Onyx Release & Mining Pools

[thelonecrouton]

I think we need to face the fact that the rapid evolution of Darkcoin, and mining pools, are simply not a good fit and never will be. Each step forward takes weeks longer than it would if miners were under direct MN control.

Centralised pools are a ball and chain for progress, and an outright security liability, plain and simple.

 

[Walter]

I think we need to face the fact that the rapid evolution of Darkcoin, and mining pools, are simply not a good fit and never will be. Each step forward takes weeks longer than it would if miners were under direct MN control.

Centralised pools are a ball and chain for progress, and an outright security liability, plain and simple.

I agree, a necessary evil as it stands... It's interesting to see the forces of economics at work here though! Centralisation and control go hand in glove with the desire and means to increase profit so it's going to be a difficult problem to solve.

 

[JGCMiner]

I think we need to face the fact that the rapid evolution of Darkcoin, and mining pools, are simply not a good fit and never will be. Each step forward takes weeks longer than it would if miners were under direct MN control.

Centralised pools are a ball and chain for progress, and an outright security liability, plain and simple.

You may be right, but in this instance the problem seems to be with Darkcoin -- as even the pools that are quick to update are have still having issues...

 

[thelonecrouton]

What amazes me is that coins that have no mining anchor still manage to make such little real progress compared to DRK. But think how much easier Evan's life would be as a developer if he wasn't constantly having to wait weeks on end for pools to get with the program.

Keep miners, just make them mine on Masternode pools: the centralisation issue goes away, the enforcement issue goes away, the updating issue goes away...

You may be right, but in this instance the problem seems to be with Darkcoin -- as even the pools that are quick to update are have still having issues...

True, but currently, rolling out a fix is still a tedious pain in the arse as we're dependent on pools to implement it.

 

[moli]

What amazes me is that coins that have no mining anchor still manage to make such little real progress compared to DRK. But think how much easier Evan's life would be as a developer if he wasn't constantly having to wait weeks on end for pools to get with the program.

Keep miners, just make them mine on Masternode pools: the centralisation issue goes away, the enforcement issue goes away, the updating issue goes away...



True, but currently, rolling out a fix is still a tedious pain in the arse as we're dependent on pools to implement it.

https://darkcointalk.org/threads/v10-15-onyx-release.2683/page-9#post-25967

 

[moli]

If you keep yelling for replacing one centralization with another centralization it's not going to help anything but just spamming this thread!

 

[thelonecrouton]

If you keep yelling for replacing one centralization with another centralization it's not going to help anything but just spamming this thread!

If you can't see the difference between 3 pools controlling 52% of the nethash vs. that nethash being distributed across 1000 Masternodes then I don't know what to say to you. Clue: one is a hell of a lot less centralised than the other.

 

[camosoul]

If you can't see the difference between 3 pools controlling 52% of the nethash vs. that nethash being distributed across 1000 Masternodes then I don't know what to say to you. Clue: one is a hell of a lot less centralised than the other.

Mining pools result in extreme centralization, and as we have seen, major hurdles in deployment and wilfull defiance.

I humbly suggest that priority list == InstantX, IP Obfuscation, MNs become fogpool.

Solo mining was all satoshi envisioned. The major flaw of mining pool hashpower consolidation has never been addressed on PoW or PoW/PoService hybrids. It's become ubiquitous that we have come to accept this serious problem as the norm instead of doing something about it.

MN [del]cloud[/del]fogpool may not be as decentralized as satoshi's original idea of everyone solo mining, but we know damn well that there is no way to avoid hashpower consolidation. Having 2 to 4 entities in power is far more consolidated than a few thousand masternodes. Stratum update: the MN mining cloud automatically adjusts your input node to the fastest/balanced input point to the cloud (fog). Since it is developed after IP Obfuscation, it's just a node ID...

The clearnet has the cloud.

DRK has the fog. Maybe we'll have DRK TOR on the to-do list again someday...

