The "test".. results, opinions

[spazzdla]

Sending BTC for a small fee currently takes hours.


My issue with the 8MB blocksize however is with cheap fees what stops someone from just spamming the shit out of it at 8mb.. 16.. 32.. etc..  Which would cause huge problems for nodes..  at 32mb the blockchain would explode in size if it was being spammed constantly.  Even 8 would bloat it pretty quick.  Thus not really solving the problem at all...


I think this proves just jumping to 20mB blocks could be quite damaging as a 200gig Blockchain would decimate decentralization.

Anyone have any other ideas on fixing this issue? I currently do not... :S..

Perhaps the answer is keep it at 1mB blocks.. and it costs a decent amount to send BTC as it is the most secure network on the planet... so it would be used to move large amounts of wealth and store large amounts of wealth.  Something like bitshares or altcoins would be spending currency..

[IIOII]

Sending BTC for a small fee currently takes hours.


My issue with the 8MB blocksize however is with cheap fees what stops someone from just spamming the shit out of it at 8mb.. 16.. 32.. etc..  Which would cause huge problems for nodes..  at 32mb the blockchain would explode in size if it was being spammed constantly.  Even 8 would bloat it pretty quick.  Thus not really solving the problem at all...


I think this proves just jumping to 20mB blocks could be quite damaging as a 200gig Blockchain would decimate decentralization.

Anyone have any other ideas on fixing this issue? I currently do not... :S..

Perhaps the answer is keep it at 1mB blocks.. and it costs a decent amount to send BTC as it is the most secure network on the planet... so it would be used to move large amounts of wealth and store large amounts of wealth.  Something like bitshares or altcoins would be spending currency..

I think the current problems are less dramatic than you think. You have to pay a bit higher fees, yes, but it does in no way hinder regular transactions.

You point out correctly that increasing the maximum blocksize is a risky decision. If decentralization is hurt, the entire Bitcoin system is at stake. In my opinion, higher fees are the only solution to spam attacks. We need a healthy fee market. Bitcoin itself doesn't scale well enough to include all microtransactions.

Solutions are already in the works. Sidechains will offer a faster, cheaper, and sufficiently reliable ways to transact in <$1 values. Small-value transactions do not need the same level of security as higher value ones anyway.

So there will be solutions on top of Bitcoin that address the problem. However even with a very conservative approach to blocksize-increases, I don't think that it will ever be a problem to send >$5 transactions without using such solutions for a reasonable fee.

[eyeknock]

Block size increase is obviously required. One of the appeals of bitcoin is fast transactions and low transaction fees, so it wouldn't make sense to have to pay higher just to get transactions confirmed faster.

exactly, i really think that yes or yes we will need to increase the block size, everybody know it the question is why they didnt decide a good solution about it? is not good about why they disagree between them.

about pay more fee to have faster transactions havent sense for me too, what we need is to made things going faster without paying more than we are doing now, this will be the real solution, bitcoin is not a bank who charges you for everything, we should never forget it.

[IIOII]

Block size increase is obviously required. One of the appeals of bitcoin is fast transactions and low transaction fees, so it wouldn't make sense to have to pay higher just to get transactions confirmed faster.

Well then Bitcoin doesn't make sense for you, because transactions are prioritized exactly this way. Bitcoin offers extremely fast transactions and extremely low transactions fees - just not for cent-value transactions.

The maximum blocksize will have to increase a bit to make sidechains possible. But it makes absolutely no sense to increase it just to include any kind of transaction dust. People will be able to cheaply throw Satoshis at each other, they just will be doing it on a secondary layer. Everybody will be satisfied and the security of Bitcoin will not be compromised.

[Argwai96]

Block size increase is obviously required. One of the appeals of bitcoin is fast transactions and low transaction fees, so it wouldn't make sense to have to pay higher just to get transactions confirmed faster.

Correct you dont want new comers to actually experience this type of event because they could feel a bit overwhelm yes cost will still be low but no everyone would be able to to have a full bitcoin experience is their transactions are taking 3 to 8 hours to confirm, an obvious bigger blog would needed in order to overcome possible transaction traffic and a normal bitcoin experience.

