QuarkCoin!
Yep, I'm a fan. Quark gets a lot of things right.
Topic: Which altcoins aim for GPU resistivity? (Read 2977 times)
Yep, I'm a fan. Quark gets a lot of things right.
About time I added MemoryCoin to this list. I think it might be first real GPU-Resistant Coin.
https://bitcointalksearch.org/topic/ann-memorycoin-267522
https://bitcointalksearch.org/topic/ann-memorycoin-267522
QuarkCoin!
Yep, I'm a fan. Quark gets a lot of things right.
I've been away for a while, but I'm coming back around to the idea that a well designed altcoin could complement/supersede Bitcoin.
There's a bewildering array of them available and it's difficult to quickly compare their properties. Could anyone bring me up-to date on any current implementations that are designed for GPU-resistance?
There's a bewildering array of them available and it's difficult to quickly compare their properties. Could anyone bring me up-to date on any current implementations that are designed for GPU-resistance?
QuarkCoin!
What about putting all of this together and make a list please ?
GPU resistance as in Botnet friendly?
What happens when botnets use GPUs?
I've been away for a while, but I'm coming back around to the idea that a well designed altcoin could complement/supersede Bitcoin.
There's a bewildering array of them available and it's difficult to quickly compare their properties. Could anyone bring me up-to date on any current implementations that are designed for GPU-resistance?
GPU resistance as in Botnet friendly?There's a bewildering array of them available and it's difficult to quickly compare their properties. Could anyone bring me up-to date on any current implementations that are designed for GPU-resistance?
The only reason someone wouldn't want GPU and faster is because they either are using their entire workplace as an overnight mining operation or a full fledged botnet.
I don't think GPU resistant at this point is a very good idea. But thats just me!
There's a bewildering array of them available and it's difficult to quickly compare their properties. Could anyone bring me up-to date on any current implementations that are designed for GPU-resistance?
It's not current implementation but old idea well worth implementing. GPU resistance is achieved by using difference in floating point math on x86 and OpenCL.
1) Instead of evaluating a total recursive function over a commutative ring of integers you could try a simpler thing. Require evaluating a specific value of primitive recursive function over the field of reals that has fractional dimension. Good starting example:
http://en.wikipedia.org/wiki/Logistic_map
just pick the value of r that is in the chaotic region.
Implement reals as long double that is 80-bit (10-bytes) on the Intel/AMD CPUs. Not all the C++ compilers really support long double, but the logistic map function is so trivial that you can include in-line assembly which will be in the order of 10 lines.
2) This is a variant of the above that embraces the enemy instead of fighting it. Implement (1) with reals as cl_double which is supported in OpenCL by NVidia cards but not by AMD cards. Then short AMD stock and go long NVDA for additiona gains.
Again, good luck!
http://en.wikipedia.org/wiki/Logistic_map
just pick the value of r that is in the chaotic region.
Implement reals as long double that is 80-bit (10-bytes) on the Intel/AMD CPUs. Not all the C++ compilers really support long double, but the logistic map function is so trivial that you can include in-line assembly which will be in the order of 10 lines.
2) This is a variant of the above that embraces the enemy instead of fighting it. Implement (1) with reals as cl_double which is supported in OpenCL by NVidia cards but not by AMD cards. Then short AMD stock and go long NVDA for additiona gains.
Again, good luck!
People keeping their optimisations to themselves maybe derives from turning people into not only interchangeable components but even into un-necessary components.
If a business can be massively expanded without taking on additional staff/operators then there is maybe not a lot of incentive to provide people with the tools to perform the functions of staff/operators, such as software enabling the operators to run the thing more efficiently.
A typical capitalist response seems to be that they are actually employing plenty of people, from the people who etch chips through the people who assemble machines and datacentres and the people who run the cloud computing services, so that anyone who wants some pay/gain/income from the capitalist's growing mining business need only go work in a chip factory or cloud service datacentre, they don't need the capitalist's secret sauce recipe to profit from the success of the secret sauce.
Partly people lead themselves into these situations too by not wanting to work. They want to just tell their computer or butler or servant or slave or whatever "provide me with with income" then go bask on a beach or whatever with income automagically coming in for hopefully a nice long time. They want people to be obsolete, even to obsolete themselves really in such a scenario. being obsolete living in a paradise sipping fancy drinks is the life of Riley, who wouldn't want to be obsolete?
