rpietila Altcoin Observer

[dewdeded]

Ahh ok. And where you know that from?

[r0ach]

Ahh ok. And where you know that from?

He knows it because most 5 star posters know that L3 cache latency is not an ASIC or GPU deterrent.  In terms of fiddling with memory to try and stop or slow down GPU/ASIC, increasing the raw amount needed is the only viable method, and even that is just semi-effective.  It depends how high the coin's market cap goes before there is enough incentive to try.  Someone who has in-depth knowledge of ASIC manufacturing could probably work out a calculation that says, "the vertcoin algorithm is effective approximately up to X million or billion dollars market cap"

[drawingthesun]

Just wanted to chime in that Monero is NOT CPU-only, if it hasn't already been covered. GPU MRO mining software just hasn't yet been developed publicly yet.

I just want to point out my views on this particular point.

If you read the CryptoNote white paper they discuss how the focus on being CPU only is a good thing, as essentially the mining is like voting, voting for transaction order, for things to be added into the coin, etc...

I believe this is a flawed premise based on the existence of cloud services and botnets.

The cloud one is particularly important as it simply states that whoever can buy enough instances for that time (that a feature gets added?) they decide the final fate.

I also believe ASIC mining is of a negative nature because it concentrates power into several manufacturing bodies and they always sell the miners above the cost they run their own farms.

I think that GPU mining is the best mining paradigm at the moment. Perhaps the idea of the proof of work being useful that also isn't gamed could also be worthwhile.

The reason I replied here is that, a lot of people might panic when the first Monero miner comes along, I say this: don't panic, it's for the best.

The only way the GPU one becomes questionable is if the GPU manufacturers start their own farms for below the cost of retail. I hope that never happens. (If someone has a good argument why that won't happen I will be happy to hear it.)

[vokain]

I think Monero has a real future.  All the same, I'd love to see someone combine the privacy of Cryptonote with GPU mining.  Botnets are a real problem for CPU mining, and ASICs are a real problem because they centralize the hashrate among a few players and don't help small time miners who eventually become users of the currency.  GPUs are really the best compromise.

Botnets shouldn't be a long term threat because if CN became widespread enough and actually stayed CPU-only, then the price of botnets would rise to the market price of their computing power. In addition, botnets have a shrinkingly finite supply, especially because as more people mine, there are less potential CPUs to be a part of botnets.

In regards to cloud-scale computing, they'll quickly be mining at a loss if enough everyday miners (ie those with a computer at home) mine at a loss since they probably won't worry as much about their relatively small electric bill vs active EC2, et alia costs. I probably mine at a loss now (given current exchange rates) but I don't mind the extra few dollars per month on my electricity bill with my otherwise underutilized processor and I would be more willing to risk just that given the potential end game than increasing my risk moreso by renting cloud computing.

[aminorex]

GPU accelleration of some part of the cryptonight hash is a certainty.  Already, CPUs with AES support are a massive win, relative to other CPUs.

[Zer0Sum]

some of you say that Litecoin wont succeed because it failed the ASIC proof claim !!!!... where and when the Devs or Coblee or any big investor pointed to litecoin as an ASIC "proof" ? they always said "resistant" which was important for the coin survival at the time....there is nothing ASIC proof, it is not if we can make ASIC for this coin or that coin but when is it profitable to make a one, and when it is made it means that the coin is going mainstream.


no one gives a fuck about the hashing algorithm except miners, investors/adopters care only about the utility and opportunities, market cap,development, services.... and dont give a shit about that hashing algorithm and pure technical stuff, and what matters to me as an investor is for the last 6 months Litecoin is really going crazy, take a look at the daily transactions volume and the trading volume against Bitcoin.

Remember, from looking at the history of litecoin price it is always lagging comparing to Bitcoin, but it always make a boom, and usually a bigger one.

I like this fucking post!!!!!

People are always down on LTC...
Because when priced in BTC it always looks like a laggard.

Since I, like most functional people, pay 99% of my bills in USD or equivalent major currency...
Let's look at commodities like LTC versus BTC in USD... you know, the green stuff that keeps you on the street:

.......... Jan 2013.....Apr 2013...... May 2014

BTC...... $16.000..... $145.00........ $600.00...... up 3,750%     
LTC........ $0.065........ $3.35......... $11.00...... up 16,900%

Wow, that's not even the interesting part...
LTC returns are magnified even more by switching to LTC when BTC is strong and vice versa...
Like right now LTC is "dead", "pointless", "asic fail", etc = time to buy LTC.

LTC is extremely important in terms of providing much needed Crypto Universe liquidity...
It's not about the hashing algo or specs like the dude said.

[aminorex]

Once someone has designed a chip it is extremely expensive to make any changes to the PoW function, but software solutions (CPU and/or GPU) can be tweaked a little at almost no cost.

Have the algorithm change pseudo-randomly, to another hash algorithm, depending on the time at which hash rate crosses a threshold.  If the space of hash algorithms is large enough, that pretty much destroys ASIC, because of the large amount of die space required to hold all the recombinant blocks, and the solution is fully decentralized.

[aminorex]

My question is what will stop the miners from dumping?

Hopefully, nothing.  More coins for me, and I don't have to pay EC2 bills.  (Actually, I mine QCN and roll it into MRO right now, but that free money is likely to dry up pretty quickly.)

[damiano]

Buy support for this is incredible.  That wall 

[smooth]

Once someone has designed a chip it is extremely expensive to make any changes to the PoW function, but software solutions (CPU and/or GPU) can be tweaked a little at almost no cost.

