This would be the less easy solution, requiring solutions to meet some threshold of "quality." This is complicated by the relative weighting of significance of Arimaa pieces being generally unknown, so scoring would always have to be assumed to have a margin of error.
so.... what would constitute "solving a position"? doing checkmate? verifying it should be fast, so the verifier can't explore all the positions to validate that for example there was an unexplored move that would avoid the checkmate.... I'm very interested on this, but I couldn't make it work. I'd encourage you to try to devise a (MOTO) solver
Yes, I wanted to make a bot but I never had time to do it. MOTO is harder than Mario, but check this out: https://www.youtube.com/watch?v=DlkMs4ZHHr8 The A* algorithm used here to solve Mario requires a heuristic function, I guess distance to the coin could be used. X-whatever hashing schemes look downright weak.....)
Yes, we agree those are bs. Chained hashing looses entropy with each new hash. Personally, I think that these concepts will lead to some very specific and very big somethings, but that is another discussion for another day in another thread.
this is the game we should all be playing: http://fold.it/portal/eventually the AI that makes the singularity may be a cryptocurrency! |
|
|
|
arimaa ( http://en.wikipedia.org/wiki/Arimaa) is easy for humans and hard for computers, but the problem is how to transform header hashes to game positions and how make a difficulty function Yes, this has lately become the canonical example of "hard for bot" game. I've been toying with the idea of "whipping this up" as a fork of MOTO, just to illustrate the point. Both of the problems you mention are pretty easily met with basic solutions. It wouldn't be great, but it would work. I'm intrigued... I can think of many ways of transforming hashes to game positions, but I don't imagine how would one decide if the work done is enough for declaring a block.. some positions will be trivial loses, some trivial wins, and some very intricate. Would the work be maximizing some fixed "score" function? The problem would be in taking the security of such a coin seriously. Although we believe Arimaa to be difficult, I'd actually have an easier time believing it could be "suddenly solved" in the way Domob might suggest than believing MOTO could be. (I'm probably wrong, but we can't really confidently say which is "stronger" and therein lies the rub.  ) of course it depends on how is arimaa implemented, however I'd say it's more likely to have a sudden breakthru in MOTO: some general strategy like "stay balanced, don't go too fast, and aim for the coin". Arimaa looks like chess: it's just exponential and you can have heurisitics but at the end of the day you have to explore a lot of positions. |
|
|
|
the tetris paper: http://erikdemaine.org/papers/Tetris_COCOON2003/paper.pdfI hope you agree that proof-of-play tetris is right out.
aargh! stop using sentences like that! you are forcing me to make tetriscoin!  The bots will just inevitably be able to do the pattern matching much more efficiently. Convince me otherwise?
we agree on that, no need to convince you apparently, bejeweled and "candy crush saga" fall in the same category: np complete, which means it would work if you use the correct difficulty function, but bots would kill humans anyway arimaa ( http://en.wikipedia.org/wiki/Arimaa) is easy for humans and hard for computers, but the problem is how to transform header hashes to game positions and how make a difficulty function |
|
|
|
many games are formally reducible to the 3sat problem and as such could be asserted to be sufficiently strong... unfortunately none of the ones I know of offhand would be very fun, they are really just academic exercises.)
..
I promise you that tetris really cannot work in any reasonable way as the proof function!
it's the other way around: any computer game can be reduced to the 3sat problem, the interesting and not always possible part is reducing 3sat to the game in question. And it turns out it can be done for tetris. All of the following problems for tetris are np complete: - Maximizing the number of rows cleared while playing the given piece sequence. - Maximizing the number of pieces placed before a loss occurs. - Maximizing the number of simultaneous clearing of four rows. - Minimizing the height of the highest filled grid square over the course of the sequence. this means tetris could be used to solve 3sat! you said it yourself: the solver is exponential. The steepness of the curve and the required length for the sequence... well, that remains to be seen.... maybe more complexity can be added by making the playing area wider, or some other small variation to the game that would add more possibilities for each move we may be getting off-topic with the tetris-specific discussion, but I'm exploring the idea because I feel a tetris-coin would be cool, even if bot-dominated and eventually unplayable by non-augmented humans, still cool |
|
|
|
Tetris can't work because there is a deterministic optimal solution if you know the upcoming sequence of blocks - which you must. The search space for each block state is only a few thousand nodes, so although the complexity of the solver is exponential in the number of pieces used the complexity curve has a low coefficient, it is not steep. (You'd need some infeasably large game before it would become any sort of challenge at all for a bot, at which point it becomes pointless to humans as a game.) It wouldn't even be a matter of level iteration to find an easy level, you could just "auto solve" any given level. A game where the primary challenge is any form of "the player doesn't know what's coming next" simply doesn't work.
but we could adapt the game: show the player all the future pieces and add a "rewind" button as in MOTO... anyway... I'm just saying it might be possible and fun (until bots)... we agree bots would win... EDIT: I read somewhere that tetris is NP-complete, so yes: the solver is exponential I was asked about it, so at least one random internet user would be interested.
hehe, not enough for me but I'd think about it if there's more than that |
|
|
|
Yes, I agree wholeheartedly here. Just sticking an arbitrary game challenge in as the proof function doesn't generally work, as most games are not sufficient. You wouldn't believe how many requests I've received to do something like motocoin, but using something like "temple run" or "tetris" games as the proof function....
