Of course everyone is being vague in this thread, that's to be expected with these things. But i've gathered that the algo is going to be a """CPU"" only one, and it's going to be heavily premined / IPO'd aka buy in now or never make anything style
Why always botnet (""CPU"" aka cloud services) mining?
Why always Premining and Proof of Stake? Ethereum is PoW only (so far) Bitcoin was mined in huge quantities by Satoshi early onBitcoin started out very decentralized and then because of SHA256 became extremely centralized
Scrypt GPU era started out very decentralized and stayed like that, but now with ASICs on the horizon is about to go the way of bitcoin Litecoin was mined in huge quantities by me and others when it was first released (unfortunately I sold most of mine while they were a few cents each because I was paying for my undergraduate education). I was running a series of AMD 6-core servers on the network right after introduction. When GPU miners came out I had several MH/s on it as well. CPU coins are never actually decentralized, they're always taken over by botnets and cloud mining. CPU mining with computers is almost never ever reasonable
Anything besides mining, especially proof of stake, just results in a tiny group at the start having everything
I think pretty much everything results in a tiny group at the start having everything, and the matter of distribution is whether or not that tiny group decides to sell to others or not |
|
|
|
MC2's PoS is similar to PoA (and dissimilar to PPC) |
|
|
|
BTW, have you taken a look at the counterparty project ? It is also interesting and worth to follow IMO and have actual code that works which is not always the case for other new projects.
Also, I am still waiting for a message.
BM-2cUkz52xDAAxM36fjAs6ZNT4GS2jvDjHGo
Message was sent out, I will attempt to resend. Everyone else who PM'd me recently, I will be adding you to the newsletter shortly. |
|
|
|
Please use the nxt thread to discuss issues relating to that currency.
What will the coin acronym be for Ethereum, I just can't imagine people getting excited about saying Ethereum all the time. There has been ETH and ETR suggested on the other forum, with ETH favoured. I have been using ETH. |
|
|
|
|
Please use the nxt thread to discuss issues relating to that currency.
|
|
|
|
|
Fun night out at bitcoin decentral (hey peter todd). Vitalik gave a presentation on Ethereum that I imagine will be posted online pretty soon, if not already.
|
|
|
|
Very nice chart and calculation. It shows why intuitively I'm not too worried about the theoretical quantity of 2 trillion. It will take a long time to reach and 1.2 trillion is well under the quantity of USD in circulation today.
This also assumes 10 minute blocks for 60 years, isn't that a little conservative? Doesn't matter what the block time or block reward is. The number if coins changes, but the proportions do not. The proportion of coins you get in the premise adjusts based on the quantity of BTC coming in. The more BTC everyone invests, the less ETH everyone receives. |
|
|
|
Here's another good chart, this assumes 10 minute blocks with 50 coins issued per block continually to miners (as Bitcoin had when it began);  As you can see, there's no 1 trillion in the near future. By year 30, we've achieved 86,724,000, approximately 4 fold of that for Bitcoin. So in many ways this premine is insanely good to investors who decide to hold onto their coins (but in my opinion, a little generous to the foundation itself). I think it would be best if these (the premined coins) were given over a span of 10-15 years to investors, as a sort of "bond". This would also ensure that the foundation would keep up with maintaining the blocktree and ensuring that is works. |
|
|
|
A few people here didn't read the whitepaper closely enough.
1.2 trillion ether will NOT be issued at launch. At a tentative price of 0.0001 BTC, X ether will be raised. The initial money supply will be 1.5X, of which founders get .25X, ethereum reserve pool .25X, and fundraiser (investors) participants 1.0X.
After 1 year, total money supply is at 2X, with the ratios:
Founders .25X / 2.0X = 12.5%
Fundraisers 1.0X / 2.0X = 50%
Reserve .25 X / 2.0X = 12.5%
Miners .5X / 2.0X = 25%
Every year miners are mining .5X, this is the inflation rate
So year 1 inflation is (1.5X increase to 2X) = 33%
Year 2 inflation is 2X to 2.5X , 25%
Year 3 inflation is 2.5X to 3X, 20%
Year 4 inflation is 3X to 3.5X, 16.66%
Year 5 inflation is 3.5X to 4X, 14.2%
I personally think the distribution model is a lot fairer than the way Nxt and mastercoin did theirs. I'm not knowledgeable enough about emunie or bitshares to comment on their model, but I think ethereum has more promise than any of these other coins PLUS a fairer distribution.
This type of distribution is less than ideal, in my opinion.  We only hit the case where the premine is 10% of the mined portion of the coin at 27 years. For MC2, I aimed for <1% at the same time. But I guess it's all based on what a person considers a "fair" distribution. You can approximately reach this with linear mining inflation using these ratios (this is after one years time, with miner production held constant): Founders 0.5X / 12.0X = 4.2% Fundraisers 2.0X / 12.0X = 16.7% Reserve 0.5X / 12.0X = 4.2% Miners 9.0X / 12.0X = 75% I have no idea how many coins Satoshi has, but I'd guess it's less than 1.05m (but I may be wrong, I know it's quite large). |
|
|
|
|
Keep trolling to a minimum, please.
I'm going to fire off an e-mail to Vitalik later today about the distribution -- I hadn't read about it before, but I think a target of 0.5-1.0% of the total ETH supply at year 30 as distributed by mining is the best option.
|
|
|
|
Hybrid, like Peercoin, approach is probably the best bet if you don't want to go 100% PoS. Solves issue 1 and 2 that you brought up.
Satoshi himself was aware of the usages of coin-age, but he left the scene before it was actually implemented. Gavin seems to be coming around now, with all the talk about 51% attacks on the Bitcoin network.
