hungry.

10 MB blocks may not be that much of a problem - 8MB is what we would get after Segwit2x, in the worst case. The fear of the "small blocker" fraction, however, is that with a solution like Bitcoin Unlimited, in a few couple of years, we'll have 20+ MB blocks, because in view of the inexistent fee market it becomes viable to send microtransactions and gambling "spam" in the Satoshi Dice tradition or even to store arbitrary data on the blockchain.

Nice copy-pasted answer. At least big blocktard sockpuppets should put some more effort into varying their answers.

Anyway, we don't need big blocks because the mempool is not full, except when it gets attacked by Roger Ver and other rich attackers:



It's objectively stupid to risk a hardfork that only contains a 2MB increase when we don't need to. If we are hardforking you should look into other interesting technology and not the dumb blocksize increase only:

https://bitcoinhardforkresearch.github.io/

We also do NOT want to hardfork with unsafe code, developed in an unsafe amount of time, risking a 40 billion market in the process.

We do not need bigger blocks now, it's objectively a mistake:


https://www.youtube.com/watch?time_continue=2941&v=iFJ2MZ3KciQ

It is not a stupid question. Don't worry many of us have worked this out, similar to yours above and equally don't see the fuss.

There is an irrational fear that the number of full nodes will shrink due to the costs above and Bitcoin will be centralised.

I can download a 200mb mp4 file in 1 minute 30 seconds. 10mb in 20 second. (both averages) Internet is £30 per month and already paid for due to other uses. Running a full node does not add any extra cost at all.

Ok. I want to ask a stupid question here. As someone who has been really out of the loop on the technical side of things and I also have really no idea of the politics behind these factions... but I want to ask what is the problem with big blocks exactly.

In terms of space on harddrives I don't see it. A 4 TB HDD is like 70 dollars on sale.

I did some quick math.

Assuming a block created every 10 minutes and is the max size(which is extremely unlikely but lets go with that) 1MB or 2MB or 3MB or 4 MB or 10 MB.

10 years from now here is how much more GB is added to the total size of the blockchain


1mb blocks - 525.6 GB
2mb blocks - 1,051.2GB
3mb blocks - 1,576.8GB
4mb blocks - 2,102.4GB
10mb blocks - 5,256GB

Even with 10 MB blocks starting now and goign to 10 years out there would not be a storage issue, for the average person. The average person(western) can afford to buy a 6 TB or a little higher HDD for like 150 bucks or something, and by the time we are 10 years out a 30 TB drive will be like 100 bucks or cheaper.

Cost of internet? On average most internet companies per month charge like 60 to 100 a month and you get between 250 GB to 1,000 GB a month downloaded before they charge you an extra 10 bucks for another 100 GB.

So even under the 10MB block size on a monthly basis you are only downloading on your node 40 GB per 4 week period. (I note that this number may be higher as others point out).

How about speed of that internet? Of course it can handle it... 10MB = 80 mb(mega bits) and  in terms of speed, so to download 10MB on an average of every 10 minutes your internet speed needs to be at the minimum be able to handle 133 kbps, yes, kilobits per second, so convert to KiloBytes per second by dividing by 8 and you get 16.5 KBPS, DIAL UP IS 56 KBPS.

If you have 80mbps download you can do 600  10MB blocks in 10 minutes.


I honestly have no idea why anyone thinks big blocks are a problem, maybe if we got to 100 MB blocks and it happened in a few years, but that means almost everyone in the world would be using BTC to make 100MBblocks full every 10 minutes every day every week every year...


So what is the issue? Please lay it out for me, I have been away for awhile.  Am I missing something? Maybe poorer people cant afford to host a node is the issue? But are they hosting it now anyways, probably not...

Please note I did edit some things as I realized I did not word some things correctly and some things were exclamatory for no reason. Thank you below for the great responses about the spam attacks, did not consider that. I guess that changes my mind then on what the real solution is, this may have made me more confused than before I posted to be honest.