Fog: Quite a bit more down-to-earth than the Federal Reserve's pie-in-the-sky guvpaper. FRN's are the new Junk Bonds.

So much awesome new stuff that has never been done before...

I believe latency will become the real reason that we don't see ASICs on DRK...

 

[illodin]

If you can't see the difference between 3 pools controlling 52% of the nethash vs. that nethash being distributed across 1000 Masternodes then I don't know what to say to you. Clue: one is a hell of a lot less centralised than the other.

I don't know shit about p2pool, but what if someone sets up a masternode, and then offers people mining through his MN less fees, so miners have incentive to mine using his MN. Wouldn't this still possibly allow him 50%+ of the hash, and if so, would it matter in a p2pool?

 

[thelonecrouton]

I don't know shit about p2pool, but what if someone sets up a masternode, and then offers people mining through his MN less fees, so miners have incentive to mine using his MN. Wouldn't this still possibly allow him 50%+ of the hash, and if so, would it matter in a p2pool?

P2pool negates any 51% concerns. Why it isn't mandatory is beyond me, it would be a win for everyone. Centralised pools are a giant liability with every coin, not just DRK. They offer no advantage to anyone except pool ops, who could just as well be making their 1% or whatever running a p2pool node.

 

[camosoul]

I don't know shit about p2pool, but what if someone sets up a masternode, and then offers people mining through his MN less fees, so miners have incentive to mine using his MN. Wouldn't this still possibly allow him 50%+ of the hash, and if so, would it matter in a p2pool?

Who says he has that much control over it?

Let the MNs come to a consensus, and send the mining client's hashpower to the MN ID of the network's choosing. Those failing to submit from the specified MN ID get orphaned, so nobody can override it...

Internal p2pool, but with a few more checks and balances.

Official DRK mining software ships with client. Make the things modular... Hashing module, protocol module... Hash on X11, use fogminig protocol... Give it a DRK address and push the button. The MNs sort out everything else. How's that for lowing the barriers of entry for mining, while also distributing it much more akin to the model of solo mining than any crypto has ever seen since the first day BTC was instamined on CPU....

 

[illodin]

Cool.. another question - if everyone's p2pool mining, wouldn't this explode the blockchain with the amount of transactions it generates? Or can we use another blockchain just for the mining payouts, and move coins from the mining blockchain to the drk main blockchain when a full 1 DRK for example has been mined?

 

[thelonecrouton]

If handling something like a fully p2pool blockchain is going to break the network, we're screwed anyway and can kiss InstanTX and mass adoption goodbye.

3000 and growing Masternodes a year from now, lightweight (eg. electrum) clients/servers... if Mastercard can cope then so can we. People may need to upgrade their crappy t1 instances though... but not for a while, a single old Celeron could handle the whole BTC + DRK blockchain, and probably a dozen others, simultaneously. Go count how many transactions per minute there are... for chunks of silicon that operate in GHz, it ain't that big of a deal.

edit: been running masternode list on 15.14, not seeing many (any) :0's any more...

 

[oblox]

If handling something like a fully p2pool blockchain is going to break the network, we're screwed anyway and can kiss InstanTX and mass adoption goodbye.

3000 and growing Masternodes a year from now, lightweight (eg. electrum) clients/servers... if Mastercard can cope then so can we. People may need to upgrade their crappy t1 instances though... but not for a while, a single old Celeron could handle the whole BTC + DRK blockchain, and probably a dozen others, simultaneously. Go count how many transactions per minute there are... for chunks of silicon that operate in GHz, it ain't that big of a deal.

edit: been running masternode list on 15.14, not seeing many (any) :0's any more...

I wonder if that dos spam had a lot to do with the inactive masternodes and causing pools to pay nothing.

 

[camosoul]

If handling something like a fully p2pool blockchain is going to break the network...

Why would this happen? Why would it even be mentioned?

If a meteor falls out of the sky and jams itself up my butt, I'm screwed... But, you know, it's really not the sort of thing I worry about...

