Bitcoin - Part 2

[ChrisBFRPKY]

What was predictable was that if Bitcoin ever 'took off', mining it would require ever-increasing processing power - and therefore ever-increasing electricity use. The knock-on effects of this were also quite predictable.

But why must increased processing power equate to increased electricity usage? Because Moore's law is dead. Chip density doubling every 2 years? Doesn't sound that dramatic, but in just a few years that geometric progression has given us computers with a million times more transistors in them. To someone like me who was playing around with computers that had only a few kilobytes of RAM, the idea that I would be using a PC with Gigabytes of RAM in it was ludicrous - and yet Moore's Law had already predicted that this would happen.

But Moore's law is failing. A modern high-end CPU or GPU is vastly more powerful than the chips used in early PC's, but also uses a lot more power. The ever-shrinking transistor size was helping to keep power consumption down because smaller transistors can operate faster without using more electricity, but it could not keep up with the insatiable demand for more processing power.

The ever increasing processing power (and therefore energy input) required to mine Bitcoins is not a bug, it's a feature which its inventor and anyone who understood how it worked was well aware off. What they were perhaps not aware of - or chose to ignore- was the scale. I suspect most early bitcoin users either didn't do the math, or were unable to comprehend it - mining Bitcoins on the their PCs and thinking, "If these things take off I might need a more powerful PC", but not considering that when the price hits $1 million due to widespread adoption they will be competing against billion dollar mining pools sucking power directly from hydro dams. But if Bitcoin was to fulfill their dreams then this development was inevitable - and totally predictable.

How widely used Bitcoin will become may not be predictable, but the amount of electricity it will use when it gets there is. Market theory tells us that the 'industry' will find ways to lower its production costs, which in this case means getting cheaper power. So 'miners' will go to locations where they can get it (ie. China). And since mining is a business not an ideology, they don't care if it's cheaper due to government subsidies or slave labor or any other 'immoral' reason (even theft, if they can get away with it).

A modern CPU or GPU uses much less power with the newer generation chips. I've been into mining since 2013, not exclusively BTC , but I always trade any other coins I mine for BTC as the end result.

I've used CPU mining for CPU algo coins, GPU mining for GPU algo coins, and ASIC mining for SHA-256 and Scrypt algo coins.

As an example, when mining the CPU coin "ROI Coin" my I-7 950 cpu draws about 125+ watts to produce about 40 H/s mining power on that algo. In contrast my Ryzen 7 1700 draws about 65 watts to produce 800 H/s mining power on the same algo. So to get the same performance and work from the old I-7 950 I'd need 20 of the I-7 950 machines to produce 800 H/s with a total wattage draw somewhere around 2500 watts to do the same work the Ryzen 7 1700 does at 65 watts.........

The GPU and ASIC chips are the same way, much more work, much less power usage.

Chris B.

 

[psionl0]

A modern CPU or GPU uses much less power with the newer generation chips.

It makes no difference how powerful computers are.

There is a sort of "Parkinson's Law" when it comes to bitcoin mining. The amount of processing power needed to mine bitcoins increases with the amount of processing power used to mine bitcoins. This is because the mining rate must remain constant regardless of the total processing power used and the hashing difficulty increases to make this so.

 

[ChrisBFRPKY]

I get that the processing power needed to mine BTC increases as the difficulty increases. More miners equals more difficulty. However, newer design chips will always have more processing power at less energy consumption. If this were not true, Bitmain would have been out of business a long time ago.

Chris B.

 

[psionl0]

I get that the processing power needed to mine BTC increases as the difficulty increases. More miners equals more difficulty. However, newer design chips will always have more processing power at less energy consumption. If this were not true, Bitmain would have been out of business a long time ago.

That doesn't address the issue. If everybody switches over to Ryzen 7 1700 then all that has happened is that the new computers have to work orders of magnitude more than the previous generation and in the end, the ones that are worked the hardest (consume most power) will mine the most blocks.

One step forward, two steps back.

 

[Roger Ramjets]

Bitcoin produces more Unique Artwork

Why the Bitcoin Price Has Crypto Traders Railing at Bart Simpson

Technically speaking, a Bart Simpson pattern occurs when a sideways action follows an unexpected spike in price, and then the price drops back again – suddenly. The price action matches the shape of Bart Simpson’s head.