To resist the botnets and massive purchasers of cloud services maybe what is needed is "the human element". keep things changing so code constantly needs to adapt, leading eventually to the code itself needing to be adapted so it can adapt to a wider range of situations than an earlier version was able to adapt to. A kind of Turing Test maybe in a way, if humans are able by observing the actions of code/scripts to predict their behavior well enough to come up with new changes of the environment that are specifically intended to trip up the code.
This already happens of course with cases such as switching to scrypt so chips etched with SHA256 cannot adapt.
But what about putting that kind of conflict/development into one coin, instead of more and more chains each with different gotchas intended to trip up a previous generation of automation do something like http://www.devtome.com/doku.php?id=cpu_mining in which the problem of initial distribution of the coins is separate from the problem of securing the currency system whose coins are being distributed.
If one person just isn't going to able to handle each and every individual task of a massively parallel collection of tasks maybe they will have an incentive to recruit people to handle a lot of threads, and in order to rake in the most profit from the work of those people equipping them with the best tools might actually make business-sense?
-MarkM-
If a business can be massively expanded without taking on additional staff/operators then there is maybe not a lot of incentive to provide people with the tools to perform the functions of staff/operators, such as software enabling the operators to run the thing more efficiently.
A typical capitalist response seems to be that they are actually employing plenty of people, from the people who etch chips through the people who assemble machines and datacentres and the people who run the cloud computing services, so that anyone who wants some pay/gain/income from the capitalist's growing mining business need only go work in a chip factory or cloud service datacentre, they don't need the capitalist's secret sauce recipe to profit from the success of the secret sauce.
Partly people lead themselves into these situations too by not wanting to work. They want to just tell their computer or butler or servant or slave or whatever "provide me with with income" then go bask on a beach or whatever with income automagically coming in for hopefully a nice long time. They want people to be obsolete, even to obsolete themselves really in such a scenario. being obsolete living in a paradise sipping fancy drinks is the life of Riley, who wouldn't want to be obsolete?
To resist the botnets and massive purchasers of cloud services maybe what is needed is "the human element". keep things changing so code constantly needs to adapt, leading eventually to the code itself needing to be adapted so it can adapt to a wider range of situations than an earlier version was able to adapt to. A kind of Turing Test maybe in a way, if humans are able by observing the actions of code/scripts to predict their behavior well enough to come up with new changes of the environment that are specifically intended to trip up the code.
This already happens of course with cases such as switching to scrypt so chips etched with SHA256 cannot adapt.
But what about putting that kind of conflict/development into one coin, instead of more and more chains each with different gotchas intended to trip up a previous generation of automation do something like http://www.devtome.com/doku.php?id=cpu_mining in which the problem of initial distribution of the coins is separate from the problem of securing the currency system whose coins are being distributed.
If one person just isn't going to able to handle each and every individual task of a massively parallel collection of tasks maybe they will have an incentive to recruit people to handle a lot of threads, and in order to rake in the most profit from the work of those people equipping them with the best tools might actually make business-sense?
-MarkM-
You mentioned an expected 2x/3x for primecoin, my guesstimate is 2 orders of magnitude higher than that...
I've really no idea - I'm seeing another suggestion that the code for the CPU could be improved 1000x - that's just about plausible too. Any code benefits might be on top or exclusive from GPU efficiencies found. Many people might keep their optimizations to themselves too.
What, to you, is the important difference - if there is one - between buying coins from your electricity provider (having the cost added to your electricity bill) and buying them from your phone provider (having the cost added to your phone bill) ? Also presumably they have internet, would being able to have the cost added to your internet bill also be fine/appropriate?
The difference is the elimination of banks. In the scenarios above, the electricity firm or comms firm has become the bank. The sale of coins can be regulated and tracked. It's fine to have banks, but to keep them honest, it's better not to need them.
How about maybe they don't even have internet, they go to an internet cafe for that, would adding the cost of coins to their coffee bill also be fine? Or putting it on their tab if the important feature is paying later rather than right away?
I don't understand your point here.
So I do not think starting difficulty too low is reasonably counteracted by pools at all.
Yes, I don't suggest pools as a way to counteract birthing shenanigans - that stuff is important, but it is inside baseball.
I'm thinking of the medium term distribution of a coin . . . and I mean after 2 or 3 months . . . a pool with CPU coin allows wide participation.
I'm focused on a use case - there's a guy who wants to register a domain name for $10 - he doesn't have a bank account or a credit card, but he's got a CPU or access to a bunch of them - i want him to be able to use crypto currency to be able to register that domain name.
What, to you, is the important difference - if there is one - between buying coins from your electricity provider (having the cost added to your electricity bill) and buying them from your phone provider (having the cost added to your phone bill) ? Also presumably they have internet, would being able to have the cost added to your internet bill also be fine/appropriate?
How about maybe they don't even have internet, they go to an internet cafe for that, would adding the cost of coins to their coffee bill also be fine? Or putting it on their tab if the important feature is paying later rather than right away?
-MarkM-
sif, prime, quark, do not have public GPU miners yet, however it does not mean they aim for GPU resistivity.
You mentioned an expected 2x/3x for primecoin, my guesstimate is 2 orders of magnitude higher than that...
You mentioned an expected 2x/3x for primecoin, my guesstimate is 2 orders of magnitude higher than that...
So I do not think starting difficulty too low is reasonably counteracted by pools at all.
Yes, I don't suggest pools as a way to counteract birthing shenanigans - that stuff is important, but it is inside baseball.
I'm thinking of the medium term distribution of a coin . . . and I mean after 2 or 3 months . . . a pool with CPU coin allows wide participation.
I'm focused on a use case - there's a guy who wants to register a domain name for $10 - he doesn't have a bank account or a credit card, but he's got a CPU or access to a bunch of them - i want him to be able to use crypto currency to be able to register that domain name.
There's the potential for mining pools too - that would allow ordinary users to get coin. That would allow for some wide distribution.
Pools seem to suffer as much as anyone else from instamining. all the small miners together in a pool or individually an orphanfest is still an orphanfest it seems.
The launchers set up the initial nodes people are to download blockchains from, and presumably their own mining nodes well connected, so maybe the people launching get a lot of blocks assuming they set up the latencies correctly in their own favour.
So I do not think starting difficulty too low is reasonably counteracted by pools at all.
On the contrary, I think pools more likely to be useful in making a reasonable starting difficulty (which seems to probably mean at least a difficulty of fire for scrypt coins that are not insanely unpopular at launch like nuggets was; two seemed fine for nuggests since it had so very few miners) less unpalatable to small miners than it would be if they were expected to solo mine.
-MarkM-
Not sure about Merchants. An exchange is all the merchant they need, and the type of merchant likely to lose the least, or even actually make some profit, by tooling up to support yet another coin.
A normal merchant probably should be using a payment processor, so it would probably not be the merchant's decision which coins to accept, as the merchant just accepts his or her own local or favourite currency from the payment processor. But a merchant who actually has to have an information technology department running coin daemons webmasters adding more web code to direct payment buttons to the right daemons and so on is most probably effectively just another little old lady bystander being run down by the pump and dump bus.
The kind of miners who mine this crap do not care whether merchants came aboard and will be left holding the bag in a few hours when the next scamcoin launches, they will be off chasing that scamcoin. The merchant is just another bagholder, the miners need bagholders so I suppose they will be saying oh forget that old piece of crap, that is yesterday's pump, its already dumped, re-tool your shopping cart and sales literature to this new launch we are doing this hour...
Merchants are just victims, won't they wise up eventually? Heck maybe they already wised up, forcing the scamcoins to make up scam-services such as gambling (scam you out of money by fixing the odds against you; gambling is inherently a scam) if they want to pretend they have adoption...
A normal merchant probably should be using a payment processor, so it would probably not be the merchant's decision which coins to accept, as the merchant just accepts his or her own local or favourite currency from the payment processor. But a merchant who actually has to have an information technology department running coin daemons webmasters adding more web code to direct payment buttons to the right daemons and so on is most probably effectively just another little old lady bystander being run down by the pump and dump bus.
The kind of miners who mine this crap do not care whether merchants came aboard and will be left holding the bag in a few hours when the next scamcoin launches, they will be off chasing that scamcoin. The merchant is just another bagholder, the miners need bagholders so I suppose they will be saying oh forget that old piece of crap, that is yesterday's pump, its already dumped, re-tool your shopping cart and sales literature to this new launch we are doing this hour...
