Changing the hashing algorithm
-
Are we considering using a current non scrypt algorithm. This may man we can reuse the current mining software and not have to develop our own. I’m thinking that may be the best option. Less work for us and a proven algo.
The trouble with doing that is it is a bit like if Bitcoin decided to change over to Scrypt at the last minute as SHA256 ASICs were released.
It’s a larger coin taking over a smaller coins algorithm.
-
That’s true and I guess we don’t want to cause trouble for other coins but are we sure that this would be a problem?
If bitcoin did change to scrypt it would be in direct competition for hashes with litecoin, Feathercoin etc etc and has essentially cut out its ASIC miners.
ASIC miners would have to find a new sha256 friendly home. And the GPU scrypt guys would have to make the choice of staying with their coin or jumping ship to bitcoin.
I suppose because of the higher value of bitcoin and the very quick decrease in hashes it would be very lucrative to mine initially at least which would likely be the deciding factor.
If we consider sha3 we would be moving into maxcoin territory. As we don’t really have an ASIC problem yet the scenario is different. So we are not alienating any/many of our loyal miners which means our hash decrease would be lower. We would have to work out at the difficulty we move over to sha3 territory how would we compare financially to maxcoin to figure out the consequences to both Feathercoin and maxcoin.
-
That’s true and I guess we don’t want to cause trouble for other coins but are we sure that this would be a problem?
If bitcoin did change to scrypt it would be in direct competition for hashes with litecoin, Feathercoin etc etc and has essentially cut out its ASIC miners.
ASIC miners would have to find a new sha256 friendly home. And the GPU scrypt guys would have to make the choice of staying with their coin or jumping ship to bitcoin.
I suppose because of the higher value of bitcoin and the very quick decrease in hashes it would be very lucrative to mine initially at least which would likely be the deciding factor.
If we consider sha3 we would be moving into maxcoin territory. As we don’t really have an ASIC problem yet the scenario is different. So we are not alienating any/many of our loyal miners which means our hash decrease would be lower. We would have to work out at the difficulty we move over to sha3 territory how would we compare financially to maxcoin to figure out the consequences to both Feathercoin and maxcoin.
I believe what wrapper meant is that we’d be the larger coin taking over the smaller coins algo.
-
no I understand that and Feathercoin has been around a lot longer than maxcoin in the above scenario.
But what im saying is that the reason for people to change which coin they mine is not based solely on the size of the coin.
Profitability and Loyalty seem to be the main two factors. in different orders depending on the individual.
If it was only size then why is anyone mining anything other than Litecoin in the Scrypt arena.
Im just trying to raise the question really. Are we sure that we will have any negative effects on the current occupants of the new algo?
-
Yes, I’m just trying to grasp the pro’s and cons.
From a technical perspective, I’m for SHA3. It’s simple, longer term ASIC resistant, mining software is already available. But the philosophy for the Feathercoin change of hash algorithm is that Litecoin owns the space, so that would apply to us taking over a smaller coin’s algorithm.
Whilst we’re battling Litecoin as the underdog, there is not the temptation to misuse our networking power. At least Litecoin have the justification of getting Scrypt off the ground (for being dismissive of smaller coins). We do benefit hugely from the common software platform and that would start to be diluted by altering the hashing algorithm.
I will support the way forward, but I see this as a fundamental decision that has got to be as right as we can get it.
-
If it was only size then why is anyone mining anything other than Litecoin in the Scrypt arena.
Im just trying to raise the question really. Are we sure that we will have any negative effects on the current occupants of the new algo?
At the current time Scrypt based coins are being attacked in various ways all the time. Whilst that is spread out over a larger number of smaller GPU miners,it is just about survivable by the larger coins networks.
But with the introduction of ASICs the power of the network is more easily concentrated in smaller numbers of hands, which give considerably more power and incentive to destroy smaller coins. Re: Terracoin and the fate of other SHA256 alternatives. At the best they needed multiple hard forks which destroys reputations.
This is what some see as the problem. I think that it is not as high a danger for second generation coins, particularly as we have developed improved settings to counter those dangers, but it is significant and needs deep consideration.
-
bitcoin-side-chains â€"â€"A huge threat
PoW Alt-Coin will encounter major challenges. LiteCoin and DogeCoin may crash. I want to be a choice PoS. Of course, this will not necessarily be the best choice. We want to make early preparations.
http://www.coindesk.com/bitcoin-core-developers-bitcoin-side-chains/
-
Hi Lizhi, I hate PoS, it is even more centralising to those with a stake, as a small miner it will make me pointless.
I don’t see how it cures the problem either, big ASICs will soon have a Big Stake.
-
I would be interested in doing some calculations on what our mining profitability would be after a change to sha3. And how that compares to maxcoin and any other coins under the sha3 algo.
If we said that our standard rate of ~3ghs (scrypt) was converted entirely to sha3.
As mining scrypt != mining sha3 on the same hardware.
And the difficulty was set accordingly for that rate of hash would we be more or less profitable than maxcoin based on their current hash/difficulty/value?
EDIT:
Gigabyte 750ti
stock scrypt is around 260khs
Stock sha3 is around 62000khs or 62mhsThat’s a huge ~238x speed increase.
This would bring our hash speed up to
~715Ghs pushing up the difficulty I’m guessing. -
-
From what I understand, SHA3 would just replace the hashing part of the algorithm , we would retain scrypt to enforce the memory usage.
In that case Maxcoin would have an ASIC designed first, and by adding memory you would make a Feathercoin ASIC. Which would seem a bit perverse.
I would need to look more into how Maxcoin are useing SHA3 to know for sure.
In the end the actual hashing rate is immaterial, the difficulty will adjust accordingly. Profitability calculations currently are non sense dreamt up by exchanges and don’t take into account any technical differences or efficiencies of the coins.
-
I want to be a choice PoS. Of course, this will not necessarily be the best choice.
Hi lizhi!
I was lurking in this thread since the beginning but felt I did not have enough knowledge to post. However I would like to trow my 2 cents on the POS system.
For what I realized:
- You keep the money long enough to build “coin age”
This strongly discourage coin usage: I use my money for buying and will lose my chances to stake. If that’s the major coin minting mean, that means you won’t get a cent out of your coins if you make them move around.
- The more you have the more you will gain
This is a no brainer: you accumulate enough money, lock them in a safe (your wallet) and wait for profit. You will need to invest a lot of real money to buy enough to stake profitably.
- What about third party wallets?
If you wish to leave your money on a third party wallet (being that an exchange, an online wallet or whatever) are you completely sure you will get the stake profits? I am not. They will probably gain from you money.
The result is a complete centralisation of the minting system with rich people gaining more money than “poor” miners. No matter how you look at it. The latter will also be discouraged to use their money as they will also lose the minting capability associated to their coins.
Please notice that I’m not talking about security but this system smells like the modern banking system:
-
If you spend your money you won’t have enough to get “interests”
-
Those with the money create the money and put it back in circulation (with a price). Did somebody said “central banks”??
-
Said “bank” will gain from your coins’ stakes.
This resemblance is unbearable for me. I want a coin which is 100% based on the proof of work. It is energy inefficient and bears all the problems you may think about but you can (at least) fix them and reward your miners with coins for their efforts. Being them big or small ones they will get paid for their share rate. This is the reason why I would also like a CPU mineable coin: in the future we will probably want to have miners installed on every single piece of tech hardware mining to compensate energy usage but that’s probably an unrealistic scenario.
Just my 2 cents…
-
- You keep the money long enough to build “coin age”
True.
- The more you have the more you will gain
In absolute figures only. You don’t mind if a bank pays more interest on a 1 million USD deposit than on a 1 thousand USD one, don’t you? The relative percentage is supposed to be the same.
- What about third party wallets?
The hard truth is once you have sent your coins out of your wallet they are no longer yours. It’s the receiver’s responsibility to recognise your ownership and return the coins per your request. The receiver collects the interest and keeps it unless there is a special agreement. If the receiver gets robbed, gets bankrupt or just runs away, you have no coins no matter what you think. Staking is much better: you are a bank of your own.
-
Has the Groestl algo been considered?
-
for anyone interested in looking Groestl was one of the 5 finalists for the SHA-3 competition.
The 5 were.
BLAKE, Grøstl, JH, Keccak, and Skein
Obviously keccak won the competition but only due to its innovative approach and not necessarily because of an increased level of security or performance.
http://nvlpubs.nist.gov/nistpubs/ir/2012/NIST.IR.7896.pdf
see section 3 for the selection process and reasoning.
-
for anyone interested in looking Groestl was one of the 5 finalists for the SHA-3 competition.
The 5 were.
BLAKE, Grøstl, JH, Keccak, and Skein
Obviously keccak won the competition but only due to its innovative approach and not necessarily because of an increased level of security or performance.
http://nvlpubs.nist.gov/nistpubs/ir/2012/NIST.IR.7896.pdf
see section 3 for the selection process and reasoning.
Thanks for providing more information. I don’t know much about hashing algos, or Groestl, but I do know it’s a new efficient algo for GPU mining (having mined Groestlcoin on my GTX 750 Tis).
-
I have been lurking for a while but it seems like the best and most popular thing to do these days with GPU coins is to optimize the algorithm for low GPU power consumption. This gives GPU miners larger electrical cost margins to work with compared to other GPU algorithms. You can also dramatically cut down on heat produced with decreased power usage resulting in lower cooling costs. We have the technical knowhow on the FTC team to come up with an algorithm and I would think in order to generate more publicity we should create our own hashing algorithm rather than recycle another. Drk coin’s algo is way to complex and unless mathematically proven there are probably collisions and shorter ways to get the same hashed result in the algo allowing the creation of super fast and efficient miners. Keep the new algorithm simple and easily provable that there aren’t any collisions or shortcuts in the hashing calculations.
-
I like the idea of creating an algorithm, that works with low power consumption on the gpu, but creating a totally new one requires much more work, as a full set oh mining software has to be developed.
Also an intense testing would be needed to be sure, that the algorithm is working as expected and robust against malicious mining attacks of various types.
-
Not only mining attacks but crypto attacks to make sure the algo itself could not be bypassed and was robust and secure.
-
An important news:
http://www.reddit.com/r/reddCoin/comments/249dnl/major_announcement_reddcoin_to_implment_new/
A new algorithm called Proof-of-Stake-Velocity (PoSV) be born.
https://github.com/reddcoin-project/reddcoin
I think a small miner can be compensated by the contribution in the community.We can set up a fund to solve this problem.
