Topic: [ANN][DASH] Dash (dash.org) | First Self-Funding Self-Governing Crypto Currency - page 1180. (Read 9723768 times)

Quote from: eduffield on April 04, 2016, 05:31:37 PM

....

Bitcoin's halving is around the corner and there is a concern that this will create mempool transaction problems on Bitcoin's network on a larger scale then previously witnessed
...

Why is the halving expected to increase Bitcoin traffic / mempool size?
...

Some predict that after halving and no price increase some miners could turn their machines off and hashrate could drop because of that more or less significantly. That would be ok but bitcoin difficulty adjusts every 2016 blocks so miners who are still mining would have more work to be done to solve a block than it should be for a hashrate like that for quite a long time (normally 2016 is 2 weeks and with lower hashrate this timeframe grows). This would mean larger time between blocks which in its turn would mean more transactions waiting in line to get into the block which could lead for confirmation problems which could decrease the price which will make it unprofitable for more miners which will turn their miners off and there you go - "death spiral".

I've been thinking about this problem too recently. Actual size usage of blocks is somewhere ~800kb, which means we could see throughput of ~400kb per block for up to a month after the halving. The reason for this is pretty simple, if the profitability of the network drops 50% in one moment, 50% of the miners should drop off. Also, the difficulty only readjusts every 2 weeks, which will take 4 weeks with 50% of the miners. If we're using ~800kb in transactions, doesn't that mean we'll start added 400kb to the blocklog... every 10 minutes? What happens after 30 days of that? A 1.72GB mempool

It's worth noting, miners didn't drop off last time. But the climate is completely different now. I have no idea what happens next, in either case it's sure to be entertaining. Wink

[1] That only applies if miners are on average mining break even. With all those Chinese miners who are basically mining for free I presume on average miners are mining with a profit. You are right about the remainder. Worst case scenario the hashrate drops so much that you would need a emergency hardfork to adjust the difficulty.

[2] Not entirely true, the hashrate "ramped up" in the summer, but you can see a decrease from around July 2012 until and somewhat after the halving (which was end november 2012 if I recall correctly).

https://blockchain.info/charts/hash-rate?showDataPoints=false&show_header=true&daysAverageString=1×pan=all&scale=1&address=

How does it come that Chinese miners are mining for free?

dEBRUYNE

legendary

Activity: 2268

Merit: 1141

Quote from: afbitcoins on April 04, 2016, 05:38:46 PM

....

Bitcoin's halving is around the corner and there is a concern that this will create mempool transaction problems on Bitcoin's network on a larger scale then previously witnessed
...

Why is the halving expected to increase Bitcoin traffic / mempool size?
...

Some predict that after halving and no price increase some miners could turn their machines off and hashrate could drop because of that more or less significantly. That would be ok but bitcoin difficulty adjusts every 2016 blocks so miners who are still mining would have more work to be done to solve a block than it should be for a hashrate like that for quite a long time (normally 2016 is 2 weeks and with lower hashrate this timeframe grows). This would mean larger time between blocks which in its turn would mean more transactions waiting in line to get into the block which could lead for confirmation problems which could decrease the price which will make it unprofitable for more miners which will turn their miners off and there you go - "death spiral".

I've been thinking about this problem too recently. Actual size usage of blocks is somewhere ~800kb, which means we could see throughput of ~400kb per block for up to a month after the halving. The reason for this is pretty simple, if the profitability of the network drops 50% in one moment, 50% of the miners should drop off. Also, the difficulty only readjusts every 2 weeks, which will take 4 weeks with 50% of the miners. If we're using ~800kb in transactions, doesn't that mean we'll start added 400kb to the blocklog... every 10 minutes? What happens after 30 days of that? A 1.72GB mempool

It's worth noting, miners didn't drop off last time. But the climate is completely different now. I have no idea what happens next, in either case it's sure to be entertaining. Wink

[1] That only applies if miners are on average mining break even. With all those Chinese miners who are basically mining for free I presume on average miners are mining with a profit. You are right about the remainder. Worst case scenario the hashrate drops so much that you would need a emergency hardfork to adjust the difficulty.

[2] Not entirely true, the hashrate "ramped up" in the summer, but you can see a decrease from around July 2012 until and somewhat after the halving (which was end november 2012 if I recall correctly).

https://blockchain.info/charts/hash-rate?showDataPoints=false&show_header=true&daysAverageString=1×pan=all&scale=1&address=

eduffield

legendary

Activity: 1176

Merit: 1036

Dash Developer

Why does the electrum client need a network of nodes? Isn't that exactly what masternodes already provide ?

