Topic: did ASIC ruin bitcoin ? (Read 8993 times)
Technology accelerates change. It seems like, of all communities, the bitcoin community should OK with that.
Some well thought out replies. I would add that the rest of technological progress will likely be pretty dynamic in the coming few decades: if we're only just at the stage where open source firearms are being developed, what happens when electronic devices start becoming homemade? The potential for someone to print their own mining rig suddenly throws up limits around the physical resources needed and the availability of the design knowledge or availability of designs. I realise we're nowhere close to printing chips right now, but I think anyone who follows the semiconductor manufacturing industry will know: they're running out of road, and on the timescales I'm talking about, too. Who will innovate the next processor fabrication technology: Intel, TSMC, or some Kazakh carbon-substrate 3D printing enthusiast, living far away from prying eyes? You can't put that particular brand of toothpaste back in the tube...
Well put, that's exactly why I stick around. I believe this ecosystem is actually top dog in the open source multiverse (had to use it
). Hell I would even say we have preppers beat. Locking down all raw commodities is where the control system logically moves next, I'm pretty sure of this. Electricity, glue, wood, water, metals, seeds, plastics (well, they got the crude oil pretty well wrapped up already, in fairness). How much good is digital hard money and 3D printers when price controls on materials force us into 21st century Dickensian pauper-dom? I might be overstating the extent to which all commodities could be controlled, but there are signs they've already started (ETF's for copper and no doubt other surprising ETF's, Blythe Masters on her "World Tour" since 2008...)
When we get EEPROM ASICs that can switch coins and crypto algos, deep into the micron chip limit, then we will have saved our project from those that would see it fail.
Whosoever can get there first, gets a hero medal from yours truly.
Whosoever can get there first, gets a hero medal from yours truly.
You mean programmable logic, like FPGAs? I think the idea actually grew out of memory chips - you can implement basic logic by having memory addresses as the input, and memory contents as the output.
It's a nice technology to have, as long as someone can actually program the requisite logic - we still don't have a Scrypt miner on an FPGA. Also, good luck developing a Scrypt ASIC without first prototyping it on an FPGA.
OTOH, if you had the ASIC tech for several different hashes, you could in theory have all of them on one chip, but why would you switch between them, instead of running them all at once?
Yes, FPGA are the EEPROMs for this. Add the coin switching to follow the most profitable mining of the moment, optimize, and away we go! Doing this on ASIC doesn't provide the flexibility for dynamic changing so you don't also get the monetary optimization which requires external market input. What you gain in speed and heat efficiency, you may lose in flexibility.
Though with FPGAs advanced to a sufficient optimization for better ASIC development, that may not be so important.
They were only profitable when they first came out
If you are commodity trading you could not care less, ASIC did not change anything to first order. However I think ASICs are bad for the currency functions.
With CPU and GPU mining anybody with an hour to read up on the topic could participate, and by word of mouth pull other users in. Let's face it, nobody is going to make the required investment in a new miner without a reasonable ROI prospect, so new miners are those that already have the hardware.
In a sense it is similar to real gold; to make a profit you need to be a large company, but this makes gold not easily accessible to consumers and thus impractical as a currency, while it is still fully functional as a commodity. Due to the lack of industrial applications of BTC (as opposed to gold) in the long run it must succeed as a currency. With small scale hobby miners quitting there is less talk about BTC, leading to less PR, which could drive BTC price down, and thereby also affect traders. To save the situation we would need much improved capabilities to pay using BTC in our everyday lives.
With CPU and GPU mining anybody with an hour to read up on the topic could participate, and by word of mouth pull other users in. Let's face it, nobody is going to make the required investment in a new miner without a reasonable ROI prospect, so new miners are those that already have the hardware.
In a sense it is similar to real gold; to make a profit you need to be a large company, but this makes gold not easily accessible to consumers and thus impractical as a currency, while it is still fully functional as a commodity. Due to the lack of industrial applications of BTC (as opposed to gold) in the long run it must succeed as a currency. With small scale hobby miners quitting there is less talk about BTC, leading to less PR, which could drive BTC price down, and thereby also affect traders. To save the situation we would need much improved capabilities to pay using BTC in our everyday lives.
When we get EEPROM ASICs that can switch coins and crypto algos, deep into the micron chip limit, then we will have saved our project from those that would see it fail.
Whosoever can get there first, gets a hero medal from yours truly.