You're right, however with some ingenuity lots of games could be adapted to fit the needs... tetris!? that's quite feasible... I'd do it but it would quickly turn hard to play and become a race on how quickly could bots iterate thru nonces until getting the easiest game... (kinda like MOTO) in this case the challenge to "find mechanisms which allow for both bots and humans to be able to mine "peacefully" together"... would be too hard I'm afraid would there really be interest in a tetris coin? |
|
|
|
it's just an idea... I wanted to see how people felt about it.... I think we would set 1RIC = 1RICē during the transition phase. (or maybe 1RIC = 1.2RICē in order to favor RIC, we'll have to think about it carefully) and you would choose if you move your coins to RICē or stay in RIC. PoW and mining would work the same... we could even have merged mining... I wouldn't abandon RIC. anon tx is getting attention, I agree that most attempts won't hold water, but the ring signature approach is the only one I believe that can actually work (at least the theory is sound, I haven't really verified current implementations). maybe we could add ring signature support and do the 512 bit nonces and the 5 mins per block things as a hard fork in RIC instead of making a new coin... but that's more work... if we like the RICē name we could do the hard fork and rename RIC to RICē and have nice new features without a new coin |
|
|
|
if so then please write it like Ricē +1 ...and add RAM requirements (shinycoin) to make it a truly a CPU-only coin. I like Ricē But right know I don't know how to make it more RAM demanding... (that is, without radically changing the PoW). After seeing the alleged speed of primecoin's gpu miner, I think that a Ric gpu miner will eventually be unavoidable. Newer GPUs will keep getting more and faster RAM anyway. I was thinking of making the nonce (offset in the block's header) be 512 bits long instead of 256. This would allow for more efficient mining putting us closer to the elusive world record. Also 5 mins per block: the wallet would sync faster and each block would have more difficulty, getting us closer to that record.... |
|
|
|
|
Hi! I've been a little sick (just a flu) and I didn't touch a keyboard in like 4 days (!) that's weird for me... anyway, I'm still working but I didn't make much progress lately.
One idea I've been fantasizing about: we could make Riecoin2, based on the new ring signature anonymous coins (Monero/Bytecoin). It would be a fork of their code but using out PoW. There should be a way to transition from RIC to RIC2. Maybe we could have a burn period where we would burn RIC to transform them into RIC2, or I could just premine a lot of RIC2, sell them for at 1 RIC on an exchange and after a while just burn the unsold ones. The latter would be easier than monitoring both chains.
What do you think? those who don't like it would be allowed to continue in the current RIC, and the fact that you would be able to use it to buy RIC2 at a fixed price (for a while) would make the price (of RIC) go up. Or would it hold RIC2 down? who knows....
|
|
|
|
|
HunterMinerCrafter, I agree with your explanations:
MOTO is the only true PoP, however I feel there has to be a distinction between HUC and the rest. After all, the game is actually embedded in the blockchain and it can be used to generate coins, even if the game itself doesn't secure the network. Maybe we can extend the definition of "mining" to any means of "creating coins", so in this case HUC could be "human mineable" but not "Proof of Play". Or we could make up another term, whatever, but it deserves a distinction.
I'd like to add that I'd bet that all PoP will eventually be mineable by bots, which would eventually outplay humans and make human-mining impossible or at least unprofitable. I like the concept and I think it can be useful to push the boundaries of AI, but don't expect to make money by playing, at least not in the long term.
|
|
|
|
I'm sorry to keep losing miners... Dude... I'm not going anywhere... I'll keep mining this coin as long as it will mine-able. You'll have to shut down the servers for taking down my miners.  Keep up the good work! yeah! that's the attitude |
|
|
|
hi! sorry I'm late for Thursday's update, been very busy. thanks for the kind words and I'm sorry to keep losing miners... riecoin really takes a lot of my time As mentioned earlier, pool code that supports stratum protocol has been published but the fastest miners that people use run the xpt protocol. So I'm working in adding stratum support to dga's miner, which is currently connecting to my pool but giving results like this: [00:00:32] 2ch/s: 0.0000 3ch/s: 0.0000 4ch/s: 0.0000 Shares total: 0 / 0
[00:00:40] 2ch/s: 0.0000 3ch/s: 0.0000 4ch/s: 0.0000 Shares total: 0 / 0
[00:00:48] 2ch/s: 0.0000 3ch/s: 0.0000 4ch/s: 0.0000 Shares total: 0 / 0
[00:00:56] 2ch/s: 0.0000 3ch/s: 0.0000 4ch/s: 0.0000 Shares total: 0 / 0
still debugging... |
|
|
|
I agree. My own riecoin ( riecoin.org) is innovative but didn't have the success that other less innovative coins had. However perseverance is another trait that -like innovation- can lead to success. I'll keep pushing it as far as I can. |
|
|
|
Should that be our next move Gatra? Would be nice to implement stratum compatibility directly into the wallet..
