Topic: [Alternate cryptocurrencies]Litecoin (Read 4190 times)

I was wondering about the economics of Litecoin...

When Bitcoin's value dropped because of the People's Bank of China's attempt to make buying difficult, the value of Litecoin also dropped. How closely are the prices of Bitcoin and Litecoin connected? And if they are seperate currencies, why should their values be connected?

I guess so.  There was an announcement somewhere along the lines of "Dear botnet owner, the servers can't handle all the traffic from your bots, so I have to block you." so that would be pretty misleading.  Or maybe that wasn't BTCG.  Then a couple days later I couldn't get any CPU miners to connect to BTCG and went into the IRC channel to ask what's up and the admin said they're all blocked on his pool, because of botnets and stale shares. 

I think you are confused.  It wasn't botnet mining that brought down BTC Guild.  BTC Guild banned a botnet and then they ATTACKED (not mined but launched a syn flood attack) the major pools in retribution.  The fact that BTC Guild kicked them off indicates that prior to being kicked off they were mining just fine on the pool. 

Well, you basically just confirmed what I was saying on the first point.  Not sure how they'd limit it to 40% CPU or something though.  I would like a way to do that on certain things.

I was under the impression that it'd be prohibitively expensive to get a server for a private pool with ~250k workers.  And that that many workers will kill any of the public pools, even with their multiple servers.  BTC Guild had that problem.  Somebody was obviously using a botnet and it DDoSed them, so now all CPU clients are banned.  I also assumed it'd be damn near impossible to manage 250k computers using a bunch of different pools.  I don't know.  I haven't read about how botnet works since back when they were controlled via private IRC channels, heh.

No.

1) lower priority @ 100% load is still 100% load.  A smart botnet controller would have the zombies hash at say 40% load which wouldn't even cause the CPU fan to spin to higher speed.

2) They don't need to solo mine.  They can either join a pool (maybe Bitcoin pools have thrown botnetters out of the pool once caught) or they could form their own "private pool".  It wouldn't DDOS them since the communications would be routine not intentional attacks.  If the botnet was too large for a single pool they could use multiple "subpools" either public or private.

So there is no problem with mining using a botnet.  With GPU based chains the limited power of botnet means the profits and options for blockchain attacks are limited.  Against a GPU hostile chain very easy for a botnet to be a significant fraction of the hashing power. 

As mentioned elsewhere, there is a problem with mining with a botnet.  People start to suspect things are wrong with their computer when it starts slowing down like it would if it was using 100% CPU to mine.  I guess if they could restrict it to only use a certain amount of CPU it would be less detectable.  I have noticeable lag even with it set on lowest priority using all cores of my 2600k. 

Also, wouldn't it take a long time for a bunch of small computers to solo mine?  And if using a pool, wouldn't it effectively DDoS the pool, like what happened with BTC Guild?

You can think up all the scenarios you want, but history in economics has shown that factors never previously considered often show up and change the game.  That's the only thing I can predict with litecoin.

Well the rate of LTC (or BTC or xCoin) generation is fixed.  More miners don't mean more LTC.  That is why blockchains have higher difficulty.

Of course the reverse is the problem with robocops logic.   More miners means it is harder to get a single LTC (as the supply is fixed).  Thus if the network is 500x as large then it would be 500x harder to 51% but it also means the average miner would get 1/500th as many LTC.

I don´t understand the logic because if it is easier to mine litecoins then the number of litecoins will growing up more quickly -> price is going down -> the interest of litecoins decrease.

The bitcoin mining is better because big organizations and companies can´t mine with their powerful cpu-driven hardware and the freaks with their watercooled GPUs are in advantage and it would be an bad business for companies to upgrade their hardware and buy powerful gpu-driven hardware.

In my optinion the GPU preferation make sense and is important to limit the influence of commercial companies like google or some big financial institutes.

And the performance of litecoin or other bitcoin similar currencies show that the interest and acceptance is very low.
And by the way it is not comparable with the bitcoin self because as the bitcoin was born the community was a lot of smaller than today.
But the litecoin has these community here for advertising and nevertheless the price don´t increase.

Lower priority of minerd 

Hello,

I'm currently trying out the Litecoin miner. It seems it's a little too aggressive, can the aggression be tinkered with somehow?

I can't make full use of my GPUs while the Litecoin CPU miner's on. This restricts its usefulness very much.

A botnet can attack only with it's CPU, because usually infected pc doesn't have a proper gpu or, even if they do, maybe they doesn't have the right drivers or what else, while every computer has a cpu.

Now, if you attack Bitcoin, it's pretty useless, a few GPUs have enough power to stop a whole botnet.

If you attack litecoin or another cpu coin, it will be very effective since ALL miners are cpu miners.

