Topic: HoboNickels - HBN - High Fast Stake - Version 2.0! More Secure, Less Intensive - page 251. (Read 478852 times)

Added a new page to the wiki!  http://wiki.hobonickels.info/index.php?title=History_of_HoboNickels

Good conversation here guys.

Just to keep expectations in line,  I can't respond to each reply and tit for tat. But I do read them and keep them in mind.

Also I do not have any immediate plans to change how this is currently working.  This is not a PoW coin it is PoW/PoS Hybrid. PoW is meant to start the coin off, with the PoS carrying the weight in the future.  Its not perfect, but it is working as intended.  There are many coins that are much more profitable to mine and dump. HBN PoW will get less and less profitable as time goes on.

The only thing I may do is put an optional pause time between PoS blocks. It will not be enforced by the network and will be easily turned off if you want to. Still not sure I will even do this.

If there are serious issues I will issue a hard fork. This is not serious it is a minor nuisance. And as the network continues to mature, this will become less and less of an issue.

If other people want to try out code changes, and submit them we can test them out, and see how they work.


Thanks all!

That is an interesting idea. That might be able to work, but I think we really need Tranz to weigh in.  Center to your whole idea is that PoW takes priority to PoS.  I am not sure I like that part of it.

The network/wallet just has to "reject" odd-numbered blocks by POS, and the wallet simply has to stop attempting to mine POS when it has detected an "Even block" has just been accepted. Thus, it would not be attempting to mine the odd one that would follow. (You would not reject POW at any point in time... if POW beats POS, it deserves that block, over a POS.)

Both would have to happen, or someone would just make a wallet that did mine POS on even and odd, if the network didn't also reject those, in addition to not attempting to waste time mining POS.

Froze my computer a couple of times, when it went nuts mining POS too... that should be throttled better. (More of a casual hunt for POS, not a full-on race for POS hunting. That too could be limited to force POS only for every other wallet... If you found one, you have to wait for the next one (even one) to be found by anyone-else, before you attempt to submit your next POS. Creates more even distribution too.)

EG...
e POW {even}
o POW {odd}
e POW {No POS ready yet}
o POW
e POS (you) {POS due, you got it}
o POW
e POS (someone-else)
o POW
e POS (you)
o POW
e POW {Beat POS to the block, it deserves it, this was actual work}
o POW
e POS (someone-else) {not you, you got the last POS}
...

Since three wallets/daemons are always needed for the network to live, this would always work. (1/4 might be spotty, unless you force 3+1 connections as the minimum for the network to live. Spotty, because it will ultimately round-robin the POS, with only 3, on every other even block. I might be playing that out wrong in my head...)

Can't get any more simple than that.

However, it would require a hard-fork, as the prior blocks before that point would have to be "accepted", for what they were, even or odd, in code. (Otherwise the network would reject the whole past chain, for validation, prior to building before that point, for any new wallet creation.)

That would also appease the diff-adjustment. So POS does not wrongfully bump-up POW diff, since it does not actually contain the x-coins reward, or may actually contain x-coins more than the reward.

Now if it forced holding more coins, before creating a POS block... That would be even better. (Instead of accepting super-small fresh-mined blocks.) Same if it had a maximum limit... You have two blocks of 100,000 coins, you get instant super reward, over a guy who has thousands of 1-20 coin blocks, which forces the system to create TONS of POS, while the other guy only creates one. But that is a whole complex set of programming. The other thing is immediately rewarding, and easy to do in code.

Not everyone has time to baby-sit miners. (Nor should that be a requirement, to mine, so you don't lose-out.)

You think that was a big swarm... Wait until there are thousands of open wallets POS mining. (Again, that isn't making the network use any less power. It makes it use more, and more wasteful power.)

It is great that it instantly pushes the hoppers off the coin, but that still leaves teh dedicated miners holding the short stick, 100% of the time.

This would reduce the losses (of unattended dedicated miners) by 50% immediately. (Hoppers will still leave immediately still... but our short-stick gets a little longer.)

Also, even if the entire network gets super-saturated with "due POS rewards" from millions of small-block rewards, the damage will never be more than half network stalling of POW reward/loss.