I actually had a few people ask me if bitcoin was broken because their transactions where pending for hours.

[spazzdla]

Block size increase is obviously required. One of the appeals of bitcoin is fast transactions and low transaction fees, so it wouldn't make sense to have to pay higher just to get transactions confirmed faster.

Correct you dont want new comers to actually experience this type of event because they could feel a bit overwhelm yes cost will still be low but no everyone would be able to to have a full bitcoin experience is their transactions are taking 3 to 8 hours to confirm, an obvious bigger blog would needed in order to overcome possible transaction traffic and a normal bitcoin experience.

I actually had a few people ask me if bitcoin was broken because their transactions where pending for hours.

If the blockchain jumps to 1TB by 2018, Bitcoin as we know it dies and JPmorgancoin will be born from the ashes.

No thx.

[achow101]

What happens when someone decides to spam the network with large fees? What if someone figures out a fee that is large enough to be considered high priority and has enough money just to keep spamming the network? Does that mean that in order to get my transaction confirmed I have to pay an even higher fee? That could be an attack on the network where someone could either drive up fees, prevent anyone else's transactions through their high fee spam, or make it too cost prohibitive to use Bitcoin at all.

[franky1]

here we go again..

bloating is no problem.. hard drives are cheap.. we dont live in the 90's where a 3gb hard drive was $200.... these days we can get a 8TB for the same price.

also the 1mb has been around for a while now. but we havnt seen a 52gb yearly growth EVER.. so dont expect a 1tb yearly growth instantly as soon as a 20mb block rule is implemented..

[spazzdla]

What happens when someone decides to spam the network with large fees? What if someone figures out a fee that is large enough to be considered high priority and has enough money just to keep spamming the network? Does that mean that in order to get my transaction confirmed I have to pay an even higher fee? That could be an attack on the network where someone could either drive up fees, prevent anyone else's transactions through their high fee spam, or make it too cost prohibitive to use Bitcoin at all.

If it costs you 10k every 10min to spam the network.. have fun bro lets see how long you can burn money.

[achow101]

What happens when someone decides to spam the network with large fees? What if someone figures out a fee that is large enough to be considered high priority and has enough money just to keep spamming the network? Does that mean that in order to get my transaction confirmed I have to pay an even higher fee? That could be an attack on the network where someone could either drive up fees, prevent anyone else's transactions through their high fee spam, or make it too cost prohibitive to use Bitcoin at all.

If it costs you 10k every 10min to spam the network.. have fun bro lets see how long you can burn money.

If you are a government, you have tons of money at your disposal. They can just keep burning money. Even better, if a miner is doing it, he can just keep recycling the same Bitcoin and do it over again.

[tspacepilot]

I'm trying to figure out if there was some kind of third test yesterday.  I saw a rather unusually large backlog of unconfirmed transactions and I wondered if it was happening again.  I never saw anything definitive though.

With respect to the test results and whatnot, I suppose that someone burnt a lot of money to prove that a lot of money can cause problems and back up the network.  If we ever actually have that kind of volume of legitimate transactions then yes, I think we would want to make a modest update on the block-size cap.  If it's just to accomodate volume tests, that's not important in the long run.  Although, again, in the long run, I suppose these tests did provide some data which can be consulted in the future at decision times---so at least that part of it is a positive.

[spazzdla]

What happens when someone decides to spam the network with large fees? What if someone figures out a fee that is large enough to be considered high priority and has enough money just to keep spamming the network? Does that mean that in order to get my transaction confirmed I have to pay an even higher fee? That could be an attack on the network where someone could either drive up fees, prevent anyone else's transactions through their high fee spam, or make it too cost prohibitive to use Bitcoin at all.

If it costs you 10k every 10min to spam the network.. have fun bro lets see how long you can burn money.

If you are a government, you have tons of money at your disposal. They can just keep burning money. Even better, if a miner is doing it, he can just keep recycling the same Bitcoin and do it over again.

Your populs is going to get pretty pissed you're burning money for no reason.

Not true unless they have 51% of the mining network and at that point there are other issues.  Other miners will take those BTC from them via fees.