Whosoever can get there first, gets a hero medal from yours truly.
You mean programmable logic, like FPGAs? I think the idea actually grew out of memory chips - you can implement basic logic by having memory addresses as the input, and memory contents as the output.
It's a nice technology to have, as long as someone can actually program the requisite logic - we still don't have a Scrypt miner on an FPGA. Also, good luck developing a Scrypt ASIC without first prototyping it on an FPGA.
OTOH, if you had the ASIC tech for several different hashes, you could in theory have all of them on one chip, but why would you switch between them, instead of running them all at once?
ASIC has made Bitcoin much more secure. This means that to perform a double-spend attack on the network would require more mining power than the majority of other miners combined, which would be very hard and insanely expensive.
that may be the problem with this wave is that a lot of people will get disillusioned that they got taken for a ride by the earlier adopters and quit or start calling cops and lawyers.
Cops and lawyers? Taken for a ride? Huh?!?
It wasn't clear to me that you were talking about BFL rather than BTC in general. I get what you were saying now. Thank you for clarifying.
that may be the problem with this wave is that a lot of people will get disillusioned that they got taken for a ride by the earlier adopters and quit or start calling cops and lawyers.
Cops and lawyers? Taken for a ride? Huh?!?
... there were a lot more fun ways to participate. I've had tons of fun learning to code FPGAs, and some might even consider it a skill worth money. I've generally only bought hardware that the user could program themselves, so that rules out ASICs - as you said, there are plenty of other uses for GPUs, and the same goes for FPGAs.
Of course, there have to be people who develop stuff, instead of investing, so this kind of fun aspect is actually quite important even from the financial POV.
Some of the altcoins have rekindled the fun to some extent, notably Primecoin from the most recent ones. It's a little sad that a lot of the talk around Primecoin is about hosted mining and profitability, while the math itself is so exciting.
Of course, there have to be people who develop stuff, instead of investing, so this kind of fun aspect is actually quite important even from the financial POV.
Some of the altcoins have rekindled the fun to some extent, notably Primecoin from the most recent ones. It's a little sad that a lot of the talk around Primecoin is about hosted mining and profitability, while the math itself is so exciting.
Here's a hidden gem for those who have read this far into the thread and can parse the lingo:
When we get EEPROM ASICs that can switch coins and crypto algos, deep into the micron chip limit, then we will have saved our project from those that would see it fail.
Whosoever can get there first, gets a hero medal from yours truly.
Did CPUs ruin bitcoin?
Now take your answer and apply it to ASIC.
+1 Now take your answer and apply it to ASIC.
nothing "ruined" bitcoin in my opinion it is still at its very beginning
Block Erupters aren't very economical pieces of kits, you'd need a good number to keep up with the difficulty rate
If one is not profitable, it's unlikely more-than-one will magically transform into being profitable - simple maths ...They make great gifts, good GPU replacements for hobby-miners and have blinking lights - what's not to like
The space race is fucking over. Hype has fallen and yes BTC more then likely needs a fix. 2 years of speculation lead to a ton of improvements on GPU and FPGA Mineing. For some it was an education. Myself included. Now that we really don't have much to tweak on we are bored and restless LOL.
It's an arms race now. No real fun in that. No real education aside from perhaps understanding ASIC Tech. I pray daily somebody will find a homebrew method to allowing one to expand on this hardware they have. But thay isn't going to happen. I for one am going to continue keeping one or two gpu rigs around. So much can be done aside from ET Search and PW Cracking.
For me it's not about money. This hosted mineing shit with 4737474GigaNanoUberBits of hash powet per person. Thats more money then brains
It's an arms race now. No real fun in that. No real education aside from perhaps understanding ASIC Tech. I pray daily somebody will find a homebrew method to allowing one to expand on this hardware they have. But thay isn't going to happen. I for one am going to continue keeping one or two gpu rigs around. So much can be done aside from ET Search and PW Cracking.
For me it's not about money. This hosted mineing shit with 4737474GigaNanoUberBits of hash powet per person. Thats more money then brains
Agreed. Looking back, the best financial investment would have been to buy lots of BTC, but there were a lot more fun ways to participate. I've had tons of fun learning to code FPGAs, and some might even consider it a skill worth money. I've generally only bought hardware that the user could program themselves, so that rules out ASICs - as you said, there are plenty of other uses for GPUs, and the same goes for FPGAs.