I have no anger of how this works right now either... But I find your suggestions interesting. I would love coin devs and knowledgable people to comment on this.  There's always space for improvment and if it can contribute to secure the coin value and network, why not debate it?

And as presstab said, it might not be 100% efficient the way it works now but for sure, people checking in on their mining will switch to other coins that rack in more profits. Who likes to pay the electricity bill just for the fun of it

I mine HBN with about 5 mhs.  I simply switched to another coin while the diff spiked, and came back after.  I know of other miners with much higher hash rates that did the same.  The network hash rate went down that day because attentive miners check in frequently, or use cgwatcher which can switch profiles automatically.

Anyways, I have absolutely no anger towards the coin or the way PoS works. The way I look at it is that I mine the coin and the coin mines itself after, so it is actually increasing my total hashrate.  

You want to save this coin...

Force POS to only happen/accept on "Even" block numbers. That allows for POW between every POS, and that WILL give you some actual security. Until the last coin is mined, then, and only then, should POS be allowed to run on every block.

Even better, only allow POS on "every other", "Even block". (1/4)

That will limit the distribution to be more evenly also, as one CPU can't just build upon its own blocks which have not even had time to distribute through the network. (That alone, decreases security, because one POS puke could all come from one fast CPU and fork the whole chain.)

Not if you are a miner...

Hard trying to convince people to mine, when they constantly get told... "It's a good thing that you worked for nothing!"

That didn't save any energy... it just made a bunch of miners mine for zero reward, making them foot the bill while POS booted the work they were doing. You just added more energy-waste from CPU mining, in addition to the unrewarded GPU miners. Where did you learn to do math?

When they leave, they will cash-out and take what little value was left in the coin with them.

That was just a small swarm of POS blocks... Imagine what happens when people actually use this coin!

You don't even have a mechanism in place to alert them that they are wasting power or time mining blocks that can never compete with POS, so they could... Um, mine something with actual value instead of just burning electricity, racking-up the electric bill for your entertainment. All so someone who is not even contributing to the network can get 2% reward, which is still more than the 0% reward the actual workers/miners are getting.

POS was a neat idea that was poorly executed, and photocopied and made even worse. It secures nothing. Just fills the chain with hyper-blocks of crap.

That's why everyone left bottlecaps. (That, and poor support)

I'd say around 50.  I'm getting 56 at 1.3MH.

Anyone got an approximate quote on coins per hour/day at around 1,000 kh/s for the current difficulty?

Thundertoe?

Hi all!

Fun day today.  

Nothing is wrong with the network, or with the hash rate calculation or the difficulty. All were acting within normal parameters.  The client tries to keep the block time generation to 30 seconds, across PoS AND PoW

In this case a user how a bunch of aged coins, that they turned on after 30+ days(max weight), and it started generating PoS blocks very quickly. The network responded by increasing the difficulty of both PoS and PoW to slow down the block back to 30 seconds.  Preference is given to PoS over PoW, as this was the intention to create a more efficient network. And what happened? Miners left the network, saving electricity and allowing PoS to carry more of the block chain.

This is a good thing. We want a very energy efficient network, and this is what we are getting!

We got hoboed lol.

Will have to wait another 5 hours for it become normal again. Maybe we will go back to normal diff of sub 1

Seems it takes very long for diff to come down. Currently 5.8 at 210 MH/s.

According to my munin graph, it's ending from what I understand, so it should be going down and down

https://s.gks.gs/img/img/02-2014/bitcoin_difficulty_week_1_.png

It started at 10 (GMT+1) this morning, and "finished" around 17. So I will guess around midnight it should be normal.
A lot of miners seems to have left, so the diff should be even lower than before.

do you know how long this will continue for?

From looking at the chain it looks like a bunch of PoS blocks are being generated. Some only ten seconds apart.  My guess is that there is a large amount trying to stake right now so it is crowding out PoW. 

looks like miners are withdrawing till PoW diff issue is sorted
time per block now 2 min 15s and hashrate dropped to 195 mh/s

I wish I knew. Anyways i guess it is cheaper to buy this coin after all. Mine something else instead.

at current 215 mh/s diff should've been about 1.5 (currently @ 6.2) - 2 min time per block vs 30s target
concerned about how long it's going to take for PoW diff to settle down