 

[eth1]

What amazes me is that coins that have no mining anchor still manage to make such little real progress compared to DRK. But think how much easier Evan's life would be as a developer if he wasn't constantly having to wait weeks on end for pools to get with the program.

Keep miners, just make them mine on Masternode pools: the centralisation issue goes away, the enforcement issue goes away, the updating issue goes away...



True, but currently, rolling out a fix is still a tedious pain in the arse as we're dependent on pools to implement it.

I would be totally against using a masternode as the pool wallet for the moment. I am operating a few masternodes aswell, and so far the stability of the masternode system doesn't make me want to use one as the pool central wallets.

Once the bugs are sorted out, that could be an idea.

 

[GermanRed+]

Mining pools result in extreme centralization, and as we have seen, major hurdles in deployment and wilfull defiance.

I humbly suggest that priority list == InstantX, IP Obfuscation, MNs become fogpool.

Solo mining was all satoshi envisioned. The major flaw of mining pool hashpower consolidation has never been addressed on PoW or PoW/PoService hybrids. It's become ubiquitous that we have come to accept this serious problem as the norm instead of doing something about it.

MN [del]cloud[/del]fogpool may not be as decentralized as satoshi's original idea of everyone solo mining, but we know damn well that there is no way to avoid hashpower consolidation. Having 2 to 4 entities in power is far more consolidated than a few thousand masternodes. Stratum update: the MN mining cloud automatically adjusts your input node to the fastest/balanced input point to the cloud (fog). Since it is developed after IP Obfuscation, it's just a node ID...

The clearnet has the cloud.

DRK has the fog. Maybe we'll have DRK TOR on the to-do list again someday...

Fog: Quite a bit more down-to-earth than the Federal Reserve's pie-in-the-sky guvpaper. FRN's are the new Junk Bonds.

So much awesome new stuff that has never been done before...

I believe latency will become the real reason that we don't see ASICs on DRK...

I like the the idea but how we can migrate from centralized pools to fogpool? If I understand it correctly, the network can split if the miners don't agree.

 

[GermanRed+]

Mining pools result in extreme centralization, and as we have seen, major hurdles in deployment and wilfull defiance.

I humbly suggest that priority list == InstantX, IP Obfuscation, MNs become fogpool.

Solo mining was all satoshi envisioned. The major flaw of mining pool hashpower consolidation has never been addressed on PoW or PoW/PoService hybrids. It's become ubiquitous that we have come to accept this serious problem as the norm instead of doing something about it.

MN [del]cloud[/del]fogpool may not be as decentralized as satoshi's original idea of everyone solo mining, but we know damn well that there is no way to avoid hashpower consolidation. Having 2 to 4 entities in power is far more consolidated than a few thousand masternodes. Stratum update: the MN mining cloud automatically adjusts your input node to the fastest/balanced input point to the cloud (fog). Since it is developed after IP Obfuscation, it's just a node ID...

The clearnet has the cloud.

DRK has the fog. Maybe we'll have DRK TOR on the to-do list again someday...

Fog: Quite a bit more down-to-earth than the Federal Reserve's pie-in-the-sky guvpaper. FRN's are the new Junk Bonds.

So much awesome new stuff that has never been done before...

I believe latency will become the real reason that we don't see ASICs on DRK...

Can the miner still cheat by having his/her own customized MN?

 

[eth1]

I like the the idea but how we can migrate from centralized pools to fogpool? If I understand it correctly, the network can split if the miners don't agree.

It can, and that's still good. The miners have a word to say here, they are the ones paying the masternodes as far as I can remember right? Doing so further centralises the masternode side of things, but having two separate entities is good for decentralisation imo.

 

[moli]

It can, and that's still good. The miners have a word to say here, they are the ones paying the masternodes as far as I can remember right? Doing so further centralises the masternode side of things, but having two separate entities is good for decentralisation imo.

You're wrong... I'm with lostcrouton here... Masternodes are owned by 1000+ individuals, while pools are owned by only 3 individuals.. Clue: MN is less centralized, get it?
:grin::grin: