• Forum has been upgraded, all links, images, etc are as they were. Please see Official Announcements for more information

Development Update - Oct 1, 2014

What about more masternode pooling services like Vertoe's? This would mean people with less than 1000DRK could create masternodes, the count would go up, but the collateral and the reward stays the same. Couldnt we try that first?
This definitely will lower the barrier to entry, and you can bet Vertoe will be a good actor.
I haven't researched how these work though. Must the coins go to the pooling service? If so, there is a trust issue there.
In NXT, one can delegate his/her proof of stake without losing control of the coin. They call it leased forging. We need a mechanism like this to remove the issue of trust.
 
Last edited by a moderator:
This definitely will lower the barrier to entry, and you can bet vertoe will be a good actor.
I haven't researched how these work though. Must the coins go to the pooling service? If so, there is a trust issue there.
In NXT, one can delegate his/her proof of stake without losing control of the coin. We need a mechanism like this to remove the issue of trust.

eduffield If Evan could develop a mechanism for trustless masternode pooling, it would definitely raise the masternode count substantially, the pools dont need to be big, just a way for 3 or 4 people to pool together to create a masternode. Maybe using the same mechanism that checks for the 1000 DRK stake, you could check for 4 independent 250DRK stakes to validate a masternode, and the shareholders would have to do masternode start from each of their wallets using the same masternode genkey, if one of the 4 shareholders moves the coins the masternode goes offline. It can obviously also be 2 people with 500DRK same principle. This could be huge.

Edit: This would be a very smart move, just like when we introduced cold wallet masternodes, without that, we wouldn’t have half the nodes we have now. Trustless masternode pooling would definitely help, regardless of anything else we do. So if it was technically possible I would do that first to see what impact it could have on the count.
 
Last edited by a moderator:
I wanted to add some insight to all the comments on here about "couldn't we just decrease the requirement for MNs to 500 DRK and double the number?" or "is increasing the payout ratio really the right way to force an increase in the number of MNs?".

If you decrease the requirement to 500, here would be the effects:

1) The cost for all the MN operators to run their MNs would double (twice as many servers to operate), assuming that they kept their 1000 DRK fully invested in MNs. This would reduce the after-expense profits for MN operators and should therefore result in a slight reduction in their number (so you will not double the number). It will also decrease the cost of each individual MN and therefore provide less of an incentive for MN operators to properly secure their servers.

2) Since the number of MNs would nearly double overnight, the frequency of payouts would drop by half for each MN in operation. This would reduce the ROI of a masternode in half from ~23% to ~11.5%, at which point many of the market participants would find that rate of return too low given this is a risky class of asset. MN operators would then start selling their DRK, decreasing the number of MNs, until the payout rate once again reached equilibrium at (I'm assuming) about 23%. This would have the effect of flooding the market with DRK and driving the price down. Basically, you will end up with the same number of MNs, a lower price for the coin, and less profits (due to the higher expenses) for the MN operators.

3) Therefore, the ONLY mechanism we have for increasing the number of MN operators using free-market mechanisms is to increase the payout to them. Playing with the MN amount needed to start one theoretically should have no effect.

I think the issue comes down to "what is the right balance to reach between incentives to miners and masternodes that will optimally provide sufficient security and sufficient number of MNs?" And that is the debate this community should be having. If the amount of hash power is sufficient to secure the coin AND provide for the number of MNs we need, then we have the answer. You could also debate what the right payout amount is (currently about 5 per block) to sufficiently incentivize both miners and MN owners, if the supply is insufficient.

One final thought.... increasing the payout ratio will also create demand for DRK (to start masternodes). This should increase the price of the coin and therefore could actually INCREASE the fiat value of the rewards being paid to miners (even if it is a lower absolute number of DRK). Therefore, we could increase the absolute amount of hashpower securing the coin while simultaneously increasing the number of MNs and achieve both goals and make both groups ultimately happy. Even if miners are better off, most will not see the correlation and will complain about their share. Some may even stop mining the coin. So be it. The mining world has an equilibrium supply of mining as well, and if that supply drops, it will leave the remaining suppliers more profitable. The "miners will all leave" argument is therefore a red herring and isn't warranted. Only if you argue that we end up with too few miners to secure the network is this truly valid.