Merchants are just victims, won't they wise up eventually? Heck maybe they already wised up, forcing the scamcoins to make up scam-services such as gambling (scam you out of money by fixing the odds against you; gambling is inherently a scam) if they want to pretend they have adoption...
Agreed - not wise for mechants to try to support nascent coins except where they have a stake in the coin (like the casinos you mention) or where there's an incentive provided by those who have a stake in the coin.
Not sure about Merchants. An exchange is all the merchant they need, and the type of merchant likely to lose the least, or even actually make some profit, by tooling up to support yet another coin.
A normal merchant probably should be using a payment processor, so it would probably not be the merchant's decision which coins to accept, as the merchant just accepts his or her own local or favourite currency from the payment processor. But a merchant who actually has to have an information technology department running coin daemons webmasters adding more web code to direct payment buttons to the right daemons and so on is most probably effectively just another little old lady bystander being run down by the pump and dump bus.
The kind of miners who mine this crap do not care whether merchants came aboard and will be left holding the bag in a few hours when the next scamcoin launches, they will be off chasing that scamcoin. The merchant is just another bagholder, the miners need bagholders so I suppose they will be saying oh forget that old piece of crap, that is yesterday's pump, its already dumped, re-tool your shopping cart and sales literature to this new launch we are doing this hour...
Merchants are just victims, won't they wise up eventually? Heck maybe they already wised up, forcing the scamcoins to make up scam-services such as gambling (scam you out of money by fixing the odds against you; gambling is inherently a scam) if they want to pretend they have adoption...
-MarkM-
A normal merchant probably should be using a payment processor, so it would probably not be the merchant's decision which coins to accept, as the merchant just accepts his or her own local or favourite currency from the payment processor. But a merchant who actually has to have an information technology department running coin daemons webmasters adding more web code to direct payment buttons to the right daemons and so on is most probably effectively just another little old lady bystander being run down by the pump and dump bus.
The kind of miners who mine this crap do not care whether merchants came aboard and will be left holding the bag in a few hours when the next scamcoin launches, they will be off chasing that scamcoin. The merchant is just another bagholder, the miners need bagholders so I suppose they will be saying oh forget that old piece of crap, that is yesterday's pump, its already dumped, re-tool your shopping cart and sales literature to this new launch we are doing this hour...
Merchants are just victims, won't they wise up eventually? Heck maybe they already wised up, forcing the scamcoins to make up scam-services such as gambling (scam you out of money by fixing the odds against you; gambling is inherently a scam) if they want to pretend they have adoption...
-MarkM-
And designed to instamine almost all the coins in just six months, maybe to make sure botnets get the lot before any special hardware can get in on it.
There's the potential for mining pools too - that would allow ordinary users to get coin. That would allow for some wide distribution. Sure botnet controllers will profit from it, but for network health, I'm not sure there's any difference between preferring gpu farms over botnets.
1. GPU Mining = Narrow distribution to GPU Farms
2. CPU Mining = Narrow distribution to CPU Botnets + (Potential) Wide Distribution to CPU owners on mining pools.
I doubt anyone will wait for the sixth months to be up though before launching another to get another six months, mixing up the hash algorithms to make it "new" and "unique"...
That's an essential point - what can make a coin 'sticky' when it can be so easily cloned and relaunched. There are three things required, and in this order . . . Developers, Speculators, Merchants
Another reason is that they are sometimes created by whiny kids who wanted to "play miner", and can't do it anymore with bitcoin with the hardware that they happen to already have.
Which is beyond retarded, of course.
Which is beyond retarded, of course.
Quarkcoin is of interest too -
https://bitcointalksearch.org/topic/ann-qrk-quark-core-010-upgrade-260031
(Looks to be a clone of "SIC - Sifcoin" but with better parameters)
https://bitcointalksearch.org/topic/ann-qrk-quark-core-010-upgrade-260031
(Looks to be a clone of "SIC - Sifcoin" but with better parameters)
And designed to instamine almost all the coins in just six months, maybe to make sure botnets get the lot before any special hardware can get in on it.
I doubt anyone will wait for the sixth months to be up though before launching another to get another six months, mixing up the hash algorithms to make it "new" and "unique"...
So I guess there is no point blowing computing power on this one, better to just keep launching, maybe catch that big botnet guy asleep on one if you keep launching enough new ones all around the clock...
-MarkM-
Jump to: