Topic: All coins exept bitcoin and scrypt coins are vulnerable (Read 2568 times)

mr_random, like everyone else you're clueless about how the ppcoin system works. You don't need enormous number of coins for 51% doublespend, you just need to wait a couple of months until your chance to find stake is as good as anyone else, and combine it with much less than 50% hashpower. To prove that you're clueless, I offer you 100,000 ppcoins if you can describe here the ppcoin protocol (feel free to copy and paste from any source you'd like, though you won't find anything).

Not necessarily. There are two issues here:

Issue 1. 51% computational attack. This can be defended by either: very strong network with many miners leading to a high network hashrate OR some proof of stake support which makes such an attack useless OR a hash algorithm which is expensive to compute (litecoins Scrypt).

Issue 2. ASICs jumping on board to mine, rasing the difficulty to an absurdly high level and then leaving the network. This can be defended by either: very strong network with many miners leading to a high network hashrate OR a difficulty adjustment algorithm that is smart enough to cope with a sharp increase and decrease in network hashpower without leading to wild difficulty swings (e.g. PPCoin's exponential smoothing)

Lol no you don't understand. Re-read my quote. A computational 51% attack isn't possible on PPCoin. That means if someone gains >51% of the hash rate power they can't do malicious things like double spending. A proof of work computational attack alone won't do anything. This is because a hybrid of proof of work and proof of stake is used. Therefore an ASIC can't launch a 51% computational attack.

Instead a person needs to attack PPCoin's proof of stake by accruing an enormous number of coins ie a 51% proof of stake attack. Whereby someone needs 51% of the coins to have a decent chance of attacking to double spend. Good luck with that.

I wasn't talking about scrypt, but as an aside: do those 10 supercomputers have more hashing power than all PCs owned by individuals combined?

If you look around, you won't see anyone on this forum that owns a top 10 supercomputer.

Pretty much why I'm alternating between mining Litecoins and Mincoins at the moment, with the occasional spot of PPcoin mining here and there.

Nonsense, executing 51% on ppcoin is a lot easier, you just need to wait 30 days and combine the stake that you collected with your hashpower for the attack. There is no security proof for ppcoin like there is for bitcoin, the ppcoin article contains just gibberish if you actually try to read it. The people who mined 300,000 ppcoins in the first couple of days will dump everything when they see that there are no more fools who believe the fantasies that they tell you.

then i guess all coins should implement PoS? or is it too late?

This is simply PPCoin's difficulty adjustment algorithm, it's independent of POS afaik... there is nothing stopping TRC from implementing it. PPCoin has shown so far that it's difficulty adjustment algorithm copes very well with ASICs or big increases in mining power. You'll notice a lot of people were telling the TRC devs to incorporate PPC's difficulty adjustment but it fell on deaf ears.

Proof of stake (POS) is what protects against a computational 51% attack.

I'm not sure that's true if mining is based on an algorithm that cannot be done spectacularly faster or with spectacularly less energy consumption on ASICs than on CPUs.

What would happen with ppc ?

1. i point my popular btc pool to ppc for a while,
2. difficulty starts skyrocketting
3. miners leave

4. repeat 1+2 when needed

how long would it takes to POS to bring difficulty down to a reasonable level back ? Isn't POS efficiency related to the number of running nodes and their coin age ?

It's price has gone from 2 cents a coin to 14.5 cents a coin in the space of 3 weeks. It's also moved up to number 3 for market capitalisation, ranked only behind Bitcoin and Litecoin. Haters gonna hate.


This is a good point. I had never read of this 'attack' before until I saw it happen with TRC.

TRC has shown that there is an additional computational vector for attack: A difficulty based attack. You can make the chain diff level so high that it slows the rate of production of blocks to the point where the network is at a standstill. Could happen for Bitcoin although you'd need a crazy powerful system to do it.

It's price has not, though.

PPCoin is immune to a computational 51% attack. This is because it uses proof of stake.

And PPCoin's intelligent difficulty adjustment has coped just fine with ASICs hopping on board.

Yes and no.

While scrypt could be vulnerable to anyone with sufficient hash power to do so, it's relatively rare for people to have such significant concentrated hash power in the form of a GPU farm compared to all the other GPU miners out there. Bitcoin (Or any SHA2 based coin) is more vulnerable because of ASICs, which a significant portion of the population don't have access to.

At FRC we have seen a good deal of 'chain-hopping' and while their seems to be no malicious intent it is annoying.  That said I disagree that the pattern is for fluctuations to grow, were seeing them get smaller with time.  The first was something like 97% drop off, the second around 80% and the latest looks to be 50% (so far).

TRC seems to have suffered increasingly severe fluctuations mainly due to errors in their patches which were meant to dig them out of a difficulty hole.  They over did it and as the saying goes, when from the frying pan and into the fire.  I don't think it's correct to extrapolate their experience to all coins.

Also we at FRC don't think any of our fluctuations were caused by ASICs (though we wisely have a member with one who's kept in reserve to buoy up the speed in just this situation).  All indications are they are the normal GPU miners who are fairly small to mid scale as we see the Cointron pool mining most blocks and they list their members, none are remotely large enough to prove they are running ASICs.

Lastly your conclusion is flawed because it ignores the ability of GPUs to perform Scrypt meaning they can jump between both hash-types their are still plenty of those machines and plenty of owners looking for places to squeeze some profit out of them before they are rendered permanently useless for BTC mining.

Scrypt is irrelevant here— they're just as vulnerable to people bouncing between coins.

Merged mining was created to solve this.

2 main things make non scrypt coins vulnurable are:
1)Difficulty swings,  a pattern starts when
 difficulty goes up-> makes it less profitable than bitcoin to mine-> people mine something else->Longer time to mine blocks-> next retarget lower difficulty->makes it more profitable to mine-> miners switch back-> difficulty ect
every time it gets a more pronounced difficulty
This may have already claimed its first victim with terracoin and some are concerned bytecoin and freicoin might be going down the same road
2)51% attacks (asic style) now with any new startup coin that runs off bitcoins algoritm all u need to do is poin a few asics at it and 51% attack it and boom dead. If you get a group with 100 asics it might pose threats to other coins

So if u are holding alt currencies , and dont want ur coin killed MINE IT CONSISTENTLY, it will help with both problems