Have the algorithm change pseudo-randomly, to another hash algorithm, depending on the time at which hash rate crosses a threshold.  If the space of hash algorithms is large enough, that pretty much destroys ASIC, because of the large amount of die space required to hold all the recombinant blocks, and the solution is fully decentralized.

Somewhat. There is an argument (undemonstrated AFAIK) that an ASIC for say X11 is still viable because the space of primitives used by the various hash functions is pretty small, allowing for a programmable ASIC. Something like what has been proposed for Etherium where the PoW has to run actual (otherwise useful) programs may be better, but this is all theory.

[aminorex]

Anonymous currency ... as the amounts grow larger, the risks aren't worth it.  If Alice has financial dealings with Bob but genuinely doesn't know and cannot discover who Bob is, then Bob will occasionally rip off Alice secure in the knowledge that nothing can be proven.  As this happens more often, Alice will learn to refuse to do business anonymously.  Once you're talking about value that is worth the effort to rip off, you're talking about a negative value proposition for 'hard' financial anonymity. ...  once they're into transactions big enough that they're afraid of getting ripped off?  Then they want to be dealing in assets that the police can trace and the courts can recover.

Transacting with an anonymous party requires proofs.  It's complicated, and it's only getting started -- it is a new thing.  It's not the primary use case for an anonymous currency.  The primary use case is transacting with friends, family, business associates, or in-person transactions. Outside of that circle, the first large area is escrowed transactions.  More advanced forms of transaction with embedded proof will be a burgeoning area of economic development at some point in the future, but it is not worth discussing them now.  They are a red herring, for the time being.

Courts are perfectly capable of recovering crypto, whether it is anonymous or not.  The standards of proof will, of course, be higher initially, and then on par with cash transactions thereafter.

[smooth]

In regards to cloud-scale computing, they'll quickly be mining at a loss if enough everyday miners (ie those with a computer at home) mine at a loss since they probably won't worry as much about their relatively small electric bill vs active EC2, et alia costs. I probably mine at a loss now (given current exchange rates) but I don't mind the extra few dollars per month on my electricity bill with my otherwise underutilized processor and I would be more willing to risk just that given the potential end game than increasing my risk moreso by renting cloud computing.

This. Run the numbers, its pretty easy to see. Cloud mining is not competitive with small scale home mining. It only works when a coin is new and unknown and home miners are small in number. If a coin gets popular enough and is easy to mine, cloud mining is not remotely viable.

This shouldn't really be surprising because of you look around at what happens in IT, cloud is often replaced once a startup reaches the scale for it. There are a few exceptions that are probably strategic in some ways that are unclear to me (e.g. netflix).

The exception being the subset of cloud mining that runs on fraudulently-obtained resources (promotion abuse, stolen credit cards, etc.) but that is basically a botnet.

[smooth]

Ahh ok. And where you know that from?

He knows it because most 5 star posters know that L3 cache latency is not an ASIC or GPU deterrent.

It most certainly is a GPU deterrent to some extent. Whether there a clever work around remains to be seen. Even if there is, the GPU factor will likely be low and still may not be economically viable (compared to near-zero-cost CPU mining).

It is less clear whether it is an ASIC deterrent. It is clearly not at some (potentially impossible) market cap, but the actual market conditions where it makes sense to try to compete with CPU R&D are not clear. Your point about putting actual numbers on this is valid.

[aminorex]

tax dodgers

extortion evaders would be my preferred term because it is more accurate.  involuntary taxation can be a form of extortion.  i am not here to debate the boundaries.  anonymous crypto makes drawing those lines a responsibility of the owner.  but there are other forms of extortion, such as kidnapping, threats of arson or bodily harm, originated by a non-state actor, which are no less desirable to defend against.  i really don't care whether the person with superior arms who is threatening me is a member of a national security organization or a freelancer.  my concern is not exposing my children to abduction, torture.

[AlexGR]

LTC is extremely important in terms of providing much needed Crypto Universe liquidity...

What does that even mean?

[darlidada]

Anyone betting on LTC to rebounce hard ? It seems we might have hit bottom at 0.017. I dont believe for a second in this coin but what do I know ? The market is stupid enough to value XC and DRK, he may be stupid enough to think litecoin value will surge as usual during a bitcoin bubble.

What does Aminorex' monkey think about that? Does he ever speculate on the idiocy of a market?

[mmitech]

Anyone betting on LTC to rebounce hard ? It seems we might have hit bottom at 0.017. I dont believe for a second in this coin but what do I know ? The market is stupid enough to value XC and DRK, he may be stupid enough to think litecoin value will surge as usual during a bitcoin bubble.

What does Aminorex' monkey think about that? Does he ever speculate on the idiocy of a market?

what do you know? you seem enough stupid to think LTC the second coin the almost old 3 years coin the most accepted, traded, used and developed coin wont surge !!!!

[AlexGR]

The market is stupid enough to value XC and DRK

What "value" does DRK have compared to LTC?

Coins number adjusted, if LTC had 4.3mn coins (same as DRK) it would cost 60-70$. And these, for what? What does LTC offer compared to DRK?

and developed coin

What exactly have they developed?

[mmitech]

The market is stupid enough to value XC and DRK

What "value" does DRK got compared to LTC?

Coins number adjusted, if LTC had 4.3mn coins (same as DRK) it would cost 60-70$. And these, for what? What does LTC offer compared to DRK?

and developed coin

What exactly have they developed?

what exactly it doesn't, everything Bitcoin have, Litecoin have first all the time... all core implementations are implemented to Litecoin then later to Bitcoin, Litecoin Dev team members always give a hand to Bitcoin developers... so how about that ?

[AlexGR]

what exactly it doesnt, everything Bitcoin have, Litecoin have first all the time...

Like what?