Because the block limit is not creating a fees market. The block reward is too high and masks the theoretical "bidding" process for block space, leaving aside the fact that wallet software does not offer users a way to update a fee on a transaction until it has been dropped from all the mempools.

However 20MB is large enough for a proper fees market to develop. Currently, blocks >900KB are carrying 0.25 BTC in fees. ~5BTC in fees for a full 20MB block and a block reward down to 6.25, means the fees and reward are similar, and the fees market should be healthy.

No one cares about this. Duplicate tx from MPcoin get discarded.


Forking without consensus is exactly what will happen if volumes suddenly ramp up filling the 1MB blocks and everyone is yelling about ridiculous delays, bad press, collapsing price, etc, and a large number of nodes feel it necessary to increase the limit, while others refuse to, welcoming the "necessary pain".

That is why Gavin has proposed re-using the block version majority process, which worked very well in 2013 for P2SH, and at least gets a big consensus in the mining power. A well-executed fork, done over many months, will be very smooth for all users.

LeMiner, fully agree.

There are not two networks. There are two sets of nodes which disagree on which is the "One True Blockchain". Up until the fork (a certain block height) all the nodes have 100% agreement, but then after the fork, a subset, "MPnodes", start to reference utxo from blocks which are unknown to the Bitcoin nodes.
These tx are rejected by Bitcoin nodes just the same way, as if I create and broadcast a tx for 123,000 BTC to send to an address I own, so I can have a nice fat bitcoin holding to rival the Winklevoss'.

Companies which accept bitcoins for products and services will be anxious to ensure that their nodes conform to the majority view and have the same blockchain that the exchanges, and payment processors do.

In that scenario:
If the utxo for x comes from an MPcoin block before the fork then the tx is discarded as a double-spend.
If the utxo for x comes from an MPcoin block after the fork then the tx is discarded due to unrecognized inputs.

The fork has no effect on anyone's existing bitcoin holding. It only affects newly mined bitcoins rewarded after the fork occurs.

edit: i.e. you don't want to mine on a fork which is not the majority fork, because the reward will be worth little or nothing.

D&T, a very informative OP.

Empirically, ~1MB blocks support 2.7 tps.

https://bitcointalk.org/index.php?topic=941331.msg10360199#msg10360199

FFS. I can't imagine the state of your car.


The owners of Bitcoin exchanges are not performing chimpanzees at a tea party. If a fork occurred and two chains were similar size then the CEOs of Bitcoin businesses would quickly agree (with Core dev and the mining pools) which chain to use (B) and which chain to ignore (C) , even if this meant suspending trading in the meantime.

This is why Gavin has consulted with many Bitcoin companies/players already, to find out what their view is. He reports that they support scalability by a big margin. Bitcoin businesses are going to want their business model to have the best chance of success, and that does not include crippling the tx throughput on the blockchain which will also cripple business models.

The rejected chain C will quickly find that the difficulty (40+ billion) is way too high for the devoted "anti" rump of miners clinging to it, and it will take several hours to mine each block. This will get worse as the coinbase rewards can't be sold for decent fiat at any significant exchange. Miners will drift to the majority chain. It will probably take 1 year for the difficulty to fall on chain C until it is usable. So chain C will need to be forked in order to slash the difficulty for the few users who still want the 1MB lmit.

I thought I would have a look at recent blocks >900KB. There are 21 in the last 6 days.