[achow101]

What happens when someone decides to spam the network with large fees? What if someone figures out a fee that is large enough to be considered high priority and has enough money just to keep spamming the network? Does that mean that in order to get my transaction confirmed I have to pay an even higher fee? That could be an attack on the network where someone could either drive up fees, prevent anyone else's transactions through their high fee spam, or make it too cost prohibitive to use Bitcoin at all.

If it costs you 10k every 10min to spam the network.. have fun bro lets see how long you can burn money.

If you are a government, you have tons of money at your disposal. They can just keep burning money. Even better, if a miner is doing it, he can just keep recycling the same Bitcoin and do it over again.

Your populs is going to get pretty pissed you're burning money for no reason.

Not true unless they have 51% of the mining network and at that point there are other issues.  Other miners will take those BTC from them via fees.

Not going to get into an argument about this. I'm just saying that it is not infeasible for someone with a ton of money to attempt to do this to the network.

[dothebeats]

What happens when someone decides to spam the network with large fees? What if someone figures out a fee that is large enough to be considered high priority and has enough money just to keep spamming the network? Does that mean that in order to get my transaction confirmed I have to pay an even higher fee? That could be an attack on the network where someone could either drive up fees, prevent anyone else's transactions through their high fee spam, or make it too cost prohibitive to use Bitcoin at all.

Good luck burning your money continuously. Even the world's richest people wouldn't do such things to prove a point on fees. But theoretically, yes, if you want to get your tx faster to confirm, you need to pay an even higher fee for it to be placed on higher priority, but I don't think that would happen anytime sooner.

[Amitabh S]

Is there an option to "withdraw" an unconfirmed transaction other than by trying to double-spend it?

[Meuh6879]

Anyone have any other ideas on fixing this issue? I currently do not... :S..

Self calculation for fees (file follow the activity of network to calculate more precisely, the fees in real time block generation).
Already activate on 0.10.2 ... but modified in the 0.11.0 RC3 (after the 1st test of spamming network)

And so, i think that it will be, again, modified in the next release after the result of the day-by-day spamming test network.

[LiteCoinGuy]

In mid-2010, when the average block size was about 1KB, Satoshi put in the 1MB block size limit because the project he had worked on for years was finally gaining real traction. People were using it!

Everyone had to run a full node in order to have a wallet and send bitcoins, so this led to risk. The risk from someone mining a series of large blocks: say, 5, 10 or 20MB. BTC was very cheap then, and miners had no money, compared to the huge amounts today invested in hardware for which they want ROI, investments large enough for them to want Bitcoin to succeed.

FPGA mining was just being discussed in mid-2010. What if a rogue miner got a head-start and kept hammering out spam-blocks killing off the fledgling interest in Bitcoin? Of course, he did not make the limit 2KB. That would quickly cripple usage, and blow out confirmation times! So, Satoshi capped the size at 1MB,

Did the blocks immediately fill to the 1MB level? No. And they won't after raising the limit a mere 800%, or even doubling that every two years.

Satoshi knew the 1MB was a risk to his long-term vision for a global competitor to the likes of VISA, but he did massive code changes, and frequently. He just assumed it would be easily increased or removed when the time was right.

The time is right.

[ajas]

The time is right

If we look at the temporal evolution of the blocksize, it was currently filled up
only half with a very slow exponentional increase. Extrapolation would result
in a hit of the 1 MB threshold in about 1 year from now on.

https://blockchain.info/en/charts/avg-block-size?showDataPoints=false&timespan=all&show_header=true&daysAverageString=1&scale=1&address=

So to my oppinion there is no urgent need for a large increase of the blocksize
which may have other disadvantages. Furthermore I do not see a need for forking the blockchain
and/or the bitcoin core code. So I think there must be other intentions for trying to do this.

Currently somebody is spamming the blockchain, obviously with the purpose to give arguments
for this fork. I am not sure what should be the best reaction on that.

-------------------------------

By the way does anybody have explanatios for the changes in temporal evolution in the graph
linked above ?:

 - strong, more than exponential groth until mid 2011
 - constant level afterwards
 - jump in mid 2012
 
A