I hope this helps frame the issues for everyone and helps with understanding why MN payments is THE key mechanism to grow the MN network.
 
I wanted to add some insight to all the comments on here about "couldn't we just decrease the requirement for MNs to 500 DRK and double the number?" or "is increasing the payout ratio really the right way to force an increase in the number of MNs?".

If you decrease the requirement to 500, here would be the effects:

1) The cost for all the MN operators to run their MNs would double (twice as many servers to operate), assuming that they kept their 1000 DRK fully invested in MNs. This would reduce the after-expense profits for MN operators and should therefore result in a slight reduction in their number (so you will not double the number). It will also decrease the cost of each individual MN and therefore provide less of an incentive for MN operators to properly secure their servers.

2) Since the number of MNs would nearly double overnight, the frequency of payouts would drop by half for each MN in operation. This would reduce the ROI of a masternode in half from ~23% to ~11.5%, at which point many of the market participants would find that rate of return too low given this is a risky class of asset. MN operators would then start selling their DRK, decreasing the number of MNs, until the payout rate once again reached equilibrium at (I'm assuming) about 23%. This would have the effect of flooding the market with DRK and driving the price down. Basically, you will end up with the same number of MNs, a lower price for the coin, and less profits (due to the higher expenses) for the MN operators.

3) Therefore, the ONLY mechanism we have for increasing the number of MN operators using free-market mechanisms is to increase the payout to them. Playing with the MN amount needed to start one theoretically should have no effect.

I think the issue comes down to "what is the right balance to reach between incentives to miners and masternodes that will optimally provide sufficient security and sufficient number of MNs?" And that is the debate this community should be having. If the amount of hash power is sufficient to secure the coin AND provide for the number of MNs we need, then we have the answer. You could also debate what the right payout amount is (currently about 5 per block) to sufficiently incentivize both miners and MN owners, if the supply is insufficient.

One final thought.... increasing the payout ratio will also create demand for DRK (to start masternodes). This should increase the price of the coin and therefore could actually INCREASE the fiat value of the rewards being paid to miners (even if it is a lower absolute number of DRK). Therefore, we could increase the absolute amount of hashpower securing the coin while simultaneously increasing the number of MNs and achieve both goals and make both groups ultimately happy. Even if miners are better off, most will not see the correlation and will complain about their share. Some may even stop mining the coin. So be it. The mining world has an equilibrium supply of mining as well, and if that supply drops, it will leave the remaining suppliers more profitable. The "miners will all leave" argument is therefore a red herring and isn't warranted. Only if you argue that we end up with too few miners to secure the network is this truly valid.

I hope this helps frame the issues for everyone and helps with understanding why MN payments is THE key mechanism to grow the MN network.

I agree with all of the above, reducing the collateral is not really an option, but what could be great is to create a mechanism for trustless masternode pooling, as I described in my post above. I don't know if this is technically possible but if it was it would be a great solution. People with smaller stakes could pool together and create masternodes with 0 risk, this will take new coins out of circulation and have a positive effect on price. If at all possible, we could do this first and see where the count settles. Having said that, a raise in masternode payouts may still be necessary later on, but I think is a worthwhile initiative if at all possible.
 
I wanted to add some insight to all the comments on here about "couldn't we just decrease the requirement for MNs to 500 DRK and double the number?" or "is increasing the payout ratio really the right way to force an increase in the number of MNs?".

If you decrease the requirement to 500, here would be the effects:

1) The cost for all the MN operators to run their MNs would double (twice as many servers to operate), assuming that they kept their 1000 DRK fully invested in MNs. This would reduce the after-expense profits for MN operators and should therefore result in a slight reduction in their number (so you will not double the number). It will also decrease the cost of each individual MN and therefore provide less of an incentive for MN operators to properly secure their servers.

2) Since the number of MNs would nearly double overnight, the frequency of payouts would drop by half for each MN in operation. This would reduce the ROI of a masternode in half from ~23% to ~11.5%, at which point many of the market participants would find that rate of return too low given this is a risky class of asset. MN operators would then start selling their DRK, decreasing the number of MNs, until the payout rate once again reached equilibrium at (I'm assuming) about 23%. This would have the effect of flooding the market with DRK and driving the price down. Basically, you will end up with the same number of MNs, a lower price for the coin, and less profits (due to the higher expenses) for the MN operators.

