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)
Topic: Which altcoins aim for GPU resistivity? - page 2. (Read 2967 times)
Given the starter advantage and the fact that Bitcoin now controls so much hashing power, it is doubtful that any other GPU or GPU resistant coin can supplement it. The best that they can hope for is to be a complement to Bitcoin
Hashing power flows from desirability and protects the network from attacks. The amount of hashing power a coin has is irrelevant to the ability of another coin to supplant it. It would be like saying a university would remain the best university because it had the most security guards. Bitcoin's desirability is its strength - this flows from recognition and acceptance.
Given the starter advantage and the fact that Bitcoin now controls so much hashing power, it is doubtful that any other GPU or GPU resistant coin can supplement it. The best that they can hope for is to be a complement to Bitcoin
Yup, they are all botnetcoins.
Exactly.CPU only coins lets me earn a ton of $$ using botnets, VPS, EC2 and the list goes on
Focusing on GPU resistance is just going to help the botnets a lot.
No kidding. People just need to give up creating contrived solutions to non existent problems.
Focusing on GPU resistance is just going to help the botnets a lot.
Unfortunately with GPUs now having 6 gigs of RAM, and faster RAM than most desktops, one GPU now has more RAM than most consumer desktops, doesn't it?
Yes . . . but the advantage of the GPU is the large number of parallel cores . . having to assign 1GB of memory means most of those cores won't be in use - you'll have 6 (underpowered GPU) cores running against your 6GB ram . . a well programmed GPU miner might come up to the same level as a CPU miner. Maybe, and at approximately the same capital cost.
Plus, a desktop is a rapidly becoming obsolete fossil that businesses maybe still use because too many of their employees who need a computer don't have a laptop or handheld yet; isn't a desktop a kind of specialised equipment less and less consumers tend to have in their homes? Maybe if first they get some kind of run your own business thing in their minds they might go out and buy a desktop but isn't it more and more likely they have a handheld or laptop or xbox or whatever?
That's a fair point. I think it'll run nearly as well on laptops with a good chunk of memory. I'm not sure about handhelds. I'd love to see a coin optimized for mining on Xbox or PS3 - but it's a bit beyond my technical capabilities at present.
Do entry level desktop machines at Walmart routinely have more than 6 gigs of RAM?
I think 2GB - 4GB is standard now. There are very few uses for more than 4GB of memory. Anyway the key metrics here are 1. GB/Core 2. Power efficiency of the machine. You might see similar efficiency on a dual core machine with 2GB as on a quad core with 4GB.
i do like the intermittent nature of their mining as you envision it though. Six billion people each running a miner 24/7 on their phone are not likely to get more than one six-billionth of the loot, but if it only takes one phone one week to mine one person's typical ongoing-average amount of coin wanted and each only wants/needs that coin once per year each could rake in up to fifty-two six-billionths of the loot...
Well I think there is a demand/supply curve here. Certainly not everyone would be mining, and if they were rational, they'd need to consider the cost of it.
3 types of miners would dominate -
1. Those with the most efficient hardware
2. Those having the greatest difficultly obtaining coin by other methods.
3. Those not bearing the costs
Hmm, those number still seem to work out to be somewhat scant. maybe we need to be using not electricity, which goes up in heat, to buy with but, rather, something that isn't burned up on use like that, so that when someone mines $10 worth of coin, close to $10 worth of what they expended to mine it goes to someone else who is also looking to mine $10 worth of value/wealth.
Real trade seems to work that way: I want to "mine" (aka take posession of, make it mine, cause it to belong to me) ten pounds of cabbage, you want to "mine" ten pounds of potatoes. We both run our "cause things to be mine, aka 'mine things'" software, my cost to run my mining app is ten pounds of potatoes, your cost is ten pounds of cabbage...
Real trade seems to work that way: I want to "mine" (aka take posession of, make it mine, cause it to belong to me) ten pounds of cabbage, you want to "mine" ten pounds of potatoes. We both run our "cause things to be mine, aka 'mine things'" software, my cost to run my mining app is ten pounds of potatoes, your cost is ten pounds of cabbage...