Of course, there have to be people who develop stuff, instead of investing, so this kind of fun aspect is actually quite important even from the financial POV.
Some of the altcoins have rekindled the fun to some extent, notably Primecoin from the most recent ones. It's a little sad that a lot of the talk around Primecoin is about hosted mining and profitability, while the math itself is so exciting.
Some well thought out replies. I would add that the rest of technological progress will likely be pretty dynamic in the coming few decades: if we're only just at the stage where open source firearms are being developed, what happens when electronic devices start becoming homemade? The potential for someone to print their own mining rig suddenly throws up limits around the physical resources needed and the availability of the design knowledge or availability of designs. I realise we're nowhere close to printing chips right now, but I think anyone who follows the semiconductor manufacturing industry will know: they're running out of road, and on the timescales I'm talking about, too. Who will innovate the next processor fabrication technology: Intel, TSMC, or some Kazakh carbon-substrate 3D printing enthusiast, living far away from prying eyes? You can't put that particular brand of toothpaste back in the tube...
Well put, that's exactly why I stick around. I believe this ecosystem is actually top dog in the open source multiverse (had to use it
). Hell I would even say we have preppers beat. 
Comparing investments against one another is always wise; I recommend diversity. I, for one, haven't put a penny into mining yet but am utterly delighted to have other folks doing so; best o' luck; break a leg.
If the only currency is Bitcoin then taking the future value of Bitcoins into account is not relevant to determining the wisdom of mining as an investment. Since there are other currencies then taking the exchange rate into account is relevant. Then again I'm sure I will be told if the exchange rate does move in favor of Bitcoin then the original investment should have skipped mining and gone directly into speculating on Bitcoin.
Mining as a charitable donation; hmm, can I deduct them from my income on my tax return?
If the only currency is Bitcoin then taking the future value of Bitcoins into account is not relevant to determining the wisdom of mining as an investment. Since there are other currencies then taking the exchange rate into account is relevant. Then again I'm sure I will be told if the exchange rate does move in favor of Bitcoin then the original investment should have skipped mining and gone directly into speculating on Bitcoin.
Mining as a charitable donation; hmm, can I deduct them from my income on my tax return?
You could structure it as a single owner business and then deduct the resulting net business loss on your tax return. Businesses which continually lose money are still businesses, the IRS just looks for a profit potential.
Comparing investments against one another is always wise; I recommend diversity. I, for one, haven't put a penny into mining yet but am utterly delighted to have other folks doing so; best o' luck; break a leg.
If the only currency is Bitcoin then taking the future value of Bitcoins into account is not relevant to determining the wisdom of mining as an investment. Since there are other currencies then taking the exchange rate into account is relevant. Then again I'm sure I will be told if the exchange rate does move in favor of Bitcoin then the original investment should have skipped mining and gone directly into speculating on Bitcoin.
Mining as a charitable donation; hmm, can I deduct them from my income on my tax return?
If the only currency is Bitcoin then taking the future value of Bitcoins into account is not relevant to determining the wisdom of mining as an investment. Since there are other currencies then taking the exchange rate into account is relevant. Then again I'm sure I will be told if the exchange rate does move in favor of Bitcoin then the original investment should have skipped mining and gone directly into speculating on Bitcoin.
Mining as a charitable donation; hmm, can I deduct them from my income on my tax return?
Some well thought out replies. I would add that the rest of technological progress will likely be pretty dynamic in the coming few decades: if we're only just at the stage where open source firearms are being developed, what happens when electronic devices start becoming homemade? The potential for someone to print their own mining rig suddenly throws up limits around the physical resources needed and the availability of the design knowledge or availability of designs. I realise we're nowhere close to printing chips right now, but I think anyone who follows the semiconductor manufacturing industry will know: they're running out of road, and on the timescales I'm talking about, too. Who will innovate the next processor fabrication technology: Intel, TSMC, or some Kazakh carbon-substrate 3D printing enthusiast, living far away from prying eyes? You can't put that particular brand of toothpaste back in the tube...
Excuse me if I don't read all 6 pages of this thread, but did anyone handle the "in a decade or two, BTC price increases so much + advanced mining hardware becomes so expensive that it basically gets taken over by corporations" scenario? I can think of a few objections, just thought I'd put it out there if it's not already been handled.