3) Therefore, the ONLY mechanism we have for increasing the number of MN operators using free-market mechanisms is to increase the payout to them. Playing with the MN amount needed to start one theoretically should have no effect.

I think the issue comes down to "what is the right balance to reach between incentives to miners and masternodes that will optimally provide sufficient security and sufficient number of MNs?" And that is the debate this community should be having. If the amount of hash power is sufficient to secure the coin AND provide for the number of MNs we need, then we have the answer. You could also debate what the right payout amount is (currently about 5 per block) to sufficiently incentivize both miners and MN owners, if the supply is insufficient.

One final thought.... increasing the payout ratio will also create demand for DRK (to start masternodes). This should increase the price of the coin and therefore could actually INCREASE the fiat value of the rewards being paid to miners (even if it is a lower absolute number of DRK). Therefore, we could increase the absolute amount of hashpower securing the coin while simultaneously increasing the number of MNs and achieve both goals and make both groups ultimately happy. Even if miners are better off, most will not see the correlation and will complain about their share. Some may even stop mining the coin. So be it. The mining world has an equilibrium supply of mining as well, and if that supply drops, it will leave the remaining suppliers more profitable. The "miners will all leave" argument is therefore a red herring and isn't warranted. Only if you argue that we end up with too few miners to secure the network is this truly valid.

I hope this helps frame the issues for everyone and helps with understanding why MN payments is THE key mechanism to grow the MN network.

Great post. You articulated what I was thinking. Cheers, mate!
 
I wanted to add some insight to all the comments on here about "couldn't we just decrease the requirement for MNs to 500 DRK and double the number?" or "is increasing the payout ratio really the right way to force an increase in the number of MNs?".

If you decrease the requirement to 500, here would be the effects:

1) The cost for all the MN operators to run their MNs would double (twice as many servers to operate), assuming that they kept their 1000 DRK fully invested in MNs. This would reduce the after-expense profits for MN operators and should therefore result in a slight reduction in their number (so you will not double the number). It will also decrease the cost of each individual MN and therefore provide less of an incentive for MN operators to properly secure their servers.

2) Since the number of MNs would nearly double overnight, the frequency of payouts would drop by half for each MN in operation. This would reduce the ROI of a masternode in half from ~23% to ~11.5%, at which point many of the market participants would find that rate of return too low given this is a risky class of asset. MN operators would then start selling their DRK, decreasing the number of MNs, until the payout rate once again reached equilibrium at (I'm assuming) about 23%. This would have the effect of flooding the market with DRK and driving the price down. Basically, you will end up with the same number of MNs, a lower price for the coin, and less profits (due to the higher expenses) for the MN operators.

3) Therefore, the ONLY mechanism we have for increasing the number of MN operators using free-market mechanisms is to increase the payout to them. Playing with the MN amount needed to start one theoretically should have no effect.

I think the issue comes down to "what is the right balance to reach between incentives to miners and masternodes that will optimally provide sufficient security and sufficient number of MNs?" And that is the debate this community should be having. If the amount of hash power is sufficient to secure the coin AND provide for the number of MNs we need, then we have the answer. You could also debate what the right payout amount is (currently about 5 per block) to sufficiently incentivize both miners and MN owners, if the supply is insufficient.

One final thought.... increasing the payout ratio will also create demand for DRK (to start masternodes). This should increase the price of the coin and therefore could actually INCREASE the fiat value of the rewards being paid to miners (even if it is a lower absolute number of DRK). Therefore, we could increase the absolute amount of hashpower securing the coin while simultaneously increasing the number of MNs and achieve both goals and make both groups ultimately happy. Even if miners are better off, most will not see the correlation and will complain about their share. Some may even stop mining the coin. So be it. The mining world has an equilibrium supply of mining as well, and if that supply drops, it will leave the remaining suppliers more profitable. The "miners will all leave" argument is therefore a red herring and isn't warranted. Only if you argue that we end up with too few miners to secure the network is this truly valid.

I hope this helps frame the issues for everyone and helps with understanding why MN payments is THE key mechanism to grow the MN network.