it's a posibility, yes. I've been thinking about this and I found no other coin has it. There should be an option to restrict ip, use SSL, and everything like the rpc port has, in order to avoid opening a new attack vector. re-thinking that, we already have getblocktemplate and blocknotify which replace the stratum functionality... maybe we could just add long-polling to getblocktemplate. I'll investigate this possibility, I think there's a standard for long polling in gbt bitcoin does not implement it. |
|
|
|
Should that be our next move Gatra? Would be nice to implement stratum compatibility directly into the wallet..
it's a posibility, yes. I've been thinking about this and I found no other coin has it. There should be an option to restrict ip, use SSL, and everything like the rpc port has, in order to avoid opening a new attack vector. |
|
|
|
Take RIC for what it is, a very valuable experiment.
Crypto as a whole is moving forward.
RIC is still valuable. Any crypto with improvements and active dev. is a good crypto. I agree, but this truly has no chance of being adopted by the public (like 99% of alts) but nonetheless has provided a few unique concepts that can help us move along as a community. I would contest with this. Many argued that Darkcoin would not be adopted by the public but its in the top 10 right now. Dogecoin was a shocker but it is also top 10 right now. Yes, strictly speaking, bitcoin was not adopted by the public either. What I consider "the public" may have heard about bitcoin, but doesn't know what it is. So any altcoin has a chance. And I'm willing to take this as far as I can! I see more long term potential in RIC than in DRK or DOGE (which I think will eventually die). However I cannot be trusted on this: when I first read about bitcoin 3 years ago I thought it was a nice experiment without real potential, and I failed to investigate deeper and mine or invest in it, which would have made me a millionaire by now. |
|
|
|
Fortunately, there's an exit, a re-framing of the process to invert the control, allow coins to petition to be rescued. The petitioners are making the categorisation and if the CoinShield process works as expected, investors will press for petitions to be opened, there will be competition to get a petition accepted, arguments will be more focused and effective, voting will become a popular contributive activity. And when CoinShield later subtly repositions itself to become the first to offer insurance to altcoin operators, no-one will be in the least bit surprised.
I'm thinking devs would create shitcoins just to get rescued, something like insurance fraud... still, a very interesting idea |
|
|
|
I see a very healthy diversity of pow algo usage:
[img edited]
(note, log2 scale for presentation purposes)
nice chart! could you please tell us the source of your data? is it number of altcoins you are comparing? glad to see prime constellations is at least mentioned |
|
|
|
dont be so sure riecoin cannot be valuable in the long run and used by mainstream people
i have a question, from research i notice ypool hits almost every block
we have all heard the drama over gigahashio and bitcoin and the stories of block witholding attacks against other pools
can someone tell me how that is not an issue here ?
not only is it one pool it is closed source,
i would not run anything closed source ever, anyone who does should reconsider
yep, you bring up very good points, but: block witholding hasn't been reported for riecoin, it hasn't been an issue probably because it's an expensive attack. ypool being closed source is not a problem because even if the pool's backend is closed, the miners are open source, so no issue there. Even if the backend were open sourced, you can never know what software is the pool running, so in that sense it's ok: no need for miners to reconsider using their open source mining software. Also, the operators of ypool are very capable and have been honest so far (afaik). There is open source pool server code but it works using the stratum protocol while the most optimized miners use xpt, so nobody uses stratum pools. Centralizaction is not good and in fact ypool had some downtime this week and it impacted us by lowering the difficulty a little bit. We haven't deployed a big PR campaing for RIC because of that: we want to have many pools before going forward, and it's been my priority for some time. I diverged from it to release new clients only because there were security fixes for openssl. tl,dr: it is an issue but not that bad, miners are open source and we are working on the rest. BTW Gatra, any news after you release 0.9.2? update Thursday here once again!
Today, I worked on my "real life" job during the morning but spent the afternoon adding stratum support to dga's b15 miner. Almost there! There are at least 2 more pools ready to go after this is ready. cheers! gatra |
|
|
|
1395 now is looooooooong time low, I contemplate about starting the unoptimized ARM miner again ypool had some problems last night, which caused diff to drop a little also, I think some cpu miners moved to new cpu coins with cryptonight and maybe "wild keccak" (boolberry), but gpu miners exist for both so I expect them to come back soon |
|
|
|
|
Show Posts
