Bitcoin Forum

I am really wondering about BTC Block Time.

This morning only 6 Block found in 2 hours. Of course it depends on luck. But it depends also an HASH Rate. This is so enormous high and only adaped every 2 weeks.

Why does the biggest coin does not solve this difficulty problem? And why still more and more miners start so that the difficulty rises without watching the price.

It's not like that.
Bitcoin mining is designed to adapt to the total hashrate of the miners based on the previous block.
The difficulty will always change to maintain the average of 10 minutes per block. "Average" means that sometimes, it doesn't guarantee that the difficulty can accurately "harden" the block discovery.
So sometimes, a new block was discovered in less than 2minutes, sometimes, it can take 30 minutes to mine a block.

Ok, but when there are "waves" of hash rate in the two weeks, the coin can not adapt.

Depending on BTC.com the difficulty will rise about 3 % in 6 days, but as I said at the moment 5 to 6 Blocks in 2 hours....

The network behaves as it was initially designed. Every 2016 blocks there is a retarget, the difficulty gets adjusted upwards or downwards so the AVERAGE time between two blocks is ~10 minutes.
You are correct tough, if the networks hashrate would drop to 50% in the first 10 minutes after a retarget, it would take about 4 weeks untill the next retarget, at which point the time between two blocks would once again be ~10 minutes. This is still "working as designed" tough, theoretically, there is no problem... Sure, blocks would probably be completely full for the next 4 weeks, a fee war could start, it would take a longer time to get a confirmation,.... But everything would sort itself out, without technical problems.

The fact that at this moment, the time between the last 6 blocks was ~20 minutes (i didn't verify this): no problem, it's completely possible... That's why the difficulty is adjusted so the AVERAGE time between two blocks is ~10minutes, and that's also why we don't retarget every 10 or 20 blocks (if we would retarget to often, the difficulty would jump up and down quite a bit, since the number of samples would be to low).

Solving a block requires the miners to increment a nonce in the block header, then create a sha256d hash of this new header and check if the result in under the current target. It's possible that after a miner receives a new block and start working on the next block, he finds a header whose sha256d hash is under the target in less than a second, it's also possible he'll never find a valid block...

Want to send some BTC, now over 1 hour no block found.

For payment in  just NOT to use

Lots of options for reducing the retargeting time, Even keeping the same difficulty protocol I think lowering it to 1 Day (360 blocks) would be plenty.  Reading old posts people were worried about crazy difficulty oscillations, but DarkGravityWave doesn't seem to have that and it retargets every block.

Yes difficulty must be addapted more fast.

And why not halven block time and reward ?

Variance happens, difficulty has little to do with. Reducing retarget times to timeframes of eg. 1-2 hours you'd lose most of the network's reliability, stability and security.

Case in point: When Bitcoin Cash hardforked last year, they started off with retargets every 144 blocks (= one day). The result was a clusterfuck of hashrate oscillation between Bitcoin Cash and Bitcoin. Sometimes Bitcoin Cash would have a block every minute, causing the retarget period to only last a couple of hours. Sometimes Bitcoin Cash would swing towards the other extreme and have only 1 block every couple hours, practically grinding to a halt for days. So in a way Bitcoin Cash was the first coin to introduce banking holidays to cryptocurrencies. Joking aside, the latter was incredibly annoying, especially since many exchanges had to up the minimum required confirmation count to 12-24 confirmations, due to the questionable security and stability of the Bitcoin Cash network.

In other words, there's a lot of side effects to keep in mind. One may argue that 2 weeks are too long a timeframe, but a certain margin of error is required as to avoid hashrate fluctuations. Then again reducing the timeframe to other conservative timeframes, like for example 1 week, you'd still have variance in block intervals like you experienced.

It is also worth noting that for the most part of Bitcoin's history it's been ahead of its block intervals anyway. As the network grows, block intervals are on average shorter each difficulty period as they should be. Hence difficulty increasing. Hashrate being added to the network actually reduces the block interval within each period. It's only that at the beginning of each difficulty period the network adjusts back up to a block interval of 10 minutes.

Why are there more miners entering the market? Well, likely there's still profits to be made if you're able to minimize costs at the right corners. Old mining hardware may be phased out and replaced by more efficient and powerful miners. Old mining hardware may be bought up and brought back online by newcomers who have no idea what they are doing.