If everything goes right, corporations will have a huge stake in Bitcoin. I suspect that in a decade or two, the only real money to be made from mining will be the transaction fees. In 10 years we will be getting ready for the block reward to drop to 3.125 BTC and in 20 years we will have just passed the reward halving to 0.78125. It's likely by then that the large bulk of the network will be large companies handling and securing transactions and collecting transaction fees. The block rewards will still be a nice bonus but I would guess that an industry will grow (is already growing) around Bitcoin and mining won't necessarily be something everyone does.
But whether corporations can "take over" Bitcoin is the real question and i think we've gone beyond that point. The people and/or organizations that are willing to put the most time, energy, ingenuity, and money into it will have the biggest stake. But collecting transaction fees, while it could be a very large industry, is necessarily going to be much smaller than the set of industries sending and receiving that money.
So yeah, it won't be something that hobbyists will get rich off, but it's hard to imagine that it would ever be something that hobbyists could no longer participate in. I'm personally looking forward to the day when I might get a chance to help a kid build a mining rig for an elementary school science fair.
If in 20 years most of the network is still running on the mining efforts of amateurs, then it will be fair to say that Bitcoin had failed.
Excuse me if I don't read all 6 pages of this thread, but did anyone handle the "in a decade or two, BTC price increases so much + advanced mining hardware becomes so expensive that it basically gets taken over by corporations" scenario?
Bitcoin mining is essentially a problem which is almost perfectly parallelized, there are no significant global restrictions, and the barrier to entry is non-existent. What you are seeing today is the gold rush phase but it will fade. Hardware will become more available, and as NRE get paid off and inventories build the price per GH/s will only fall. This means that there is no realistic scenario where a small miner can't compete on with big miners (especially when you consider those with reduced or so called "free" power).
For example say a 1 PH/s miner someday costs $1000x well a 1 TH/s miner would probably be ~$x. Yes larger units are marginally cheaper per GH/s but not to a point that it would make smaller units unviable. Sure if you have 1/1000th the hashing power of a big entity, then your reward is only 1/1000th but your capital and electrical costs are ~1/1000th as well. People look at the USB Block Eruptors says it is a "bad" deal and assume that means small units will cost too much. The reality is the price of miners right now has little to do with cost and more to due with demand. Friedcat sells out of every batch of BE. As a merchant the best possible price is the highest price where you still continually sell out. Any higher and you are stuck with inventory, any lower and you still run out you just makes less. As demand falls so will the price (it already has to some extent from 2 BTC to 0.3 BTC). Eventually the margins on these small miners will be very low and they will be produced in hundred thousand lot units. I could see entities like ASICMiner getting out of the retail business all together instead they mass produce chips and sell them to OEM who build the end product (the AMD/NVidia model). USB Miners will lose their "premium" relative to larger units as they become more available and there is more competitions.
Mining isn't really that attractive for corporations, most wouldn't want to touch Bitcoin mining with a 10 foot pole. Bitcoin mining is a somewhat unique industry where your profitability matters less on what you do but on what other people (you can't control) do. For example a lot (maybe the majority) of early ASIC miners (orders in 2012 and 2013) will have negative net profits. This is because you can do everything "right" (order when difficulty is low, pick the right supplier, have low electrical costs, buy the most efficient gear, etc) and still lose due to tens of thousands of clueless noobs who combined deploy so much hashing power that the revenue for everyone falls. You can't stop them from doing that which means a big entity can't protect their margins. Sure they can optimize cost but price spikes, a bunch of noobs think they will become millionaires in a month buy a 100 PH/s and then price falls. Everyone is stuck in the same low/no profitability boat. That scenario is a nightmare for any CEO. Period. It isn't something anyone with a lot of capital will be dreaming about.
Maybe I am an optimistic but I don't see it being an issue. Mining will eventually be a very boring, low profit business. Many hobyist will mine hoping to break even as an alternative way to acquire coins (essentially buying them from the power company). I would be far more worried about centralization in the financial services space to a tiny handful of massive global players. By financial services I mean bitcoin dealers, brokers, exchanges, merchant processors, etc. This is more of a risk because government will put up barriers to entry and those that can afford those costs will have a vested interest to see those costs become larger to kill off any future competitors. This has already started to happen in the sense of nonsensical high cost regulation and one should only expect it to accelerate as Bitcoin becomes a bigger deal. Luckily that isn't a core aspect of Bitcoin itself but it is something to closely watch.
Jump to: