Something Odd Is Happening at Bitcoin’s Largest Mining Pool Read more: http://w

[alyssa85]

http://www.nasdaq.com/article/something-odd-is-happening-at-bitcoins-largest-mining-pool-cm756139

Something Odd Is Happening at Bitcoin's Largest Mining Pool

Blocks on the Bitcoin network have been increasingly full lately , except for some of the ones mined by AntPool , which is the largest mining pool on the network . In the past 24 hours from the time this article was written, every non- empty block mined by someone other than AntPool was practically full of transactions, nearly up to the 1 MB limit.

Let's take a closer look at the details regarding these smaller blocks.

Recently, AntPool has been mining a number of blocks with sizes of around 99 KB, 369 KB and 860 KB. There were dozens of blocks mined around these specific sizes during the month of February. During the times these blocks were mined, everyone else on the network was filling blocks with transactions up to the 1 MB capacity limit.

In addition to the non-full blocks mined by AntPool, the mining pool also created 16 empty blocks in the month of February. The total amount of transaction capacity lost by the network during this time as a result of AntPool's small blocks is not difficult to estimate. Numbers shared by BitFury's Alex Petrov show AntPool's average mined block size in February was around 100 KB less than other mining pools of comparable size.

[Carlton Banks]

Strange.

Why would the largest miner on the network be stifling the capacity we already have? The owner of AntMiner claims to support the on-chain transaction model. This news seems to contradict that position.

[flora_digitalis]

Strange.

Why would the largest miner on the network be stifling the capacity we already have? The owner of AntMiner claims to support the on-chain transaction model. This news seems to contradict that position.

The owner is pro Unlimited Coin and wants to artificially create the impression of capacity problems to force a shift in public opinion towards bigger blocks. It's the same old strategy the Andresen/Hearn/Ver crowd and their shills is using since months.

[franky1]

its not about band camp arguing. its about network propogation..

even BTCC (pro segwit) has near empty blocks
455807 (Main Chain) BTCC Pool 00000000000000000107a158529ac9448946981489894a00075b2a9968f567b7 0.27mb  2017-03-05 03:39:05

the reason is rather apparent
455907 2017-03-05 03:39:05
455806 2017-03-05 03:38:44
hint: 21 seconds

in the 21 seconds to see 806 exists.... then get 806... btcc started mining its own new block with transactions it knew were NOT in 806(because its their own tx's or other reasons so they know they were not relayed at 806) so deemed it safe to include them. rather than do the usual empty block philosophy of mine new while validating old including none at all.

but btcc was LUCKY enough to get a block solution to 807 before being able to fully validated 806 and so didnt have time to add more tx's to their block after confirming 806. because 807 had a solution.


in short.. if you see a block not at over 900k.. and instead low kb.. check the time between blocks.. if its short. EG under 2minutes.. its usually about the empty block game of starting a new block BEFORE validating the old block thus not able to fill the new block because you are stil unsure whats truly 'unconfirmed' to allow into a new block.


if a block is near empty and its taken ~5min+ and no sign of any holdups in previous block due to sigop spamming.. then and only then treat it as intentional ignoring unconfirm additions

[Kprawn]

Strange.

Why would the largest miner on the network be stifling the capacity we already have? The owner of AntMiner claims to support the on-chain transaction model. This news seems to contradict that position.

The owner is pro Unlimited Coin and wants to artificially create the impression of capacity problems to force a shift in public opinion towards bigger blocks. It's the same old strategy the Andresen/Hearn/Ver crowd and their shills is using since months.

Can this be seen as an "attack" on Bitcoin, like Luke Jr is saying about the "vote" testing that are going to be done. This is very selective and can

very easily go unnoticed. https://www.reddit.com/r/Bitcoin/comments/5xkvc1/psa_were_running_a_stress_test_of_our_blockchain/ ... If you

combine this with the vote "spam" then this can easily lead to more congestion. 

[franky1]

as for the "spam" and people presuming its BU driven....

check out the unconfirmed mempool sizes over the last year
https://blockchain.info/charts/mempool-size?daysAverageString=7&timespan=1year

apart from a blip in june/july (hmm ill get to this later)
mempool size was 2mb-3mb..

then in october(hmm ill get to this later). (around the 10th october) it started to get into 4mb.. and started rising...and rising. and rising..

hmmm i wonder what new feature was introduced in october that needed the community to get frustrated and delayed to "sell" the community into thinking this feature being the promise to fix the issue..

hmm.. oh yea.. segwit.

as for june july. (CLTV+CSV)

what a coincidence.. spam attacks right around the times blockstream devs want pools to think certain features need to be added to be stepping stones to something that (*may* cough wont cough) eventually fix spam attacks..

[unamis76]

This has been ongoing for a lot of time and has been previously discussed, probably in the mining section... Pool operators don't always (or at all) do it on purpose. This has to do with "SPV mining" and as said before, time between blocks (franky1 explains it nicely).

That being said... Since when is Nasdaq monitoring the blockchain?

[odolvlobo]

My guess is that they are testing propagation times or orphan rates of blocks of different sizes.

BTW, there is no such thing as "congestion". If someone's transaction is not being included in a block, it is not because there is "congestion" blocking their transaction. It is because other people are willing to pay more to get their blocks included sooner. Paying more to get a transaction included can hardly be called "congestion" or even "spam".

[ImHash]

Nothing odd just as usual, I always thought mining bitcoins is only possible as long as you include transactions into blocks but now I understand after finding an empty block in hashnest it's all good and normal go poke your noses some where else please.

[AngryDwarf]

The answer is really contained in post 4. I believe it's called SPV mining, where they only validate the previous block header and not the transactions in it before attempting to find the next block. So it would seem that miners are still more interested in the block rewards than the transaction fees in it.

[deisik]

Strange.

Why would the largest miner on the network be stifling the capacity we already have? The owner of AntMiner claims to support the on-chain transaction model. This news seems to contradict that position

As to me, there is nothing strange in that

In fact, I have been expecting something like that. Miners are just trying to squeeze more profits later by stifling the capacity now. It is a bit counterintuitive (I understand your pain), but that's how things work in general (and worked out in the past numerous times). With more transactions unconfirmed today, miners will get more profits tomorrow. Folks will be forced to pay higher fees, and gradually they will get accustomed to high fees and then the pools will fully exploit that. This is quite in line with what diverse monopolies are doing every day all over the world in any field. Any monopoly is evil unless it gets bridled somehow, and miners are simply using and abusing their position

[Hydrogen]

Recently, AntPool has been mining a number of blocks with sizes of around 99 KB, 369 KB and 860 KB.

This seems to suggest increasing block size won't do much to speed up unconfirmed transactions or reduce fees.

Is there a reason to increase block size if many 1MB blocks are only 10%, 25% or 75% full?

[AngryDwarf]

Protocol has to strip miners choice of transaction inclusion, otherwise #4 and #11 comments applies.

[LLec]

This is the reason for all the slow down lately?
It is causing unnessary delays in the blockchain from sending and just sending more business for already saturated websites like ViaBtc which I am assuming is the only of it's kind running to accommodate all those requests to speed up their sends.

[AngryDwarf]

Vast majority of blocks are full to the limit.

[Variogam]

Recently, AntPool has been mining a number of blocks with sizes of around 99 KB, 369 KB and 860 KB.

This seems to suggest increasing block size won't do much to speed up unconfirmed transactions or reduce fees.

Is there a reason to increase block size if many 1MB blocks are only 10%, 25% or 75% full?

While SPV mining is one reason for not full blocks (as franky1 nicely explained), the second reason is sometimes miner adds transactions with certain fee/byte only. In the past I have seen blocks around 500-900KB, with no transaction included under for example 25 Satoshi/byte.

So thinking there going to be always full blocks filled up to maximum capacity is wrong, its the miners who decide what transactions to add and some define spam as for example under 25 Satoshi/byte transaction not worth adding even if there is available space.

Because during some days/hours people send more transactions (peak times), it make sense to increase block sizes above 1MB, so the Bitcoin user experience is good even at peak times.

[franky1]

So thinking there going to be always full blocks filled up to maximum capacity is wrong, its the miners who decide what transactions to add and some define spam as for example under 25 Satoshi/byte transaction not worth adding even if there is available space.

you can thank the pools that support blockstream for that possibility of occurance.
the removal of reactive fee acceptance ( natural demand) and sticking to blockstreams 'average fee' relay rules. means even in low demand average fee estimate rules are still high. leading to ignoring certain low fee tx's during low demand

[Victorycoin]

Strange.

Why would the largest miner on the network be stifling the capacity we already have? The owner of AntMiner claims to support the on-chain transaction model. This news seems to contradict that position

As to me, there is nothing strange in that

In fact, I have been expecting something like that. Miners are just trying to squeeze more profits later by stifling the capacity now. It is a bit counterintuitive (I understand your pain), but that's how things work in general (and worked out in the past numerous times). With more transactions unconfirmed today, miners will get more profits tomorrow. Folks will be forced to pay higher fees, and gradually they will get accustomed to high fees and then the pools will fully exploit that. This is quite in line with what diverse monopolies are doing every day all over the world in any field. Any monopoly is evil unless it gets bridled somehow, and miners are simply using and abusing their position

More disturbing when one take into account the fact that majority of mining rigs are situated in China and monopoly of all things in the hand of the Chinese doesn't bide well for the rest of the world. Something need be done to address this ugly part of Bitcoin before it metamorphosed into a point of departure.

[AliceWonderMiscreations]

More disturbing when one take into account the fact that majority of mining rigs are situated in China and monopoly of all things in the hand of the Chinese doesn't bide well for the rest of the world. Something need be done to address this ugly part of Bitcoin before it metamorphosed into a point of departure.

And what exactly have they done to harm bitcoin?

They are people, just like those of us who do not live in China.

It is in their best interest for bitcoin to succeed, they invest a lot of money in it.

Why would they damage the network, and what evidence do you have to suggest they have tried?

[odolvlobo]

...
So thinking there going to be always full blocks filled up to maximum capacity is wrong, its the miners who decide what transactions to add and some define spam as for example under 25 Satoshi/byte transaction not worth adding even if there is available space.

Why wouldn't a miner include a 25 s/b transaction? It certainly doesn't cost them that much to include it. Why wouldn't it be worth including it?

Because during some days/hours people send more transactions (peak times), it make sense to increase block sizes above 1MB, so the Bitcoin user experience is good even at peak times.

In that case, it would be natural for the fees to be higher during times of high demand and lower during times of low demand. No change of block size is necessary.