simply buying gold is not an investment, it's speculation

real estate may be an investment if your rent it or do something with the real estate that gives returns, otherwise it's also speculation or another form of saving

in general, it's better to invest if you can afford the risk

"undoubtedly" "more secure"? it's not even a consensus design since the method for deciding consensus (which chain is preferred when a fork is presented) is not explained!     EDIT: but it undoubtedly is interesting, agreed on that

this was not yet answered:

first tell me the decision method, then we can discuss if it is more or less secure. Myriad looks more secure in some aspects, but it doesn't have POS. Those hybrid altcoins that have POS are not undoubtedly more secure.


Anybody knows of a working implementation of "Follow the satoshi"? in many cases the owner of the satoshi will not be online... I wonder how they deal with that... and how do they make sure the randomization is not manipulated.

Yes, the problem of the mod 4 idea is that miners are idle until it's their turn again. An idle miner consumes less than a working miner, but it sill consumes a lot of power. I think that if can't find a way around this then it's a show-stopper.

Letting all the 4 "channels" run in parallel raises 2 immediate questions:

- When is a block confirmed? 6 pos blocks is not as secure as 4 blocks from different methods. This can be solved saying "1 confirmation is whenever you have 1 block from each method". So 1 confirmation requires at least 4 blocks, and 2 confirmations require at least 8 blocks, but can take many more blocks in case of bad luck.

- How to weigh two competing chains for a reorg? Imagine two chains that have the same length and the same blocks up to height n-1, but block n is method A in one chain and method B in the other one. Which one has more work (or weight, or "trust")? Since different methods have different unrelated difficulties it's not easy to compare. Myriad solved this using factors resulting from empirical tests: scrypt is assumed to be 1000 times as hard as double-sha256, so now the work for different algorithms can be compared. But how do you compare POS vs POW? most hybrid coins don't: they just allow POS to rewrite POW blocks by assigning a constant minimal "trust" to POW blocks.
If you want to avoid someone from rewriting history you can assign exponentially decreasing work to consecutive blocks from the same method. Example: after a SHA3 block, if the following block is also SHA3 then you multiply it's work by 0.5 and so on until a block from another method is found.

wait... you have that same problem with the mod 4 idea: if there are 2 chains that only differ in the last 4 blocks but one chain has more difficulty in the mod 0 block and at the same time less difficulty in the mod 1 block. Which chain is the best?

Oh... I see it now, what I wrote there is not correct... you say that you can get to any % you want as long as you keep growing exponentially... I wouldn't worry about that though...

mmm, I don't think so... he gives asymptotical bounds for the memory used by the sieve

Hi!


Other people made better pool servers and miners and I merged a couple of pull requests from others, but I'm still the main developer of Riecoin Core.


Yes, I still have gmail and it's still riecoinorg.
My @riecoin.org e-mail auto forwards to my gmail

I was referring to another work from the same author. The same man who worked on the "the sum of three primes" problem has recently been working on a method for making sieves that uses less RAM: that's the work that may (or may not) be helpful, but I still couldn't find any links to an actual paper.

Hi, you can create a pull request on github or send me the source of the affected file and I'll merge it. Thank you.

you may want to read about replace by fee

http://bitcoin.stackexchange.com/questions/10733/what-is-replace-by-fee



Also, the average time per block is not 10 minutes. It would be if difficulty were constant. But difficulty is growing because the average is less than 10. Somewhere between 8 and 9 minutes (which is still far from what Itskok suggested)

but the chances of overtaking get lower as time goes by, so it will never get to 50% if you have less hashpower than the network

Hello beautiful people!

I just read about this: http://www.scientificamerican.com/article/new-take-on-an-ancient-method-improves-way-to-find-prime-numbers/
It's supposed to be a new algorithm that improves Eratosthenes' sieve, so maybe it could be used to optimize Riecoin mining. Does anyone have a link to an actual paper about this? all I could find was just the abstract...

cheers and keep mining!

gatra

updated the OP to reflect this

Just to be clear, for practical purposes what knightdk says is true. But from a cryptoanalysis point of view, it's slightly more likely to have a collision.

Another idea: if the NSA has a backdoor for sha256, then it is possible that backdoor doesn't work for double sha256.

This is not exactly true, the collision resistance is weaker than the weakest.

Hashing many times protects from some preimage attacks. It's sometimes used to make the hashing slower on purpose (like in WPA, in order to make cracking slower and more expensive). Doing it twice for mining makes sense because it prevents some mining algorithm optimizations and also because ASICs for sha256 existed before bitcoin, but not ASICS for double sha256. So hashing  twice may have been a way to prevent those ASICS from working.

Hashing more than once does increase the probability of collisions, however that increase is negligible. Consider the case of double sha256:
output = sha256(sha256(input))

If one of the two sha256 have a collision, then the double sha will have a collision too. So it's weaker: at least one of the two hashes is required sufficient for the double sha to collide. That probability of having one of two hashes collide is certainly higher than having only one hash collide.

However, consider that if double sha256 has a collision it is because at least one of the two iterations of sha256 did collide. So you see, it is still very unlikely because it still requires a collision in sha256. And as someone mentioned, a collision in sha256 is in itself more important news than having found Satoshi.

I consider it safe to assume that CSW didn't find a collision in sha256.

trying to get RIC up to date with changes introduced to BTC up to 0.12.1 ..... doing it right is a lot of painstaking work... too many changes, too many decisions...

Well, leaving aside my doubts about PoS' security, it is true that implementing PoS between super-super-blocks would work. However the point of RIC was to prove that PoW could be improved to get some useful results, so I'm more inclined towards a PoW-only solution.

The length of the primes ("difficulty" is the length of the base prime in binary digits) is adjusted so that we find one block (each block has one sextuplet) every 2.5 minutes. There is some variance, but we adjust the difficulty a couple of times a day so we don't get far from the 2.5 min target.
Once a week we work on a sextuplet that is about 30 times harder so it takes us 70 minutes (in average of course), and we call that a superblock. So we have a superblock once every 4032 blocks.

The numbers being a multiple of 173# is just a result of how the (mining) software makes the sieve and is not mandated by the protocol.

I read coinkite had to close their wallet service because of too much ddos: constant attacks over three years
it's insane!

Yes, but all our base primes start with 10000000 in binary so I'd round it up instead of down: 3445 just to be safe


Agreed. 50% higher in number of digits means more than ten times of actual work because of the exponential nature of this stuff. One order of magnitude higher, however still doable.

I'm trying to find a way of keeping frequent blocks instead of super-blocks, because some traders rely on the speed of the coin for arbitrage. I don't want to hurt them with transactions that take many hours.

but... challenge accepted! we are going to have to break it

I was thinking about the differences between Riecoin mining vs just finding a sextuplet. In Riecoin we are forcing the the first digits of the base prime to match the hash of the block, so the first 250 or so binary digits are "fixed" and change about once every 2.5 minutes. This means we have to make a new sieve each time and gives us a "disadvantage", in the sense that there is a loss in efficiency.
But compare our efforts: in two years, we found many sextuplets in the range of 500 digits. This is far from the 1037 of the new record, but hey! we found 400000 of those! We have much more computing power than 18 phenoms.

If we could focus work on larger numbers without impacting the coin's transactions we could reach 1037 digits in weeks intead of years. I've been working in this direction and I'll put my ideas in written as soon as I understand them
Meanwhile, we are getting the attention of more people insterested in Maths making them look into cryptocurrencies. That makes me happy