111* days estimated for Difficulty to retarget. Could this kill PTS?

[maqifrnswa]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

You can break up that 500 PTS and send out as transaction fees. That alone should be enough to tide over the problem; not to mention will also generate a buzz.

With only 20% left to mine, it doesn't seem worth the trouble to hard fork it. Even if you do, if you are just paying to implement KGW, then 500 PTS sounds a lot (though the challenges may be different 'coz this is not just any scrypt coin, and so maybe worth it).

there are two economic problems with fixing this with fees alone:
1) retarget occurs every ~2000 blocks, which now is ~3500PTS in mining revenue. Current hashrate is down to <50% of what it was previously. However, offering 500 PTS will only increase payouts by ~14%. Therefore, you will need a lot more than 500 PTS to entice miners back.
2) Once you've enticed miners back, you've artificially decreased the time to retarget, which means that the drop in difficulty won't match what the market thinks it should be. You will need to do a second bailout, most likely. See below edit. Actually, these bailouts can help PTS find the correct market price faster than if you let it oscillate.

the biggest bang for the buck would be patching the PTS code to prevent the timewarp attack, and then to consider the cost benefit analysis of a hardfork and compare it to how much it would really cost to bring down PTS difficulty with bailouts.

EDIT: there is an additional danger to doing nothing, and a benefit to paying PTS fees out to stabilize mining. Once difficulty drops down, miners will come back fast at the low difficulty, and mine the 2100 rapidly. Then they will turn off once retarget happens again, and this will continue (most likely it will be a stable oscillation). Paying PTS transaction fees, in this case, will dampen the oscillation and actually help the mining market stabilize.

[maqifrnswa]

another interesting academic question... PTS is different than any other altcoin. If every miner in the world suddenly stops mining PTS, it still has value. PTS, at that point, has become just like AGS. An untradeable asset that represents a share in future I3 DACs.

If every miner in the world stops mining litecoin, it becomes worthless. PTS is different.

not sure if the time travel patch was implemented. if it wasnt, i will gladly prove you wrong by attcking the chain, simply reversing to the checkpoint block and branching off with static difficulty applied for blocks up till the very latest, then i raise the difficulty and pass the current block number at a raised diificulty, continued mining pushes me a hundred blocks ahead, then poof...your wallet is empty an i have the longest chain.

even if the fix is applied later, the sheer amount of dumping i would have done will irreversibly devalue PTS. ask TRC holders, they thought they were safe, and they were tethered to the BTC chain, yet they still got a huge kick in the ***

thanks, that's a well described deficiency beyond the simple idea of "a dead chain is dead"

[bitbadger]

If KGW is flawed, then just change the adjustment period to some lower number of blocks.  It's not exactly hard.  Ideally, the block retarget period would be specified in only one place in the code. 

As an alternative, the coin MazaCoin (MZC) launched with a 4-block retarget, based on the last 90 blocks.  It seems that this would be easier to implement and possibly without the KGW weakness, which seems to be based on the fact that KGW retargets every block, and is based on a smaller number of recent blocks (the most recent 11 blocks, IIRC?).

On the other hand, it seems that the KGW flaw is likely more exploitable in coins with fast block times.  There are lots of scrypt alts employing KGW that have a target block time of 120 or 60 seconds.  So time-malleability (the fact that a later block can have an earlier timestamp) can come into play a lot easier than with a coin with longer block times.

[sumantso]

I still say put in big transaction fees and that will bring miners in. May also add some publicity.

[Schwede65]

hm, there seem to be problems with KGW

DRK-thread:
https://bitcointalk.org/index.php?topic=421615.8580

Cryddit wrote there:
"The kgw exploit is real.

The problem is that the software will accept blocks that are timestamped earlier than previous blocks, (for good reasons) up to the median timestamp of the previous 11. 

Kgw counts these as very fast blocks and adjusts the difficulty up by its 20% maximum.  Once.   But you can use that once to roll the timestamp back after five blocks that are long enough to drive the difficulty 20%lower each. 

So the exploit is to start mining your own block chain, time stamping several blocks into the future far enough to get the maximum downward adjustment of difficulty each time.  Then jump backwards in time getting the maximum upward difficulty once.  Rinse, repeat,  and your chain gets ridiculous low difficulty and you mine a whole lot of blocks while advancing your timestamp not much further than the main chain.

When the time in the main block chain catches up, you release your own chain and force a reorganization, which allocates to you all the coins mined for the last umpty ump blocks."

the DRK-dev eduffield is "cleaning" KGW for DRK: DarkGravity

please have a look after this while implementing KGW in PTS, maybe DarkGravity works better

[jsidhu]

Is there nodes running the testnet? Ip?

Or do I have to do a test in a box approach.. creating a new testnet?

Also give me a list of good nodes on live.. preferabble dns.. Ill update the dnsseed list so client connects without add adding nodes manually. The ann thread has some nodes but they dont connect. This is unacceptable.

[barwizi]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

I offer the fix with checkpoint using kimoto gravity well implementation.. github.com/sidhujag/PTS

But testz said bounty isnt valid and to test it myself.

Are you going back on your word here or whats going on?

Here is my answer:

Sorry, but I don't think we will going to fork the network, I will ask bytemaster, or you can open the poll in the forum about this.

Regards,
TestZ

PS: Test net works if you want, you can check it by your self.

As you can see, I dont say that "bounty isnt valid".
Yes, BitShares-PTS has working testnet and you can check your Kimoto Gravity implementation.

yes, test it first.

[NewMine]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

I offer the fix with checkpoint using kimoto gravity well implementation.. github.com/sidhujag/PTS

But testz said bounty isnt valid and to test it myself.

Are you going back on your word here or whats going on?

Here is my answer:

Sorry, but I don't think we will going to fork the network, I will ask bytemaster, or you can open the poll in the forum about this.

Regards,
TestZ

PS: Test net works if you want, you can check it by your self.

As you can see, I dont say that "bounty isnt valid".
Yes, BitShares-PTS has working testnet and you can check your Kimoto Gravity implementation.

[testz]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

I offer the fix with checkpoint using kimoto gravity well implementation.. github.com/sidhujag/PTS

But testz said bounty isnt valid and to test it myself.

Are you going back on your word here or whats going on?

Here is my answer:

Sorry, but I don't think we will going to fork the network, I will ask bytemaster, or you can open the poll in the forum about this.

Regards,
TestZ

PS: Test net works if you want, you can check it by your self.

As you can see, I dont say that "bounty isnt valid".
Yes, BitShares-PTS has working testnet and you can check your Kimoto Gravity implementation.

[NewMine]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

I offer the fix with checkpoint using kimoto gravity well implementation.. github.com/sidhujag/PTS

But testz said bounty isnt valid and to test it myself.

Are you going back on your word here or whats going on?

[jsidhu]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

I offer the fix with checkpoint using kimoto gravity well implementation.. github.com/sidhujag/PTS

But testz said bounty isnt valid and to test it myself.

Are you going back on your word here or whats going on?

[barwizi]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

i'll put a pull req for the checkpoint.

[Hukkel]

The lack of idea of what a share is and how to keep investors happy is baffling me.

Wether you think so or not, you are not operating in the world of stocks. You are operating in the world of cryptocurrencies.
Profitabilities are high and short term.

Barwizi seems to be the only one understanding this.
BTS - protoshares are your base shares. The entire operation depends on the PTS and its value. Wether you believe this or not. The complete desinterest in PTS will be a very big problem. Don't think that by releasing another DAC will make it another cashcow for a couple of weeks. You haven't show anything yet, BTS-X is not even here yet. You are losing the attention of investors very quickly.

[jsidhu]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

You can break up that 500 PTS and send out as transaction fees. That alone should be enough to tide over the problem; not to mention will also generate a buzz.

With only 20% left to mine, it doesn't seem worth the trouble to hard fork it. Even if you do, if you are just paying to implement KGW, then 500 PTS sounds a lot (though the challenges may be different 'coz this is not just any scrypt coin, and so maybe worth it).

By looks of it it can get exponentially harder to solve a block as people stop mining, especially around the retarget time... so may not be right to think linearly that only 20% is left when on the timescale of PTS lifetime it may seem alot longer.

[sumantso]

Ok... I will offer a 500 PTS bounty for the best solution to this problem.   Testz, as the maintainer, of PTS right now will be responsible for reviewing and applying it and producing a release.

In the mean time, we should have an updated checkpoint ASAP to minimize any damage from potential attacks.

You can break up that 500 PTS and send out as transaction fees. That alone should be enough to tide over the problem; not to mention will also generate a buzz.

With only 20% left to mine, it doesn't seem worth the trouble to hard fork it. Even if you do, if you are just paying to implement KGW, then 500 PTS sounds a lot (though the challenges may be different 'coz this is not just any scrypt coin, and so maybe worth it).

A hard fork is a lot of effort & support to do RIGHT.

As I said, just publicise and send out big transaction fees instead.