Bitcoin Forum
November 09, 2024, 02:24:34 PM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 [2] 3 4 »  All
  Print  
Author Topic: Don't we need to increase block weight/size?  (Read 721 times)
Maveth13
Full Member
***
Offline Offline

Activity: 336
Merit: 112


View Profile
January 15, 2018, 02:07:48 PM
 #21

Increasing the block size would solve our current situation, but how would we get most of the network to accept the change? In line with that comes many cons. Main reason is that it would increase hardware and bandwidth demands, thereby increasing cost to operate a node.

There's a somewhat 'controversial(?)' reason as to why core has not yet increased the blocksize. I think it's still being censored here, so if anyone wants the link to the reddit discussion send me a pm.
Carlton Banks
Legendary
*
Offline Offline

Activity: 3430
Merit: 3080



View Profile
January 15, 2018, 02:12:45 PM
 #22

please let me know what
specific point you would like to disagree on

Everyone's a bank in Bitcoin
Everyone's a bank on the Lightning network

and also note that I am not against
Segwit which kinds of brings file compression to the blocks using "Extended blocks" to
store data in BC because short term it is a good fix

That's totally wrong

Vires in numeris
RGBKey
Hero Member
*****
Offline Offline

Activity: 854
Merit: 658


rgbkey.github.io/pgp.txt


View Profile WWW
January 15, 2018, 02:47:01 PM
 #23

agreeing with everyone who says "blocks aren't full yet", or "Bitcoin needs validation performance increases first so the network can handle it"

Validation performance didn't even enter my mind, that's a huge point. You wouldn't just need more storage space, doing things like running a full node on an rpi3 would become more difficult.
Anti-Cen
Member
**
Offline Offline

Activity: 210
Merit: 26

High fees = low BTC price


View Profile
January 15, 2018, 03:40:14 PM
 #24

Everyone's a bank in Bitcoin
Yes if they hold the public key

Quote
Everyone's a bank on the Lightning network
No because not everyone becomes a hub and keeps lots of channels open because it costs
$30 in Tx fees to close them and get the balance committed to the BC

lets draw this out and Bob sends Alice $1.00

Bob---->You---->Hub1<><><><>Hub2<------Alice

You become a relay node that lets Bob share your major hub because you use Bitcoin
five times a day, always online and share the channel.

Hub1 settling up with Hub2 cost him money (in debt to hub2) so he wants you to chip in with the bill
and takes fees from you for the service but Hub1/2 settling does not mean that Bob is forced to settle
up with Hub1 or you or even Bob

Question
1. Hub1 and you fall out, fees to high but hub1 owes you $500 so explain settlement of the public and private ledgers
2. Bob owes you $1 but the channel is no use because you have no out going channels so what happens to Bob

and also note that I am not against
Segwit which kinds of brings file compression to the blocks using "Extended blocks" to
store data in BC because short term it is a good fix

Quote
That's totally wrong

Yeah, maybe you should take your issue up with this guy
https://www.youtube.com/watch?v=OFfBRzh9HmU
because i think people will see that my comment was totally right


Mining is CPU-wars and Intel, AMD like it nearly as much as big oil likes miners wasting electricity. Is this what mankind has come too.
synthlock
Newbie
*
Offline Offline

Activity: 10
Merit: 0


View Profile
January 15, 2018, 03:41:27 PM
 #25

agreeing with everyone who says "blocks aren't full yet", or "Bitcoin needs validation performance increases first so the network can handle it"

Validation performance didn't even enter my mind, that's a huge point. You wouldn't just need more storage space, doing things like running a full node on an rpi3 would become more difficult.

Double SHA256 is not a huge bottleneck in terms of validation; if anything, it would just be the merkle tree size. Running full nodes on a Raspberry Pis is a bad idea in general. Low network performance and moderate to OK CPU performance.
Anti-Cen
Member
**
Offline Offline

Activity: 210
Merit: 26

High fees = low BTC price


View Profile
January 15, 2018, 03:59:53 PM
 #26

Raspberry Pi

Amazing little devices for the price so are you talking about the block files being on the network (LAN)
or coming from a USB disk drive attached to the device ?

200gb from RJ45 would be slow and even USB3 D-Drive (thinks they do USB3) would be slow
so are you talking about plunging a SETA drive into a Raspberry here ?

Reading back down the BC for a current transaction to check all part coins history making up
the wallet without a major large index in memory would involve reading near 200gb of block files
so hats off to anyone that gets a full node working on one of these $40 devices.