Yes, but I think that's a somewhat wider discussion.
An algorithm that utilizes the maximum power of a standard desktop would be ideal. The idea I'm looking at currently requires a large amount of memory per cpu core. 1GB to 2GB with lots of branching. This fits the profile of desktops/servers/laptops.
Unfortunately with GPUs now having 6 gigs of RAM, and faster RAM than most desktops, one GPU now has more RAM than most consumer desktops, doesn't it?
Plus, a desktop is a rapidly becoming obsolete fossil that businesses maybe still use because too many of their employees who need a computer don't have a laptop or handheld yet; isn't a desktop a kind of specialised equipment less and less consumers tend to have in their homes? Maybe if first they get some kind of run your own business thing in their minds they might go out and buy a desktop but isn't it more and more likely they have a handheld or laptop or xbox or whatever?
Do entry level desktop machines at Walmart routinely have more than 6 gigs of RAM?
i do like the intermittent nature of their mining as you envision it though. Six billion people each running a miner 24/7 on their phone are not likely to get more than one six-billionth of the loot, but if it only takes one phone one week to mine one person's typical ongoing-average amount of coin wanted and each only wants/needs that coin once per year each could rake in up to fifty-two six-billionths of the loot...
Hmm, those numbers still seem to work out to be somewhat scant. maybe we need to be using not electricity, which goes up in heat, to buy with but, rather, something that isn't burned up on use like that, so that when someone mines $10 worth of coin, close to $10 worth of what they expended to mine it goes to someone else who is also looking to mine $10 worth of value/wealth.
Real trade seems to work that way: I want to "mine" (aka take posession of, make it mine, cause it to belong to me) ten pounds of cabbage, you want to "mine" ten pounds of potatoes. We both run our "cause things to be mine, aka 'mine things'" software, my cost to run my mining app is ten pounds of potatoes, your cost is ten pounds of cabbage...
-MarkM-
I was thinking along the same line. Coud computing and computing-as-a-service make cpu-only coins the same kind of arms race as a GPU compatible coin.
The aim is to make dedicated mining unprofitable.
Let's say you invest in the most specialized equipment, but still find that you need to invest $12 of electricity to get $10 of coin back. You wouldn't do it - you're not even making your variable costs back let alone your fixed costs or profit.
But a casual user, who needs $10 of cryptocoin to register a domain name - he might spend an extra $5 in electric rather than try to set up a wire transfer to MtGox.
Ideally we want a very broad base of people mining small amounts of coin.
Can you refresh our memories as to why we want to resist GPUs?
Sure. One problem with Bitcoin is distribution. A relatively small number of users currently hold most of the Bitcoins. Newly generated Bitcoins are minted by a relatively small number of miners. To acquire coins, you need to interface with the traditional banking system, using intermediaries subject to governmental and commercial attack. These barriers hinder the adoption of cryptocoins.
GPU mined coins fare slightly better, with a user able to acquire a GPU and mine on a relatively equal basis. However this takes technical expertise and initial setup investment.
Let's say a user wants to register a domain name with namecheap for $10 . . . it would be ideal if he could convert approximately $10 worth of electricity into a crypto coin and spend it.
One solution is to leverage existing hardware so that specialized miner who invests in optimized hardware doesn't gain a disproportionate advantage because casual users don't have an investment cost - they already have the computer, it's a sunk cost. The only technical expertise required is to download and install QT wallet style software - maybe with automatic connection to a pool.
An algorithm that utilizes the maximum power of a standard desktop would be ideal. The idea I'm looking at currently requires a large amount of memory per cpu core. 1GB to 2GB with lots of branching. This fits the profile of desktops/servers/laptops.
So really we need, or should want, to be aiming at handheld devices, as mass sales of homebound/desktop devices keep falling while sales of handheld devices keep climbing?
That's not a bad idea - handheld devices are designed to have a very low power consumption. I wonder how fast a handheld device could burn through $10 of electricity - I think $10 is a good basic unit for creating an amount of currency that is usable online - for registering a domain name for example. It raises an interesting question - are handhelds optimized to do anything (that could form part of an algorithm) that other hardware is not.