Well, obviously I would agree the hybrid model is the way to go, and Vitalik was interested in it but thought it was beyond the scope of his project I guess I'll keep hacking away at MC2 |
|
|
|
Bitcoin has a current inflation of about 11%
Ethereum seems to have a initial inflation of 50%, based on the info.
Why not proof-of stake? Fixes the inflation and the 51% attack problem.
Cool project though. Can't wait to get more info at Miami.
Vitalik had trouble solving the "fork" issue, that is, that when two forks arise in a PoS coin, there is no incentive for PoS miners to select one over the other. That is, they are incentivized to mine both simultaneously at the same time. He thought he may have solved it today, but didn't seem terribly interested in pure PoS, especially because you need to typically do a 100% premine. There are other people I know who previously wanted pure PoS chains arguing about them for one reason or another now. I would guess that Satoshi himself/themselves also considered this (as it seems like the more logical means to construct a cryptocurrency network), but abandoned it for whatever reason. |
|
|
|
PS: TacoTime - can i ask why you felt it was appropriate to put up an announcement thread for someone elses company, and project? As per several comments by Charles in this thread, this has contributed to numerous headaches for them, as they were not quite ready to make this public. I think it would have been more professional to request permission to put them "on blast" like this ahead of schedule. Then again you did say you are a competitor ...
Vitalik didn't seem overly upset when I saw him in person today, but as I stated before, if the Ethereum staff has an issue with the thread they're welcome to PM me. Considering the whitepaper and repository have been public for a while, I don't think it was a huge issue. That being said, Vitalik is a really bright guy and I think he'll be the first person to successfully implement contracts and DAOs. He has a ton of neat ideas for things to use contracts for, and a very fast mind. My only huge reservation about the alt chain (tree) is that transactions and contracts are kept on the same layer for both security and liability. I would like to have these things kept separately; I worry that judges/political panels might consider contracts to be an entirely different thing than transactions because they represent actual instructions that may be criminal, and could make their propagation illegal. This might push the chain/tree over to the dark net; I feel like it'd be safer to potentially have two blocktrees, one for simple Bitcoin like equity transactions, and one for decentralized programming, and to have these two trees interact with and validate one another. The other thing is that executable code could bring with it some security issues, which also scares me. But it was just my thoughts on that matter, and it doesn't mean it'll be the case. I plan to invest some money even he decides this is a silly idea, but I'd feel more comfortable with them split. I'm going to try my best to be at the Toronto Bitcoin meetup @ Bitcoin Decentral tomorrow night, if anyone wants to catch me there. |
|
|
|
Hi!
Any news on the 0.05 version?
Im getting a wierd problem....
I have a second rig that i have configurated like my main rig...
When i do start, it start hashing but doesnt accept nothing!!! Its stays on ZERO for long as i wait until i press stop...
I have no ideia whats going on, everything is correctly configurated, all the drivers updated, etc...
Sometimes it works, sometimes it won´t...
Can anyone helpme out?
Thanks...
Best Regards,
LPC
I need more information. What does running the .bat file in /cgminer/ look like? |
|
|
|
|
All persons have been e-mailed the most recent newsletter on BM who had asked for it.
I will try to push another series of updates later on this week to all the BM addresses.
|
|
|
|
I was aware of this since about a week from the i3 forums. Apparently creator of ethereum is an ex founder of invictus.
I also read the white paper seems promising, however I see no mention of an IPO like OP says or am I missing something?
There's supposed to be a mastercoin style IPO where a certain number of USD/BTC yields some ETH |
|
|
|
Hm, what's up with the massive jump in diff, is Cointerra and/or Hashfast shipping in large quantity? Does this qualify for releasing more jupiters/etc?
Might not be a bad idea for the network since the whole GHash.io drama went down
Hashfast shipping in large quantity, hahaha. |
|
|
|
What I don't understand is who will do the computing for the Turing-complete "contracts"?
So a "contract" is supposed to be some script that is embedded on the blockchain. People can send messages to the contract that it will take as input to some computation. Maybe a simple example would be a vending machine. I send some bitcoins with a message "I want a Coke" to the vending machine contract. The vending machine processes my input and directs the physical vending machine in front of me to vend a Coke.
But that code has to run on a computer somewhere. Does it run on a miner's computer? If that's true then what prevents me from injecting malicious code? For example code that never halts and ties up resources?
Even if you don't have malicious code I could imagine people building complicated scripts that might tie up computation resources. How do you create an effective incentive scheme for miners then? You can't just use the code size because there is no relation between code size and resources required to run the code.
It looks like they try to mitigate this by taxing operations with fees: TXFEE (100x) - fee for sending a transaction
NEWCONTRACTFEE (100x) - fee for creating a new contract, not including the memory fee for each item in script code
STEPFEE (x) - fee for every computational step after than first sixteen in contract execution
MEMORYFEE (100x) - fee for adding a new item to a contract's memory, including when first creating a contract. The memory fee is the only fee that is not paid to a miner, and is refunded when memory from a contract is removed.
DATAFEE (20x) - fee for accessing or setting a contract's memory from inside that contract
EXTROFEE (40x) - fee for accessing memory from another contract inside a contract
CRYPTOFEE (20x) - fee for using any of the cryptographic operations |
|
|
|
I'm not knowledgeable about how fast GDDR5 is vs. DDR3, but I don't think it's an order of magnitude right?
About half an order of magnitude in terms of BW The main problem with vRAM tends to be latency, but it's less of an issue if you plan your implementation well |
|
|
|
|
Show Posts