In the current bitcoin price action, the BTC is halfway fitting the definition of the Bart Simpson pattern. A full Bart will develop if the cryptocurrency undergoes a sudden bearish correction, such that it erases its previous gains. Should that happen, bitcoin is looking at a downside target towards $3,355. That’s what a straight Bart Simpson head can do.

A dinosaur, a cartoon character - is Bitcoin drawing portraits of itself?

But the glass is half full. The bitcoin price has not entirely invalidated the bull flag formation, which became a central point of discussion in our previous analysis. The BTC/USD rate is still consolidating inside a little descending channel, which could mean there is still a chance that the pair would extend its upside momentum.

Ever the optimists these bitcoiners - all the technical indicators may be pointing down, but still they cling to the dream!

 

[psionl0]

Ever the optimists these bitcoiners - all the technical indicators may be pointing down, but still they cling to the dream!

Of course! TA is a scientifically valid tool in this thread.

 

[Samson]

Of course! TA is a scientifically valid tool in this thread.

Yes indeed it is. I have noted that Bitcoin behaviour is exactly predictable by algorithm. I will make specific predictions on request.

 

[Roger Ramjets]

the new computers have to work orders of magnitude more than the previous generation and in the end, the ones that are worked the hardest (consume most power) will mine the most blocks.

One step forward, two steps back.

Yes. Counterintuitively, the more processing power you throw at Bitcoin the more it needs, and the cost of that processing becomes its 'value'.

This is similar to what happens with gold. As the price goes up, so the does the cost of mining - because the higher price makes more difficult mining techniques profitable. Current annual gold 'consumption' is less than 1,000th of above-ground reserves, so if all the mines closed down today we would have enough to last the next 1000 years. We don't need to mine any more gold!

If gold was a normal commodity it would be grossly oversupplied and the price would plummet. But because people see it as a 'store of value' we continue to waste energy and damage the environment mining it. Fiat currencies avoid this waste. With paper - and now electronic - money, the cost of providing that 'store of value' goes way down, especially when the money supply needs to be increased.

But to 'Satoshi' and his anarcho-libertarian friends, a low cost flexible money supply is anathema. So he built into Bitcoin a digital version of the wasteful process used to mine gold, only worse. The sick part is that it was totally unnecessary.

 

[Mycroft]

You seem to be arguing that bitcoin prices are like a roulette wheel - you can never tell what number is coming up next.

In theory you could predict the roulette wheel. If you could somehow measure the mass, speed and direction of the ball as the croupier released it and factor in the mass, speed and direction of the wheel, it should be possible to predict how the ball will bounce and ultimately where it would end up.

In real life there is no way to measure those things with the speed and accuracy you would need, nor could you do the calculations fast enough to make a bet. Where the ball lands is just applied physics, but for any real life purpose it's random.

I would argue that it is like a biased roulette wheel. You still don't know what number is coming up next but you know that some numbers are more likely than others.

My argument is that because demand is driven by the unknowable thoughts and feelings of millions of bitcoin investors, there is nothing you can look at and measure in order to make useful predictions.

 

[Samson]

In theory you could predict the roulette wheel. If you could somehow measure the mass, speed and direction of the ball as the croupier released it and factor in the mass, speed and direction of the wheel, it should be possible to predict how the ball will bounce and ultimately where it would end up.

In real life there is no way to measure those things with the speed and accuracy you would need, nor could you do the calculations fast enough to make a bet. Where the ball lands is just applied physics, but for any real life purpose it's random.



My argument is that because demand is driven by the unknowable thoughts and feelings of millions of bitcoin investors, there is nothing you can look at and measure in order to make useful predictions.

I think chaos theory down to quantum level rules out that roulette notion. But price action can certainly be predicted with probabilty that knocks 50/50 into a cocked hat if simple algorithmic ideas are utilised.

 

[Mycroft]

I think chaos theory down to quantum level rules out that roulette notion. But price action can certainly be predicted with probabilty that knocks 50/50 into a cocked hat if simple algorithmic ideas are utilised.