Aham. Thanks for clearing this out 

LTC is young crypto currency - time will tell is it good enough for market.
I don't want to invest in AMD rig so I'm sticking with LTC 

It doesn't have to be in first 6 hours.  The block chain won't be locked again for 3.5 days.  So attacker spending starting hour 7 and then initiates building private block chain and performs double spend once it has larger chain than the legit chain.   If it was that simple to "solve" the 51% dilemma with token effort it would already be solved.

Looks BTC isn't immune to a 51% attack either however the massive size and the "multiplier" by using more efficient hardware (GPU) makes that attack many magnitudes more difficult.

OK, you can forbid business transactions bigger than 1k in the first 6hrs after difficulty change right?
I'm not saying that LTC can substitute BTC but for fast small transactions it is ideal solution.

And yes, LTC network has potential to grow that big, because if you spent $0 on your mining rig, its not a problem to have just $10/week (+electricity)

You think LTC will be not bigger than BTC or double the size of BTC but 100x the size of BTC (and even then it woulnd't be as resistent to botnets)?  

Your right if someday LTC has 160x as many miners as BTC then it would have equivelent resistence to botnets and other commodity malicious CPU power.

Of course with that many miners the block rewards get split to almost nothing per miner.  Transaction volume would need to be hundreds of millions of dollars per day in order for fees to be sufficient to support the hundreds of thousands of miners that would require.

You are technically right.  If LTC had hundreds of millions of dollars in transaction volume it would support a mining/hashing network of hundreds of thousands of miners and that would be sufficient to be resistent today's botnets.


No you can't. That is a fallacy.  After difficulty change an attacker can work on "bad chain" in private.  This applies to any block chain.  You would have no knwledge it is happening until the "bad chain" is 2 to 3 blocks longer than good chain and is published to the network.  It would invalidate the good chain instantly and overwrite any transactions since bad chain began.

Youre right but:

You said that now it takes 50 computers to overpower one good BTC node.
For LTC is now 3:1 but if LTC network was 100x as large it would be 300 computers to overpower one good LTC node.

I expecting more miners on LTC in future than on BTC because everyone has CPU but not AMD rigs.

And, you can stop any transactions in time of 51% attack. How much will attackers have patience after 2-3 fails?

I think you miss the point it isn't the nominal value it is the relative value.

Say average botnet computer is 10MH/s (generous) on Bitcoin network.
The average "good node" based on btc-pool watch is 500MH/s.

Thus it takes 50 computers to overpower one good node.  Even if Bitcoin was 10% of its current size that would be a challenge for a botnet.

Not if average high end CPU LTC node gets ~15KH/s lets say average Botnet computers are 1/3 as powerful, 5KH/s.  It only takes 3 computers to overpower one average good node on LTC network.  Even if LTC network was 10x, even 100x as large it would still be vulnerable to botnets.

It will always be vulnerable to botnets because the playing field is level and botnets can amasse huge amounts of computing power very cheaply.  The decision to make a chain GPU-hostile is dubious as best.  That vulnerability will become more practical if/when LTC ever begins significant transaction volume.  More volume, more value = more potential gain in any attack.

Block locking won't stop a double spend attack.  3.5 days is an eternity to double spend which only involves spending large amount of coins, performing 51% attack, spending coins again and sustaining 51% attack for 6 confirmations.  Completing double spend on an exchange (tranfering LTC -> BTC and then withdraw) would take only hours not days.

Specialized hardware HELPS blockchains by making it harder for attackers to acquire efficient attack nodes.
Commodity hardware HELP attackers by making it easier to achieve 51% of network power cheaply.

Are you aware of block locking at every difficulty change (every 3.5 days)

Rogue system admins can loose their job, EC2 instances are not so cheap (even now you need about $200/hr)

And tell me, how much hashing power did BTC network have on the first month of its live?

I wouldn't say Bitcoin has "solved" botnets but Bitcoin is botnet resistant due to its massive hasing power.  Average CPU has say 10MH/s on Bitcoin algorithm (probably less than that when you consider some botnet nodes are entry level and older systems).  Thus 8TH/s is at least 800,000 CPU.  The low performance of the average botnet node (due to specialized hardware like high end GPU) and massive size of Bitcoin network makes a botnet attack improbable but I wouldn't go as far as to say Bitcoin "solved it".

GPU-hostile blockchains make the playing field between average "good node" and average "bad node" much closer.  Achieving parity with an enemy that outnumbers you is likely a bad decision.   Simply put it is far easier for bad guys to amass large amounts of CPU power for "cheap".  Not just botnets, but rogue system admins, EC2 instances, etc.

How Bitcoin solved problem of botnets then?
I know that they were on Bitcoin network.