Possibilities for gains always igniting an arms race. Maybe abstractcoin is the most resistant as most people don't want to get involved in a race for thinking 
.
. Can you refresh our memories as to why we want to resist GPUs?
Are we trying to get it down to a pure battle of who throws the most money at mining so that people who throw their money at chip and board manufacturers don't gain an advantage over people who throw their money at Walmart?
That is, do we hope that by using commodity parts the mass demands for the parts will end up making them more cost effective than more-efficient parts that don't have a lot of outlets selling mass quantities of them?
So really we need, or should want, to be aiming at handheld devices, as mass sales of homebound/desktop devices keep falling while sales of handheld devices keep climbing?
-MarkM-
Are we trying to get it down to a pure battle of who throws the most money at mining so that people who throw their money at chip and board manufacturers don't gain an advantage over people who throw their money at Walmart?
That is, do we hope that by using commodity parts the mass demands for the parts will end up making them more cost effective than more-efficient parts that don't have a lot of outlets selling mass quantities of them?
So really we need, or should want, to be aiming at handheld devices, as mass sales of homebound/desktop devices keep falling while sales of handheld devices keep climbing?
-MarkM-
I was thinking along the same line. Cloud computing and computing-as-a-service make cpu-only coins the same kind of arms race as a GPU compatible coin.
Can you refresh our memories as to why we want to resist GPUs?
Are we trying to get it down to a pure battle of who throws the most money at mining so that people who throw their money at chip and board manufacturers don't gain an advantage over people who throw their money at Walmart?
That is, do we hope that by using commodity parts the mass demands for the parts will end up making them more cost effective than more-efficient parts that don't have a lot of outlets selling mass quantities of them?
So really we need, or should want, to be aiming at handheld devices, as mass sales of homebound/desktop devices keep falling while sales of handheld devices keep climbing?
-MarkM-
Are we trying to get it down to a pure battle of who throws the most money at mining so that people who throw their money at chip and board manufacturers don't gain an advantage over people who throw their money at Walmart?
That is, do we hope that by using commodity parts the mass demands for the parts will end up making them more cost effective than more-efficient parts that don't have a lot of outlets selling mass quantities of them?
So really we need, or should want, to be aiming at handheld devices, as mass sales of homebound/desktop devices keep falling while sales of handheld devices keep climbing?
-MarkM-
I'm adding Primecoin to this list. It's difficult to say how efficient a GPU miner will be, I'm seeing estimates of 2x - 4x more efficient.
A GPU miner for YAC has been released for a few weeks now. Just an FYI.
I'm adding Primecoin to this list. It's difficult to say how efficient a GPU miner will be, I'm seeing estimates of 2x - 4x more efficient.
Both OneCoin and YbCoin have CPU-only miners right now.
Thanks - and if I understand correctly, these both use the same method as YACoin for GPU resistivity. . . which is to make GPU mining less efficient over time by increasing the N in Scrypt(N,1,1) hashing.
Yup, they are all botnetcoins.
Quote
The scrypt-jane modification allows the N in scrypt(N,1,1) to change. Litecoin uses scrypt(1024,1,1)
http://bitcoin.stackexchange.com/questions/11487/how-does-the-cpu-hashing-algorithm-used-in-yacoin-differ-from-scryptThe N events halve hashrate, difficulty and block rewards change all the time aswell, fun coin imo.
http://yacexplorer.tk/graphs.htm
Both OneCoin and YbCoin have CPU-only miners right now.
SIC, YAC, eMunie in the near future.
Thanks for orienting me on where to start my research.
Do I have these right?
SIC - Sifcoin
'Blake=>BMW=>Groestl=>JH=>Keccak=>Skein' . . . a complex chain of hashing requiring a lot of development effort to bring GPU miners online.
YAC - Yacoin
Uses scrypt, but changes the parameters dynamically in response to some change in difficultly . . . currently that parameter does not prevent efficient GPU mining.
eMunie
Bit confused about this one - quite a different setup.
Jump to: