Topic: ASICS killing BTC ? - page 4. (Read 15908 times)

The number of coins created remains constant; that's what difficulty is for.  So there is indeed no surplus of bitcoin supply.

Memory is a lot cheaper than people seem to realize.  Slap some DDR3 sockets on your ASIC board and you have 16gb for like $60.  I understand that memory is more expensive than computation power, but not forever, and not nearly to the degree everyone here seems to think.  The reason ASICs don't exist for Litecoin right now is because no one sees the same profit margins because adoption is so much lower.  Litecoin is based almost PURELY on speculation, as opposed to bitcoin that actually has a small viable market.  In addition, Litecoin is mainly mined by people who like mining but can't mine bitcoin at a profit any longer.

This. ASICS make mining harder for new people joining Bitcoin, but they also provide much more support for blockchain.

tell that to Rassah, dude is reminding me of Denzell in Training Day when Ethan Hawke walks away from him

http://www.youtube.com/watch?v=6KrNpxODiDA

people are walking away Rassah, DEAL WITH IT OR ADAPT

What are you talking about? You're putting words in my mouth, or not at all understanding what I said....

 I'm doing fine. my jalapeno cost $149 - in 7 days, I've generated just about 1 coin. I'm FAIRLY positive that it'll generate another coin at some point. So, I'm not blaming ASIC's at all on my "loss" (which hasn't occured). I'm blaming ASIC's for driving many people away from Bitcoin. I'd hardly call myself "conned". Sure, it took a lot longer than expected, but I'm certainly glad I declined my friends offer to split the cost of several GPU's back in May to run while I waited for this.... That would have been a disaster. But no, I'm not part of the problem, driving away ordinary peoples interest in Bitcoin, I fully acknowledge that.

But still, hear me out... Bitcoin without people to drive interest in it isn't going to amount to anything. If you think that network hash rate is more important than network participants, well, I would be shocked.

This is exactly what I feared when the BFL disaster began.

A fool gets conned by a group of smooth talking hucksters and pays for a golden goose laying ASIC device.  There is plenty of evidence that the hucksters have no idea how to build such a device, but greed parts the fool and his money.  Enough fools throw money at the hucksters that some real players take note and start developing their own devices.  The hucksters promise '2 weeks' every few days, and that keeps the fool dreaming of sugar plums rather than investing in alternatives.

When the real players deliver and the hucksters still can get their shoes tied, the fool's dreams of sugar plums fade slowly.  Eventually difficulty rises so much that the fool realizes gifting his money to hucksters was a huge error, and he starts counting all he he lost by trusting the hucksters.

Who does the fool blame?  Not himself, not the hucksters, he blames Bitcoin!

Let's say I'm a notorious hacking group or even a government that has backdoors into hundreds of thousands of person computers across the globe.

What's to stop me from using CPU mining to pull off a 51% attack?

ASICs - that's what.

With CPU/GPU mining obsolete, that huge network of compromised machines is worthless for a 51% attack.

I think SHA-256 is probably okay for proof of work.  Bitcoin, however, should include something other than just proof of work.  I am curious about the possibility of tying proof of work to proof of identity via private keys.  A block would have to have both a hash of a certain difficulty and be signed with a valid private key.  The hard-code of client could prohibit reject any block not signed by a valid private key, and to discourage centralization, only 4 out of the last 100 blocks could be signed by the same private key.  I just made up 4 out of 100... obviously these could be picked carefully after some statistical analysis.  We would first need to decide just how dispersed we wanted the hashing to be.

Then the real questions becomes 'what determines if a private key is valid?'.  Maybe those keys could be mined themselves, via scrypt.  Maybe it could be based on a web of trust model that somehow limited each person to just on current private key.  The private keys could wear out... they could be good for only 100 blocks total.

Of course we would have to be extremely cautions in picking a method to make sure it couldn't be easily gamed and lead to greater centralization.  There would be a market for valid private keys, and that is okay and expected.  I would hope there is someway the web of trust idea could be implemented.  Bitcoin, after all, is a service in the name of us people.  This project is not about trusting no one, but more so trusting everyone.  We trust ourselves - as a community - right now.  The web of trust could start with the core devs and valid keys would be propagated throughout the bitcoin community.  One person, one valid key.  The details of how that could be enforced I don't know yet.  But I hope some of you will be able to build on this.

I would hope that big miners who, unlike regular users, actually have a huge incentive to keep mining, would also be aware of this issue, and would base themselves in multiple countries, just to make sure they don't get taken down all at the same time. Plus I would hope that all the users will go to where all the hardware and software development is taking place, and all the hardware and software developers will go to where all the users are, and not just where all the mining takes place. (hardware as in ATMs, hardware wallets, etc) I.e. if this ASIC thing and centralization is really an issue, then people won't leave bitcoin for anything else until there is a catastrophic failure from this issue, which may turn people off from crypto-currencies completely for a long while.

I don't know, for roughly 1 BTC you too can be an ASIC Miner. I don't see a centralizing factor unless the little guys decide it just isn't worth it.

because Joe Smith the miner actually CREATED something, whether it is digital or physical, it still exists. Joe Smith the shareholder simply threw money at someone in the hope that they knew what they were doing and walked away. He isn't going to spend his coins in the slightest, just let him know when its time to cash out.

Joe the miner is the reason developers create services for coins because they know he will spend them

Joe the shareholder is the exact type of person who Bitcoin was designed to combat, a get rich quick suit who couldn't even spell Bitcoin if you gave him all three vowels

so why don't you stop worrying about offending other people's sensibilities and realize that by putting the CREATION process in the hands of few, it completely kills the whole decentralization argument.

Which is the only thing keeping Bitcoin from going belly up, it being too decentralized for world governments to try and attack.