Block_height Block_time Size_KB Num_Tx Fee_BTC

341192   31/01/2015_7:10   976.54   2518   0.3303735

341226   31/01/2015_13:01   903.09   1106   0.15922165

341305   1/02/2015_1:05   959.89   1320   0.20086461

341313   1/02/2015_2:19   912.48   693   0.2979134

341350   1/02/2015_7:49   975.67   1735   0.263689

341505   2/02/2015_5:11   976.5   735   0.16244257

341517   2/02/2015_7:02   947.72   746   0.14308729

341519   2/02/2015_7:36   957.16   482   0.13242593

341520   2/02/2015_8:02   976.45   1964   0.28018485

341525   2/02/2015_8:35   973.77   844   0.187877

341737   3/02/2015_19:00   976.49   1291   0.18828618

341765   4/02/2015_1:23   976.49   2549   0.35075577

341773   4/02/2015_2:45   941.58   1635   0.22893775

341794   4/02/2015_6:05   976.46   2380   0.31873266

341798   4/02/2015_6:50   971.04   1759   0.24739191

341810   4/02/2015_8:30   976.17   1705   0.25575189

341944   5/02/2015_4:08   976.54   2268   0.30628975

341951   5/02/2015_5:35   974.37   1635   0.23624007

341963   5/02/2015_7:53   976.46   1911   0.27444743

341965   5/02/2015_8:33   976.52   2105   0.2951799

341967   5/02/2015_9:24   973.82   2675   0.33440304

Average block size = 965KB
Average number of tx = 1622 (or approximately 2.7 TPS)
Average block total fee = 0.25 BTC

These blocks are near to the 1MB limit and do not really permit that much real-world business to be transacted on the main-chain. If the number of people using Bitcoin, worldwide, increased to 10 million then everyone would only be able do one transaction per 43 days. This ignores that some blocks will always be mined small or empty while the reward remains high.

20MB blocks would offer 54 TPS and an average total fee per block of 5 BTC (based on this evidence) which means that fees would be comparable with the reward once the average block size hits 20MB and the reward is down to 6.25 BTC (within 6 years time).

The limit increase permits a much healthier outcome for the Bitcoin network and ecosystem.

Luckily, on a bitcoin forum, it is highly unlikely anyone would have serious savings in Brazilian reals.

Exactly. The existing 1MB limit is not keeping the blocks at an average of 30% to 40% full today, just as it was not keeping them 10% full three years ago. If the block size limit became 20MB tomorrow then it might still be four years before the first 20MB block occurs, and five years before they are common.


This worries me too. It is a serious risk. Once the blocks are an average of 75% full the network goes into a red-zone where a bunch of deliberately small or empty blocks can pile up the unconfirmed tx causing delays of hours for confirmations. It is a very cheap attack. Just sent some coins to an exchange to sell? Come back tomorrow if this attack is in progress. Yet some people are more worried about a few 20MB spam blocks being mined, which the network would hardly notice.

Kenji, you need a chart where the red candles are green, and the green ones are red. Then you can really let loose trading on your instincts.

Exactly! I also focus only on how many btc I am accruing. The amount of $$ is irrelevant.

I remember reading that theymos likes Iain Banks' sci-fi Culture novels. It was probably apparent to him early on that Bitcoin is a currency that would fit into that civilization. I think that DACs go to the next level with incredible potential, easily the backbone of an economy this size of the Culture.
Back on Earth, the future world economy will be transformed, and far more efficient, if DACs take off.

The side-chain (off-chain) solutions are also a compromise between performance and de/centralization, and logically it becomes a rats-nest:

Consider the case where 1MB is the block size limit and global bitcoin tx volumes are high enough to require 10MB blocks. Then at least 9 sidechains are needed. Unless one sidechain can have 9MB, or even 10MB blocks, which begs the question of why the BTC mainchain can't have 10MB blocks! Also wouldn't a 10MB sidechain just take over from the crippled 1MB mainchain?

If the offchains or sidechains are centralized in order to raise performance then there is a central point of failure (and central point for surveillance, and central point for government kill-switch) for 90% of the bitcoin transaction volume. This defeats the whole point of the decentralized peer-to-peer network. If the off-chains/sidechains are decentralized then they need a large network of full nodes. Where are they coming from?

Gavin's testing of 20MB blocks shows that loading (disk io) for these is 4x faster than 20 1MB blocks. So there are economies of scale which are not easily seen with lots of sidechains.  The off-chains/sidechains may have a role for micro-tx or faster block times, or 2.0 data storage, but don't help with handling tx volumes while still maintaining a fully decentralized payments system and currency.

Word clouds: https://ihb.io/2015-01-27/news/word-cloud-federal-reserve-vs-satoshi-15935

No mention of Paul Krugman either. Obviously don't want their nascent project to get torpedoed at birth from the Ivory Towers of La-La-Land.

They started using Raiffeisen in mid 2014

Only up to 100k euros.  Nothing compared to Bitstamp's funds.

http://www.bsi.si/en/guarantee-scheme.asp?MapaId=1047

Bitstamp, please consider moving all your customer's fiat out of this sick dog of a bank!

http://www.bloomberg.com/news/2015-01-20/raiffeisen-junior-debt-signaling-distress-as-currency-woes-mount.html