Quote from: DirtyKeyboard on Today at 07:18:57 AM

Quote from: matt4054 on Today at 06:25:52 AM

Quote from: Paashaas on Today at 05:59:32 AM

Why is there such a discrepancy between this and the counter from coinbase?

It appears the answer is, because Coinbase. Which is slightly troubling, because isn't coinbase handling all of the behind the scenes, off the chain transactions for the US ETFs? I really want to hate that font, but just can't.

In other news, how many forced HODLers were created this weekend? With ETF BTC being mostly frozen in time over the weekend. How many weekend BTC traders do you all think are employed by the ETFs. Or is it all, just do what the customer asks, during normal business hours?

It's because nobody actually knows the instantaneous hash rate of the Bitcoin network and it fluctuates wildly. Here's a three month chart of hash rate from ycharts.com:

Any calculation of time-to-block must include the hash rate but it can't be known with any certainty due to the probabilistic nature of mining. The Bitcoin protocol drives block time toward 10 minutes using the difficulty but this difficulty is only adjusted once every 2016 confirmed blocks. We don't even know the number of blocks remaining because it's possible (not so probably anymore) that there could be orphaned blocks. The only fixed variable we have for the calculation is the number 840,000 which is the block at which halving takes place. The second most certain variable we have is the current block which we can get by querying our bitcoin node with "getblockcount"; it's 839337

So there are almost certainly 663 blocks left then things just fall apart from there. If you decide to use 10 minutes per block, you get 4 days 14 hours and 30 minutes until the halvening... dammit, they just found another block while I was typing that. Okay, make it 4 days 14 hours and 20 minutes. BUT if you use 10 minutes in your calculations and don't revisit your countdown clock for ~4 years - you're going to be way off.

You can get fancier by using an average time-to-block in your calculation but then you have to pick a time frame for your average and after all that, if you don't regularly revisit your equation, then you're going to pull a Coinbase. My guess is that they paid someone to code a fancy halvening calculator for them and decided not to pay for ongoing support of the code. While it is still possible that the Coinbase countdown clock is the correct one, based on what we know presently, it's probably not.