What's your algorithm?

 

[Samson]

What's your algorithm?

Look for a reversal then a reaction back to a defined zone then take a position and then take a profit.
It is pretty simple, best explained with pictures but I will show on a bitcoin chart when I have time.

 

[Belz...]

I think chaos theory down to quantum level rules out that roulette notion.

If you're suggesting that quantum effects could change the outcome of the roulette wheel, you're wrong. At that scale the effects average out by several orders of magnitude.

 

[psionl0]

My argument is that because demand is driven by the unknowable thoughts and feelings of millions of bitcoin investors, there is nothing you can look at and measure in order to make useful predictions.

This is a silly argument.

It's like saying that you can't plan a road network because demand for roads is driven by the unknowable thoughts and feelings of millions of motorists. Yet with proper sampling, we not only know where to build a road, we also know how big to make the road.

Under your argument, insurance companies would not be able to set premiums because accidents are driven by the unknowable thoughts and feelings of millions of people (so they can sack their actuaries).

 

[The_Animus]

Yes indeed it is. I have noted that Bitcoin behaviour is exactly predictable by algorithm. I will make specific predictions on request.

Like this one from 1/19/2018

Cock eyed optimist.
$500 by end of january and penny dreadful by june.

Nailed it!

 

[Dr.Sid]

Something is happening none the less ..

 

[lomiller]

It's like saying that you can't plan a road network because demand for roads is driven by the unknowable thoughts and feelings of millions of motorists. Yet with proper sampling, we not only know where to build a road, we also know how big to make the road.

Under your argument, insurance companies would not be able to set premiums because accidents are driven by the unknowable thoughts and feelings of millions of people (so they can sack their actuaries).

These analogies fails because insurance traffic patterns have the exact thing everyone has been telling you bitcoin is missing. Eg roads have an underlying utility. People live in one place and need to get to where thy shop and where they work and roads help them get there. If you know how many people live in a place and how jobs there are in another you can make reasonably accurate predictions of how much traffic there will be between them before a road is ever built. You don't build a road then try to measure how big a road you needed. Insurance is similar, rates are predictable before you ever sell a policy, you create a policy and then try to figure out the rates based on how many of them you sell.

Bitcoin on the other hand has no such underlying utility driving it’s use. It’s more like people saying to themselves “it seems like a lot of people are driving these days so I had better get on the road first”. The problem is that there is no reason for any of them to “drive” anywhere, they are simply doing it to “get there first”.

The premise behind technical analysis is that even if you don’t know what underlying value is driving price trends you can make reasonable predictions abut it based on the behavior of buyers and sellers. This only works if buyers and sellers are making rational decisions based on underlying value. This is only somewhat true at the best of times and when it comes to something like bitcoin where price speculation itself is the entire basis for trends it’s one persons technical analysis being pitted against another, and ultimately someone’s version of technical analysis needs to loose.

 

[Mycroft]

This is a silly argument.

It's like saying that you can't plan a road network because demand for roads is driven by the unknowable thoughts and feelings of millions of motorists. Yet with proper sampling, we not only know where to build a road, we also know how big to make the road.

Under your argument, insurance companies would not be able to set premiums because accidents are driven by the unknowable thoughts and feelings of millions of people (so they can sack their actuaries).

Lolmiller said it pretty well. Actuaries and city planners have a lot of real data to work with, unlike Bitcoin investers.

Samson thinks he can build an algorithm to predict Bitcoin price changes, but the data he says he's going to look at is the very recent changes in Bitcoin prices. I'm skeptical that he can do this, but let's see.

 

[Dr.Sid]

Well his past price predictions were pretty bad ..

 

[psionl0]

Actuaries and city planners have a lot of real data to work with, unlike Bitcoin investers.

That wasn't your argument. Your argument was that bitcoin prices couldn't be predicted "because demand is driven by the unknowable thoughts and feelings of millions of bitcoin investors".

Samson thinks he can build an algorithm to predict Bitcoin price changes, but the data he says he's going to look at is the very recent changes in Bitcoin prices. I'm skeptical that he can do this, but let's see.

Why bring Samson into this? Are you attempting to create a false association?