But today again same problem I really hate that. I send a transaction with normal fee and now waiting at the computer for about 2 hours and it is still not confirmed.

Why all crypto people use this Bitcoin Bullshit for payment. Yes for payment in Crypto it is bullshit.  ??? ??? ???

And dont tell about Lightning, this is child foot age. And nearly same as Master Nodes.

This probably happened because you have picked a fee which was too low.
Usually 1 sat/B transactions get confirmed pretty fast, but the mempool started growing a bit today. This happens in cycles.
You can check the current status here: jochen-hoenicke.de/queue/#1,8h (http://jochen-hoenicke.de/queue/#1,8h).

How high was your transaction fee ?
If you really need it to be confirmed as fast as possible, you might bump the fee (assuming you have the RBF-tag set; depends on the wallet you are using).




Bitcoin is the most proven and secured payment system around.
Each other crypto is either a clone or some untested system which massive vulnerabilities.




The LN is completely different from master nodes. You seem to have a misunderstanding of the LN.
If you want to get a deep understanding you might read want this article (part1 (https://bitcoinmagazine.com/articles/understanding-the-lightning-network-part-building-a-bidirectional-payment-channel-1464710791/) , part2 (https://bitcoinmagazine.com/articles/understanding-the-lightning-network-part-creating-the-network-1465326903/)).

If you instead don't want to spend 15 minutes of your life, you might look at the wikipedia article (https://en.wikipedia.org/wiki/Lightning_Network) instead.

Both sources should make it clear that the LN in not similar to the concept of master nodes.

So the fee was 6000 sat per kB. I think that is ok.

I tried to use Lightning network, all wallets even the offcial one I used were "not compaible" to the test I made (this picture you can paint for 1 sat per pixel)

What is the real difference. Some special nodes confirm the transaction faster like instant send in Dash?? Thats it, nothing really special 

That's 6 sat/B. 10 would have been needed for an instant confirmation today. Can be seen here (https://jochen-hoenicke.de/queue/#1,8h).




The lightning network still is in beta.
And most of the wallets available currently do only work with the testnet (except if you make changes to the configuration, then it is possible to use them on the main net).
So if you didn't explicitly configure them to work on the mainnet, they were configured for the testnet.




No. Thats not how it works.
At least use 3 minutes of your precious time to check the link i have posted (to have at least a bit of knowledge regarding LN):

Not necessarily true.. Eclair has a pretty stable mainnet version for android which I've been using for several weeks without a hitch. Lnd and c-lightning can also be run on the mainnet without having to reconfigure them (although you need a synced node for both)

I was under the impression that most wallets were for 'testnet only'. And that you could configure them to run on mainnet, but it was recommended not to do so.
It may have been like that a few months ago. But unfortunately i don't remember, i could also be wrong.

So you tell something now in beta is the future of payment. Otherwise you say something about prooved and secure.

Bitcoin was the first crypto that is ok. But dont treat it all like al holy cow. There have been some optimising with fork and thats ok too.
And also a 51% attack of BTC is possible. 2 pools more then 50% of hash rate. Now it is not useful but think of the future

Thanks for clearing the misconception of 2 weeks. so it is exactly 2016 block (14 days *24 hours * 6 block in an hour).

@OP.
Few days back I checked the hash rate graph and I found that difficulty always  increased with each passing month and it never dipped so I do not think that we ever encountered any such situation where re target might happened after more than 2 weeks.

Every thing has been in beta once. The LN is still in development stage. Did you expect the LN would be built within a few weeks ?




This is right. The bitcoin blockchain is the most secured one.
Not a single attack on bitcoin could really damage its network.
And since it has the most hashrate (and the most 'work' being used to mine) it is the crypto which needs the highest 'investment' to get attacked.




Everything is possible.
But both pools would effectively go bankrupt if they would decide to perform 51% attacks.
They would be left with worthless hardware and coins without any value anymore.

These pools actually are the most ambicious one letting bitcoin NOT get attacked. Thats how they make their money.

This simply isn't true, hashrate (thus difficulty) has decreased many times in the past.

I know, I was quoting the general trend , even if it dipped it made up for the dip.
https://i.imgur.com/padtUMl.png

I understand , my comment "it never dipped" was wrong , but in my mind I was seeing a linear line that is going up almost every month.

Mmmmh thats nothing new. My problem is when the core client suggest a fee I use it. And when I use the fee for 20 min and it takes 3 hours with twice double the fee, it is lets say difficult for payment

By the way one of my post was deleted. So as I said crypto community should not see BTC as the holy cow.

This is completely unrelated to the post you were quoting.
A transaction with a calculated fee (the fee is just an estimation of getting a confirmation after X minutes) does not have to confirm within this timeframe. This is depending on multiple factors.
The amount of transactions inside the mempool could increase, making your fee which currently was totally fine to get a confirmation the next block now isn't enough to get it confirmed within 10 blocks.
Or no blocks are being found at the moment. The difficulty still stays the same. And so does the average block time of 10 minutes. But 2 hours without a block isn't that unusual. But 3 Blocks within 10 minutes aren't that unusual either.

All these calculations of a fee are just a estimation. If you want to better estimate the fee, take a look at https://jochen-hoenicke.de/queue/#1,8h (https://jochen-hoenicke.de/queue/#1,8h). You will be able to guess the fee to use pretty clearly.

But still,  OP  is paying more fees for a speedier transaction (might user want to payment to be processed (a single confirmation) within an hour.
User put high fees in transaction , so that it will be confirmed in the next block.
If next block is not getting mined in next 3 hours then doing payment in next 1 hour is impossible for user whatever the fees user is paying.
Actually OP is saying he want to do speedier transaction by putting more fees but block mining speed becomes a limiting factor.

n a more simple words, Block time is the time which the networks takes to generate one extra block or the time taken from one Block to another.No one can say exactly how much Block Time will it take to mine another. The 10 minute window is just an average. There is a generation calculator that will tell you how long it might take to mine a block.

Someone with good statics knowlege could calculate how average the block time is really. So normally it should be a Gauss normal distribution with the maximum at 10 mins. But it is not, cause e.g. miners are activated or deactivated and difficulty can not react. Look how many blocks are found exactly 10 mins after the last one or even 11 or 9 min.

In theory bitcoin have blocks each 10 minutes, that mean 6 blocks each hour, and i say in theory because as you say before, some time we see a low number of blocks in lot of time. but at the same time i have seen before 3 blocks back to back.

As you say, bitcoin haven't fix this issue yet, and this was one of the fork reasons, bitcoin cash brings a solution where the difficulty changes each block. So if you don't like bitcoin for this reason i recommend you to give a try to bch. And lets wait to see if bitcoin take the solutions that bch brings for this issue.  

Who said it is an issue ?
This isn't an issue at all. It is necessary.

The adjustment each 2016 blocks is to keep a 10 minute average.




Changing difficulty each block is a horrible approach.
Mining a block is based on luck. And in a period with 2 or 3 blocks being mined successively the algorithm would adjust the difficulty to a level which practically wouldn't be mineable anymore.
Not to mention what kind of potential for abuse this offers.

That is the reason the interval of 2 weeks (2016 blocks with a 10-minute block time) has been chosen.
To adjust fast enough to not create a death spiral and to adjust slow enough to not get affected too much from short streaks with multiple blocks caused by 'luck' or an heavily increasing hashrate.




 ::)




An algorithm for difficulty adjustment should be a working feature (e.g. BTC). Not an issue or something 'necessary' which needs adjustments and optimization to somehow work stable (e.g. BCH).

As I mentioned before it is never a 10 min average. It is NOT a Gauss normal function.

The reason I do this discussion is that people dont want to hear about other crypto for payment. I dont understand

There is no maximum on the time between blocks, so there is no maximum at 10 minutes. The distribution time between blocks is not a normal distribution. Rather it is poisson with the true mean at ~9.7 minutes. The difficulty adjusts to make it 10 minutes, but new hardware coming online pulls the time down. Since it is poisson, the time can vary a lot, from 0 minutes to several hours, there is no maximum at 10 minutes.

The true mean for bitcoin since genesis block up to now, is ~9.4 minutes last time I calculated this divergence. How did you get 9.7?

Then you, merchant, payment processor or payment gateway should consider use/accept zero-confirmation, it's not too risky if it's for small payment and any explorer/nodes don't detect double-spend attempt (with sufficient fee obviously).
CPFP also help if you want to use your bitcoin immediately.


AFAIK that was mean in few years ago. My rough calculation also show 9.4 minutes mean time.