But god damn, some of these fucking people are trying really hard to make that not true anymore

I've come to agree with the OP's assessment, Dan Kaminsky agrees with us as well.  ASICs (and in fact every generation of mining improvement including GPUs) has a centralizing effect on mining and the effects are cumulative and each the generations have come too fast for any kind of re-democratization to occur.  And the current ASICs on the market now are not remotely the end of the line, theirs is a LOT more power that can be squeezed out of smaller feature chips.

Kaminsky thinks all this centralization weakens the network against regulatory pressure because the miners are so small in number but so individually BIG that they are worth targeting (theirs also collusion potential amongst miners).  I think that long before that happens the user base will vote with it's feet and move to what ever coin is more decentralized simply based on the threat that such a thing MIGHT happen.  Currently LTC is that obvious go to coin but it will probably get ASICs some day too (and soon too if it supplants BTC).

The long term solution is probably a chain that mixes a large number of hash algorithms together so as to make ASICs quite wasteful and expensive to develop as you would need a specially designed and fabricated chip for every algorithm.  Even then it might be necessary to defensively rotate in and out some algorithms if they are targeted for ASIC development.

I agree that ASICS factories are bad to decentralization in short term, however the market is free and anyone can try do best anytime.

Crying is free too. In Portuguese we say "o choro é livre".

Way to miss the point .... here so it won't offend your sensibilities ...  If Joe Smith average AsicMiner shareholder gets x BTC in dividends and sells them how is that any different than Joe mining x BTC and selling them.

As for me not personally owning any shares ... WTF does that have to do with the price of tea in China.

Rant all you want, but it was inevitable.  Any time people see an opportunity to make money by getting ahead of the pack, they will, even if those profits are short lived.

Let me go back a few years.  I had an nVidia card that could only mine about 66 MH/s, which wasn't a long but I could earn a few coins a month, enough to offset the costs when coins were only $2-5. People then started buying 5970 and other cards and built mining specific rigs.  People like me that didn't believe at the time that spending a several hundred dollars on a dedicated miner was worth it, dropped out and stopped mining.  The mining started going to the guys building the dedicated GPU rigs with multiple cards, often buying from craigslist or ebay to snatch up as many cheap cards as they could.

Along comes ASIC, now you have people spending money upgrading to ASICs while those with the GPU mining can't compete and are getting pushed out of the market.

It's the same game.  I have an Avalon, but I know that within 6-9 months, the difficultly will be so high that I probably won't make enough to offset the cost of electricity.  Once I reach that point, my options are to either spend my saved up coins on a better, more efficient rig, or drop out.

So even though ASICs are taking over, a year from now, there will be Gen 2 ASICs that are twice as fast using half the power and will be much more profitable.  Join the crowd or get out of the way.

Simple. I ordered my ASIC out of greed nearly a year ago. Running the numbers and salivating just like everyone else. In the intervening period, seeing what's happened and extrapolating what will happen I think it's clear Asics are a net-negative to the currency. Sure, the network has many more hashes than it had a year ago. But that's not what's important. What it's starting to lack, and what will never come back, is the excitement that everyone had about bitcoin when it was something that everyone could participate in, to varying extents.

Put another way, bitcoin would be stronger, more valuable to everyone with 1/10 the network hashing power but 100x the miners, than this consolidation we have now. When the only investment required was a graphics card that many people already had, they participated and were excited about it. When the network devolves to exclude them, when its left with a few pillars of power (asicminer, a hand full of Avalon owners and a handful of mini rig owners), all of which require significant investment, the outside world will shrug their shoulders amd look on. As a consumer, for instance, I know that I have not been clamouring for an e-currency with irreversible transactions to send to vendors who may or may not be cooperative if there's an issue with their product. And I certainly don't wish for the means for my friends to send me money at no charge, only for myself to have to be hit with a wire fee if I need that money as cash. Especially not when they could just hand me cash or a check.

I very much doubt many other consumers wish for that either. But when they can at least participate, they then get excited about it and talk up those features. Don't believe me, that's fine. But that's my stance.  In the meantime, I'll continue on in my hashing, knowing or feeling like this is a dead end.

Asic's shouldn't be compared to what the GPUs did to BTC because of it's pricing, availability and the massive jump in hashpower they bring.

I'm getting tired of this..  "everything is ok now... see everyone can buy them... " garbage !

@leopard2 Those large numbers of households will never see a return on their investment, in Bitcoins. Notice I did not mention f i a t. (pardon my swearing)

The story that this will bring stability to the network is BS, and will only help to fatten the hoarders. 1BTC .. F that !

HT xD

and over time ASIC hash rates will go up and the cost for them will go down. I'm sure if you look through the ancient posts you'll find the same complaints and statements from people who couldn't wrap their minds around big expensive GPUs.

some people here need more sleep, apparently

why makes monopolization of mining power, BTC less peer-to-peer? Gold is perfectly peer-to-peer even though most people don't wash it themselves...

Who mines and with what is irrelevant, as long as no entity is over 51%.

The peer-to-peer part, in case you have forgotten, is not in the mining but in the usage.

And again it is very well possible that ASIC technology results in a larger number of households mining. Asics are much smaller, cooler and easier to run than GPU miners.

BTC
No offence taken lol.. I thought the deleted part was easy to understand.

Who knows what will happen. That depends on how many are willing to pile on, and how low the gpu : node ratio will affect Bitcoind.  1:1 is ideal and means that we are not required to divert all of our hashpower for this to have a significant effect.

Go ahead.. keep mining as you were. All I'm asking for is, a minimum of, 1 piece of mining hardware and 1 Bitciond node per person.

HT xD

p.s. Scrypt is asic proof.. for now...