Pretty sure ROI stays at approx 23%, ie 50% of original payout, but also 50% of original investment... Only difference should be increased server cost
 
Pretty sure ROI stays at approx 23%, ie 50% of original payout, but also 50% of original investment... Only difference should be increased server cost

Whatever the right ROI calculation is, we win nothing in terms of network security by reducing the collateral, is just the same, so it defeats the purpose.
 
I agree with all of the above, reducing the collateral is not really an option, but what could be great is to create a mechanism for trustless masternode pooling, as I described in my post above. I don't know if this is technically possible but if it was it would be a great solution. People with smaller stakes could pool together and create masternodes with 0 risk, this will take new coins out of circulation and have a positive effect on price. If at all possible, we could do this first and see where the count settles. Having said that, a raise in masternode payouts may still be necessary later on, but I think is a worthwhile initiative if at all possible.

I'll start thinking about a trustless mechanism like the cold/hot wallets.
 
Whatever we decide, we'd better do it fast, markets don't like uncertainty....according to BitcoinWisdom, someone just sold a Duff on Mintpal! :eek:
 
I agree with all of the above, reducing the collateral is not really an option, but what could be great is to create a mechanism for trustless masternode pooling, as I described in my post above. I don't know if this is technically possible but if it was it would be a great solution. People with smaller stakes could pool together and create masternodes with 0 risk, this will take new coins out of circulation and have a positive effect on price. If at all possible, we could do this first and see where the count settles. Having said that, a raise in masternode payouts may still be necessary later on, but I think is a worthwhile initiative if at all possible.
My two reactions to this argument...

1) I don't think it would have a huge effect on demand for MNs (though I could be wrong). My rationale is that $2,500 is not a lot of money... at least there are a sufficient number of people out there with $2,500-3,000 to allow the market to reach equilibrium already. If there happen to be a few people on the sidelines because that is too much money for them and they end up entering the market through a share in a MN and this drove the expected frequency of reward down for the remaining MNs, I suspect you would see a near equal number of MN operators exit their MNs (having no real effect).

2) A correlation in economics would be a stock split. Yes, it would make it so that the cost to get into the stock or MN would go down, but the payout is also lower. So when a stock splits, it might move up a little bit, but not much of an effect. You certainly aren't going to see us go from 800 to 2000 masternodes overnight. With stocks there is sometimes a gain of 5% or something, but you could argue that much of that gain is simply coming from the signals management is saying about their confidence in the stock by making such a change, not due to the "affordability" going down.

3) DRK has already been as high as what? $16,000 for a masternode? If your theory were correct, wouldn't we have seen a huge spike in the number of MNs already when the cost dropped to less than $3,000 today? It really hasn't moved much. This is further evidence that the MN network is acting as any economic asset with a return on investment would. The absolute cost is largely irrelevant.

So for me, the evidence is that trustless MN shares would be a lot of work without much, if any, impact on the number of MNs. It also introduces the complexity of how to make sure the trustless solutions deployed take the proper steps to secure the servers properly? So in that sense, you still have to "trust" the MN operator to do the job properly. It's not "trustless" in that sense.
 
My two reactions to this argument...

1) I don't think it would have a huge effect on demand for MNs (though I could be wrong). My rationale is that $2,500 is not a lot of money... at least there are a sufficient number of people out there with $2,500-3,000 to allow the market to reach equilibrium already. If there happen to be a few people on the sidelines because that is too much money for them and they end up entering the market through a share in a MN and this drove the expected frequency of reward down for the remaining MNs, I suspect you would see a near equal number of MN operators exit their MNs (having no real effect).

2) A correlation in economics would be a stock split. Yes, it would make it so that the cost to get into the stock or MN would go down, but the payout is also lower. So when a stock splits, it might move up a little bit, but not much of an effect. You certainly aren't going to see us go from 800 to 2000 masternodes overnight. With stocks there is sometimes a gain of 5% or something, but you could argue that much of that gain is simply coming from the signals management is saying about their confidence in the stock by making such a change, not due to the "affordability" going down.

3) DRK has already been as high as what? $16,000 for a masternode? If your theory were correct, wouldn't we have seen a huge spike in the number of MNs already when the cost dropped to less than $3,000 today? It really hasn't moved much. This is further evidence that the MN network is acting as any economic asset with a return on investment would. The absolute cost is largely irrelevant.

So for me, the evidence is that trustless MN shares would be a lot of work without much, if any, impact on the number of MNs. It also introduces the complexity of how to make sure the trustless solutions deployed take the proper steps to secure the servers properly? So in that sense, you still have to "trust" the MN operator to do the job properly. It's not "trustless" in that sense.
In remote wallet setup proper server security doesn't really matter because coins are not moved to server wallet - they stay secured in your local (cold) wallet
 
My two reactions to this argument...

1) I don't think it would have a huge effect on demand for MNs (though I could be wrong). My rationale is that $2,500 is not a lot of money... at least there are a sufficient number of people out there with $2,500-3,000 to allow the market to reach equilibrium already. If there happen to be a few people on the sidelines because that is too much money for them and they end up entering the market through a share in a MN and this drove the expected frequency of reward down for the remaining MNs, I suspect you would see a near equal number of MN operators exit their MNs (having no real effect).

2) A correlation in economics would be a stock split. Yes, it would make it so that the cost to get into the stock or MN would go down, but the payout is also lower. So when a stock splits, it might move up a little bit, but not much of an effect. You certainly aren't going to see us go from 800 to 2000 masternodes overnight. With stocks there is sometimes a gain of 5% or something, but you could argue that much of that gain is simply coming from the signals management is saying about their confidence in the stock by making such a change, not due to the "affordability" going down.

3) DRK has already been as high as what? $16,000 for a masternode? If your theory were correct, wouldn't we have seen a huge spike in the number of MNs already when the cost dropped to less than $3,000 today? It really hasn't moved much. This is further evidence that the MN network is acting as any economic asset with a return on investment would. The absolute cost is largely irrelevant.

So for me, the evidence is that trustless MN shares would be a lot of work without much, if any, impact on the number of MNs. It also introduces the complexity of how to make sure the trustless solutions deployed take the proper steps to secure the servers properly? So in that sense, you still have to "trust" the MN operator to do the job properly. It's not "trustless" in that sense.
I agree. Anyone who wants to can pool with others and put up a Masternode - I run two such. But at the end of the day my partners are trusting me not to screw it up, and would be doing so even if we were using a multisig address to ensure I didn't run off with their coins, which would be a royal pain in the arse if any coin-moving maintenance (even just monthly payouts) was required. Sometimes a little trust can be a good thing - it builds communities and forges bonds. So, I too don't think multi-owner MNs would be worth the effort to try and implement within the core functionality of the coin.

If Masternodes need further incentivising, it either comes from the block reward, or from some other service that Masternodes provide, for example a small fee for InstantX?
 
My two reactions to this argument...

1) I don't think it would have a huge effect on demand for MNs (though I could be wrong). My rationale is that $2,500 is not a lot of money... at least there are a sufficient number of people out there with $2,500-3,000 to allow the market to reach equilibrium already. If there happen to be a few people on the sidelines because that is too much money for them and they end up entering the market through a share in a MN and this drove the expected frequency of reward down for the remaining MNs, I suspect you would see a near equal number of MN operators exit their MNs (having no real effect).

2) A correlation in economics would be a stock split. Yes, it would make it so that the cost to get into the stock or MN would go down, but the payout is also lower. So when a stock splits, it might move up a little bit, but not much of an effect. You certainly aren't going to see us go from 800 to 2000 masternodes overnight. With stocks there is sometimes a gain of 5% or something, but you could argue that much of that gain is simply coming from the signals management is saying about their confidence in the stock by making such a change, not due to the "affordability" going down.

3) DRK has already been as high as what? $16,000 for a masternode? If your theory were correct, wouldn't we have seen a huge spike in the number of MNs already when the cost dropped to less than $3,000 today? It really hasn't moved much. This is further evidence that the MN network is acting as any economic asset with a return on investment would. The absolute cost is largely irrelevant.

So for me, the evidence is that trustless MN shares would be a lot of work without much, if any, impact on the number of MNs. It also introduces the complexity of how to make sure the trustless solutions deployed take the proper steps to secure the servers properly? So in that sense, you still have to "trust" the MN operator to do the job properly. It's not "trustless" in that sense.

I disagree with you in the following sense, creating trustless masternode pooling and raising the masternode reward are not mutually exclusive. You could do both if it was necessary, but trustless masternode pooling would have a very positive political effect by opening participation to people with smaller stakes, while increasing the network security. So I think this should be done first to see where the count settles.