Electrum uses a centralized service to provide SPV. We'll eventually provide that service in a decentralized way, but that's a later release.

eduffield

legendary

Activity: 1176

Merit: 1036

Dash Developer

Quote from: UdjinM6 on April 04, 2016, 05:37:56 PM

On another note here is what Adam Back proposed on twitter few hours ago:

Quote

Halvening: if miners were concerned they can softfork difficulty down 50% pre and release at halvening. Harmless just speeds blocks for week

https://twitter.com/adam3us/status/717023841529114625

Not sure if he is serious or what... Huh

Kidding... I hope?

afbitcoins

legendary

Activity: 2101

Merit: 1061

Why does the electrum client need a network of nodes? Isn't that exactly what masternodes already provide ?

UdjinM6

legendary

Activity: 1318

Merit: 1040

On another note here is what Adam Back proposed on twitter few hours ago:

Quote

Halvening: if miners were concerned they can softfork difficulty down 50% pre and release at halvening. Harmless just speeds blocks for week

https://twitter.com/adam3us/status/717023841529114625

Not sure if he is serious or what... Huh

eduffield

legendary

Activity: 1176

Merit: 1036

Dash Developer

....

Bitcoin's halving is around the corner and there is a concern that this will create mempool transaction problems on Bitcoin's network on a larger scale then previously witnessed
...

Why is the halving expected to increase Bitcoin traffic / mempool size?
...

Some predict that after halving and no price increase some miners could turn their machines off and hashrate could drop because of that more or less significantly. That would be ok but bitcoin difficulty adjusts every 2016 blocks so miners who are still mining would have more work to be done to solve a block than it should be for a hashrate like that for quite a long time (normally 2016 is 2 weeks and with lower hashrate this timeframe grows). This would mean larger time between blocks which in its turn would mean more transactions waiting in line to get into the block which could lead for confirmation problems which could decrease the price which will make it unprofitable for more miners which will turn their miners off and there you go - "death spiral".

I've been thinking about this problem too recently. Actual size usage of blocks is somewhere ~800kb, which means we could see throughput of ~400kb per block for up to a month after the halving. The reason for this is pretty simple, if the profitability of the network drops 50% in one moment, 50% of the miners should drop off. Also, the difficulty only readjusts every 2 weeks, which will take 4 weeks with 50% of the miners. If we're using ~800kb in transactions, doesn't that mean we'll start added 400kb to the blocklog... every 10 minutes? What happens after 30 days of that? A 1.72GB mempool

It's worth noting, miners didn't drop off last time. But the climate is completely different now. I have no idea what happens next, in either case it's sure to be entertaining. Wink

ddink7

legendary

Activity: 1120

Merit: 1000

....

Bitcoin's halving is around the corner and there is a concern that this will create mempool transaction problems on Bitcoin's network on a larger scale then previously witnessed
...

Why is the halving expected to increase Bitcoin traffic / mempool size?
...

Some predict that after halving and no price increase some miners could turn their machines off and hashrate could drop because of that more or less significantly. That would be ok but bitcoin difficulty adjusts every 2016 blocks so miners who are still mining would have more work to be done to solve a block than it should be for a hashrate like that for quite a long time (normally 2016 is 2 weeks and with lower hashrate this timeframe grows). This would mean larger time between blocks which in its turn would mean more transactions waiting in line to get into the block which could lead for confirmation problems which could decrease the price which will make it unprofitable for more miners which will turn their miners off and there you go - "death spiral".

However some say that that's not going to happen because either price would increase thanks to increased scarcity or because of press covering halving event or simply because miners rent their electricity ahead of a time and a) they have no choice but to mine because they already payed and b) they have made million $$ investments into mining and should have foreseen this situation already.

I have no idea who is correct, so I'll just grab some popcorn and prepare for fireworks Grin

Cool, makes sense. Thanks!

I have thought about the halving as well, and I just don't see a significant drop in network hash rate. Since the mining ASICs have already been purchased, their cost was recognized up-front. Now the only cost to keep them running is the cost of electricity. I foresee only the oldest and most energy-inefficient miners dropping off the network. Everybody else should (in theory) continue running their miners, as long as the expected reward exceeds the expected electrical costs.

I predict a 10-20% reduction in hashpower...with a slower growth in hashrate unless higher prices come into play.

UdjinM6

legendary

Activity: 1318

Merit: 1040