Mining is CPU-wars and Intel, AMD like it nearly as much as big oil likes miners wasting electricity. Is this what mankind has come too.
Carlton Banks
Legendary
*
Offline Offline

Activity: 3430
Merit: 3080



View Profile
January 15, 2018, 06:11:32 PM
 #27

Everyone's a bank on the Lightning network
No because not everyone becomes a hub and keeps lots of channels open because it costs
$30 in Tx fees to close them and get the balance committed to the BC

That's totally wrong.

There is no $30 fee (it's a bid-based supply/demand fee, not a fixed number). And there's no necessity to close channels often.

And there's every incentive to open plenty of channels. And free fees will undermine any attempt to turn Lightning routing into a business.


and also note that I am not against
Segwit which kinds of brings file compression to the blocks using "Extended blocks" to
store data in BC because short term it is a good fix

Quote
That's totally wrong

Yeah, maybe you should take your issue up with this guy
https://www.youtube.com/watch?v=OFfBRzh9HmU
because i think people will see that my comment was totally right

No, it's totally wrong. I'm talking to you, not some youtube video. If there's someone on that video who knows what they're talking about, they're definitely not saying anything resembling what you said.

Your ability to communicate facts about Bitcoin is completely inverted in relation to your confidence: you think you're perfect, but you sound like you're talking in your sleep after getting drunk the night before

Vires in numeris
Virtual miner
Sr. Member
****
Offline Offline

Activity: 728
Merit: 275



View Profile
January 15, 2018, 06:46:11 PM
 #28

Sure, it's bad that fees are high, but high fees are not going to kill Bitcoin, because they are driven by open market, they are not installed by one side only like it happens in some centralized systems.
But centralization can kill Bitcoin, you can't undo the damage of big blocks if the blockchain became bloated past certain point.
Of courde high fees can't kill bitcoin, it even didn't affect it overally but don't you hear words from people who says so high fees make it useless to make transaction? Bitcoin is decentralyzed for me, safest option and great currency. These high fees does shit job, I can't receive transaction quickly without ultra high fees. And yes, some people prefer high fees and spammed network because they use it for theirself. Block size must be increased and I think currently even 4MB will be absolutely enough.
Can the block size be increased using merely a soft fork? Because hardforks these days are just becoming useless and as they create an alter chain most of the people are reluctant to move on. Increasing block size will surely cater most of the needs of our present time and 4MB would cater need of a pretty long time.
RGBKey
Hero Member
*****
Offline Offline

Activity: 854
Merit: 658


rgbkey.github.io/pgp.txt


View Profile WWW
January 15, 2018, 08:47:43 PM
 #29

Sure, it's bad that fees are high, but high fees are not going to kill Bitcoin, because they are driven by open market, they are not installed by one side only like it happens in some centralized systems.
But centralization can kill Bitcoin, you can't undo the damage of big blocks if the blockchain became bloated past certain point.
Of courde high fees can't kill bitcoin, it even didn't affect it overally but don't you hear words from people who says so high fees make it useless to make transaction? Bitcoin is decentralyzed for me, safest option and great currency. These high fees does shit job, I can't receive transaction quickly without ultra high fees. And yes, some people prefer high fees and spammed network because they use it for theirself. Block size must be increased and I think currently even 4MB will be absolutely enough.
Can the block size be increased using merely a soft fork? Because hardforks these days are just becoming useless and as they create an alter chain most of the people are reluctant to move on. Increasing block size will surely cater most of the needs of our present time and 4MB would cater need of a pretty long time.

IIRC the block size cannot be increased without a hard fork. The reason segwit effectively increased the block size was because it moved some data outside of the transactions, but didn't change the block size limit. I could also be very wrong.
Anti-Cen
Member
**
Offline Offline

Activity: 210
Merit: 26

High fees = low BTC price


View Profile
January 15, 2018, 08:50:32 PM
 #30

There is no $30 fee (it's a bid-based supply/demand fee, not a fixed number). And there's no necessity to close channels often.
oh yes so next time I will use the number $30-55 to ensure I cover today fees up to the max we have seen

Quote
And there's every incentive to open plenty of channels.
But if you owe money then you end up paying $30 tx fees to close the channel on top of fees to hub/bank
as settlement is done on BC so good incentive to people with money coming in but not going out. T&C for all I know
between client (me/You) and bank could prevent you passing the $30-55 fee on to the hub/bank and needs a deposit to
cover this situation.

Quote
And free fees will undermine any attempt to turn Lightning routing into a business.
I am not sure quite what you are saying here, competition between banks/hubs is what I think your
getting at so please clarify

Quote
No, it's totally wrong. I'm talking to you, not some youtube video. If there's someone on that video who knows what they're talking about, they're definitely not saying anything resembling what you said.

No you are wrong and from your comment it is clear you did not watch it so save the speculation and lets deal in facts please.
Quote
Your ability to communicate facts about Bitcoin is completely inverted in relation to your confidence: you think you're perfect, but you sound like you're talking in your sleep after getting drunk the night before

Sorry but insults from you just pushes the point home that your line of thinking does not have any legs to stand on.


Mining is CPU-wars and Intel, AMD like it nearly as much as big oil likes miners wasting electricity. Is this what mankind has come too.
Anti-Cen
Member
**
Offline Offline

Activity: 210
Merit: 26

High fees = low BTC price


View Profile
January 15, 2018, 09:03:30 PM
 #31

The reason segwit effectively increased the block size was because it moved some data outside of the transactions, but didn't change the block size limit. I could also be very wrong.

I think you are right and it sort of compresses the block but not by using Gzip but moving some of the data
to what they call an "extended block" so maybe the files will look more like

BLK000001.dat
BET000001.dat

Here I am guessing but from a coding perspective it will make it more complicated having to read/parse two files
and will break existing code so I suspect I am missing something key here.

Mining is CPU-wars and Intel, AMD like it nearly as much as big oil likes miners wasting electricity. Is this what mankind has come too.
Carlton Banks
Legendary
*
Offline Offline

Activity: 3430
Merit: 3080



View Profile
January 15, 2018, 09:16:47 PM
 #32

Quote
And free fees will undermine any attempt to turn Lightning routing into a business.
I am not sure quite what you are saying here, competition between banks/hubs is what I think your
getting at so please clarify

Who's going to make money out of a p2p network if there are nodes on it charging zero routing fees? You can charge as high a price as you like, nodes will simply route around your node.

There's no incentive to charge high fees (or any fees) if it makes your node less well connected. And there's no cost to running a node if you have a home and a computer. So your talk of centralised hubs is complete nonsense, they'll be sat around routing nothing (and costing the same amount to run as the nodes who do route traffic)


Your ability to communicate facts about Bitcoin is completely inverted in relation to your confidence: you think you're perfect, but you sound like you're talking in your sleep after getting drunk the night before

Sorry but insults from you just pushes the point home that your line of thinking does not have any legs to stand on.

It's true, you're incoherent and rambling alot of the time

Vires in numeris
Ranker
Newbie
*
Offline Offline

Activity: 26
Merit: 0


View Profile
January 15, 2018, 09:18:27 PM
 #33

We are going to need changes to the protocol, eventually.

Perhaps the Lightning network is going to enough - I'm sure the devs are going to know best what to do.
Anti-Cen
Member
**
Offline Offline

Activity: 210
Merit: 26

High fees = low BTC price


View Profile
January 15, 2018, 09:35:08 PM
 #34

There's no incentive to charge high fees (or any fees) if it makes your node less well connected. And there's no cost to running a node if you have a home and a computer. So your talk of centralised hubs is complete nonsense, they'll be sat around routing nothing (and costing the same amount to run as the nodes who do route traffic)

How does anyone find a hub if we don't have coordinators which is semi-centralized not that I object
but they should admit this instead of pretending otherwise.

Routing is fine, like relays on Tor network and these nodes don't hold private ledgers and therefore do not
need to settle accounts. Yes no trouble but now we get to banks/hubs that hold private ledgers "Off-Block"
and charge fees like in the little diagram I did you.

Lets deal with settlement of hubs as channels close please instead of going around the houses

Quote
It's true, you're incoherent and rambling alot of the time

I deal in facts, logic and reason and I provide links in a lot of cases plus I earn a good living in software
so maybe I have trouble talking to people that are slot-machine bandits, have vested interests and cannot
get the point across using "Facts, logic and reason"

Mining is CPU-wars and Intel, AMD like it nearly as much as big oil likes miners wasting electricity. Is this what mankind has come too.
Anti-Cen
Member
**
Offline Offline

Activity: 210
Merit: 26

High fees = low BTC price


View Profile
January 15, 2018, 09:45:04 PM
 #35

Perhaps the Lightning network is going to enough - I'm sure the devs are going to know best what to do.

In the beginning, like five years or more ago I would agree with you but that's not the game being played
out today because money and greed has taken over so lets look at a case history of ETH !

Ticking along fine, Crypto-Kitties went basaltic using ETH "Smart Contracts" and the network became
unusable so did the miners put the fees up via "Gas" to reduce load to silly prices or do they all work as
a team to deal with the problem that seems to have been resolved in no time.

Contrast this with Bitcoin as the development teams tries to hide behind "Consensus" and pretending they
are skinning LN out as Rome burns.

  

Mining is CPU-wars and Intel, AMD like it nearly as much as big oil likes miners wasting electricity. Is this what mankind has come too.
Maveth13
Full Member
***
Offline Offline

Activity: 336
Merit: 112


View Profile
January 16, 2018, 12:53:38 AM
 #36

Even though i know the advantages of SegWit and LN, don't we need to increase block weight/size since i think current block weight/size won't be enough if more people adopt/use Bitcoin & some people might think LN isn't really useful since they still need to pay big fees to open/close LN channel?
I know raising maximum block size/weight is risking decentralization, but i think it won't be hurt to increase it not too much as long as there's data/consensus which says most nodes still can run even after raising maximum block size/weight?

What do you think? Also, CMIIW.

You "think" it would not hurt to increase it. Ok, I want achow or someone who has a deeper understanding of the block sizes and block propagation to make a comment on OP's "thinks it would not hurt to increase the block size".

OP, then why are the Core developers unwilling to do it? There has to be a good technical reason, correct?

Increase cost in running nodes, and would hinder decentralization, usual and still hypothetical reasons they give out.

The real reason lies on the "head maintainer" of the core github repository, Wladimir Van der Laan. He decides which software changes will go in the official core releases. Here's what he thinks about this issue:

Quote
Yes, [my position remains the same]. If we’ve learned anything from the 2008 subprime bubble crisis it should be that nothing ever keeps growing exponentially, and assuming so can be hazardous …. Changing the rules in a decentralized consensus system is a very difficult problem and I don’t think we’ll resolve it any time soon.
coder0x15
Jr. Member
*
Offline Offline

Activity: 36
Merit: 3


View Profile
January 16, 2018, 12:55:00 AM
 #37

The block size allready increase, after SegWit
Wind_FURY
Legendary
*
Offline Offline

Activity: 3094
Merit: 1930



View Profile
January 16, 2018, 04:37:29 AM
 #38

Do you deny these hub things take transactions of the block-chain that we were told was the best thing since sliced bread
Do you deny these hub things charge fees and keep a private ledger  
Do you deny these hub things use a degree of centralization

disagree all you like but I will call these "Hub things" banks because they are


Anyone can open a Lightning channel with anyone else, and anyone can also open as many channels as they like. But I guess you're going to call anyone with more than 1 channel a bank Roll Eyes



Carton, would it be fair to say that the Lightning Network would function like Ripple but with a few differences like, Lightning channels are backed by real Bitcoins while a Ripple "trust" channel is backed by nothing but an IOU?

I would like to add that OP looks like a Ripple supporter.

██████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
█████████████████████████
██████████████████████
.SHUFFLE.COM..███████████████████████
███████████████████████
███████████████████████
███████████████████████
███████████████████████
█████████████████████
████████████████████
██████████████████████
████████████████████
██████████████████████
███████████████████████
███████████████████████
███████████████████████
███████████████████████
███████████████████████
███████████████████████
██████████████████████
██████████████████████
██████████████████████
███████████████████████
███████████████████████
███████████████████████
███████████████████████
███████████████████████
███████████████████████
███████████████████████
.
...Next Generation Crypto Casino...
analyststins
Newbie
*
Offline Offline

Activity: 24
Merit: 1


View Profile
January 16, 2018, 05:45:58 AM
 #39

The battle that needs fought now (again, imo) is adoption of the lightning network. It's a drastic change from the benefits provided by segwit, which was more about fixing transaction malleability. If we could only fight one more battle, and it had to be raise the block size or get people to move over to the lightning network, the latter is going to work for a lot longer than the former. Because it's important which battles we choose here, because people will quickly get tired of needing to make changes to use Bitcoin.

Can't agree more with this. What we need to do is education and awareness on what exactly LN is. OP itself thinks Lightning Network increases fees. All the transactions occur off-chain and once the payment channel is closed you settle it in the main chain. It is similar to how you open a debt account in the local grocery store where lets say the paper that records your debt is stored in a box with two padlocks and each of you has the key.
Anti-Cen
Member
**
Offline Offline

Activity: 210
Merit: 26

High fees = low BTC price


View Profile
January 16, 2018, 03:48:16 PM
 #40

Carton, would it be fair to say that the Lightning Network would function like Ripple but with a few differences like, Lightning channels are backed by real Bitcoins while a Ripple "trust" channel is backed by nothing but an IOU?

Are you saying that BTC is backed by something else like gold and not just IOU's

Ripple is fantastic apart from it being owned by bankers but lets face it the BTC miners
and development team are taking us for dick-heads with the speed and fees so maybe
your being a little bias towards XRP with all due respect


Mining is CPU-wars and Intel, AMD like it nearly as much as big oil likes miners wasting electricity. Is this what mankind has come too.
Pages: « 1 [2] 3 4 »  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!