I do believe the masternode count would go up substantially, as the rationale for running masternodes at this stage of the game is more about supporting the coin than realizing real returns, I havent cashed out a dime from my masternode profits yet. Plus, people are already pooling for masternodes, so a nice portion of the masternode count we already have comes from pooled masternodes, they are just not trustless and some people could even get hurt.

We are missing DRK supporters that hold less than 1000DRK and would like the opportunity to support the network and probably leave their coins trading in exchanges.

Finally, your argument that with an increased masternode count the reward would go down is still true if the count just goes up right now for no reason without introducing the trustless masternode pooling, so like I said increased reward may still be necessary later on, but trustless masternode pooling should come first, then we see where the count settles and last we play with the reward if necessary since that is politically way more delicate.
 
Last edited by a moderator:
My two reactions to this argument...

1) I don't think it would have a huge effect on demand for MNs (though I could be wrong). My rationale is that $2,500 is not a lot of money... at least there are a sufficient number of people out there with $2,500-3,000 to allow the market to reach equilibrium already. If there happen to be a few people on the sidelines because that is too much money for them and they end up entering the market through a share in a MN and this drove the expected frequency of reward down for the remaining MNs, I suspect you would see a near equal number of MN operators exit their MNs (having no real effect).

2) A correlation in economics would be a stock split. Yes, it would make it so that the cost to get into the stock or MN would go down, but the payout is also lower. So when a stock splits, it might move up a little bit, but not much of an effect. You certainly aren't going to see us go from 800 to 2000 masternodes overnight. With stocks there is sometimes a gain of 5% or something, but you could argue that much of that gain is simply coming from the signals management is saying about their confidence in the stock by making such a change, not due to the "affordability" going down.

3) DRK has already been as high as what? $16,000 for a masternode? If your theory were correct, wouldn't we have seen a huge spike in the number of MNs already when the cost dropped to less than $3,000 today? It really hasn't moved much. This is further evidence that the MN network is acting as any economic asset with a return on investment would. The absolute cost is largely irrelevant.

So for me, the evidence is that trustless MN shares would be a lot of work without much, if any, impact on the number of MNs. It also introduces the complexity of how to make sure the trustless solutions deployed take the proper steps to secure the servers properly? So in that sense, you still have to "trust" the MN operator to do the job properly. It's not "trustless" in that sense.

Using "cold" wallets would be enough to protect the "shareholders," just as is currently done with Masternodes
 
I agree. Anyone who wants to can pool with others and put up a Masternode - I run two such. But at the end of the day my partners are trusting me not to screw it up, and would be doing so even if we were using a multisig address to ensure I didn't run off with their coins, which would be a royal pain in the arse if any coin-moving maintenance (even just monthly payouts) was required. Sometimes a little trust can be a good thing - it builds communities and forges bonds. So, I too don't think multi-owner MNs would be worth the effort to try and implement within the core functionality of the coin.

If Masternodes need further incentivising, it either comes from the block reward, or from some other service that Masternodes provide, for example a small fee for InstantX?

Trusting someone with managing the masternode server is trivial, trusting them with your coins is not. A lot more people would pool for masternodes if it was trustless that is a reality, rewards can be changed after that if it is still necessary.
 
Pretty sure ROI stays at approx 23%, ie 50% of original payout, but also 50% of original investment... Only difference should be increased server cost
I stand corrected, and apologize for my rush without checking the facts.

It can be calculated with: ((1/b)*c*d*e)/p = ROI per year
b is the total amount of Masternodes
c is the amount of blocks per day
d is the days in a year
e is payment per block won
p is the price per MN

So with that in mind the current profitability can be calculated by

((1/900)*576*365*(5*0.20))/1000 = 0.23 (23% per year ROI in DRK)

The scenario where it drops to 500 and twice as many MNs would be:

((1/1,800)*576*365*(5*0.20))/500 = 0.23


I really don't like being wrong, but at least I'll own up to it. Thanks for the correction. My other points remain, though. Increased cost would drive some out of the market and reduce incentives to properly secure the servers and provide adequate resources (e.g., RAM, CPU, etc) to properly support the network.
 
Back
Top