Jan 12th to Approx Jan 27th diff thread (3.5%)

[Rum152]

I like to think that independent miners would keep running to secure the network but simple economics might keep that from happening. I see a possibility that large mining farms have so concentrated the mining power that if they were to all shut down one could switch on just long enough to do some damage. The concept of logical self interest isn't all encompassing and can't for tell every possible motivation for an attack.

The economics of home mining are not as cut and dry as the cost of electricity verses the value of the bitcoin mined.

I for example pay ~7.4 cents per KwH for electricity. My two S5s use roughly 1200 watts per hour. However since I am running them in my apartment I no longer need to have my heat on that uses roughly 2000 watts per hour. Granted I would not need my heat on 24 hours per day however this should be accounted for by the fact that the electric usage of the S5s is less then that of my heat. As a result I essentially have free electricity for my S5s while it is cold enough to need heat in my apartment.

It is obviously profitable for me to run my S5's now even if I was paying 7.4 cents per KwH, however next winter when this would likely no longer be the case, if I have not sold them (and have turned them off due to lack of my ability to run them profitably) then I would likely turn them back on for heat and to generate a small amount of bitcoin in the process.

[suchmoon]

I like to think that independent miners would keep running to secure the network but simple economics might keep that from happening. I see a possibility that large mining farms have so concentrated the mining power that if they were to all shut down one could switch on just long enough to do some damage. The concept of logical self interest isn't all encompassing and can't for tell every possible motivation for an attack.

I doubt that any hashrate drop will be so disruptive as to create any attack risk. Even if it drops by half you would still need a massive 150 PH/s to do a 51% and where would it come from?

I'm sure the network will be fine, diff drops, some miners switch off, perhaps exchange rate goes up a bit if coins are mined 10-20% slower until next adjustment, in other words it should be able to balance itself out.

[QuestionAuthority]

I like to think that independent miners would keep running to secure the network but simple economics might keep that from happening. I see a possibility that large mining farms have so concentrated the mining power that if they were to all shut down one could switch on just long enough to do some damage. The concept of logical self interest isn't all encompassing and can't for tell every possible motivation for an attack.

I doubt that any hashrate drop will be so disruptive as to create any attack risk. Even if it drops by half you would still need a massive 150 PH/s to do a 51% and where would it come from?

I'm sure the network will be fine, diff drops, some miners switch off, perhaps exchange rate goes up a bit if coins are mined 10-20% slower until next adjustment, in other words it should be able to balance itself out.

Let's hope.

[Rum152]

I like to think that independent miners would keep running to secure the network but simple economics might keep that from happening. I see a possibility that large mining farms have so concentrated the mining power that if they were to all shut down one could switch on just long enough to do some damage. The concept of logical self interest isn't all encompassing and can't for tell every possible motivation for an attack.

I doubt that any hashrate drop will be so disruptive as to create any attack risk. Even if it drops by half you would still need a massive 150 PH/s to do a 51% and where would it come from?

I'm sure the network will be fine, diff drops, some miners switch off, perhaps exchange rate goes up a bit if coins are mined 10-20% slower until next adjustment, in other words it should be able to balance itself out.

I think he does make a valid point. As it stands right now (prior to the massive price drop) both bitmain and SP are having trouble selling their equipment, as evidenced by their massive discounts, group buys, expanding into reversible payment methods, ect.

The current selloff and any potential future selloff will only exaggerate this.

This will result in someone potentially being able to purchase a large amount of the network hashrate very cheaply with the possibility of being able to potentially attack the network. 

[QuestionAuthority]

I like to think that independent miners would keep running to secure the network but simple economics might keep that from happening. I see a possibility that large mining farms have so concentrated the mining power that if they were to all shut down one could switch on just long enough to do some damage. The concept of logical self interest isn't all encompassing and can't for tell every possible motivation for an attack.

I doubt that any hashrate drop will be so disruptive as to create any attack risk. Even if it drops by half you would still need a massive 150 PH/s to do a 51% and where would it come from?

I'm sure the network will be fine, diff drops, some miners switch off, perhaps exchange rate goes up a bit if coins are mined 10-20% slower until next adjustment, in other words it should be able to balance itself out.

I think he does make a valid point. As it stands right now (prior to the massive price drop) both bitmain and SP are having trouble selling their equipment, as evidenced by their massive discounts, group buys, expanding into reversible payment methods, ect.

The current selloff and any potential future selloff will only exaggerate this.

This will result in someone potentially being able to purchase a large amount of the network hashrate very cheaply with the possibility of being able to potentially attack the network. 

Yes, that scenario is exactly what I'm talking about. I worry less about an attack when the price is stable and profitable for miners. When it becomes unprofitable and dropping or stabilizes at an unprofitable price point that creates an opportunity to exploit the weakness of the network.

At some point in the future I expect businesses like BitPay to take over mining in a big way. When this happens the network will be more centralized between a few businesses but safer because their profit will really exist in their primary business model. Mining, for them, will only be a sidebar necessary to secure their business investment. They will have no incentive to do anything but keep the network safe. Today the network is unstable because it's controlled by large pools/farms that have no other method of funding the mining operation except the mining. When they switch-off then anyone with enough power can switch-on and ultimately control the network. It would only take a small window of opportunity to do this.

[Rum152]

I like to think that independent miners would keep running to secure the network but simple economics might keep that from happening. I see a possibility that large mining farms have so concentrated the mining power that if they were to all shut down one could switch on just long enough to do some damage. The concept of logical self interest isn't all encompassing and can't for tell every possible motivation for an attack.

I doubt that any hashrate drop will be so disruptive as to create any attack risk. Even if it drops by half you would still need a massive 150 PH/s to do a 51% and where would it come from?

I'm sure the network will be fine, diff drops, some miners switch off, perhaps exchange rate goes up a bit if coins are mined 10-20% slower until next adjustment, in other words it should be able to balance itself out.

I think he does make a valid point. As it stands right now (prior to the massive price drop) both bitmain and SP are having trouble selling their equipment, as evidenced by their massive discounts, group buys, expanding into reversible payment methods, ect.

The current selloff and any potential future selloff will only exaggerate this.

This will result in someone potentially being able to purchase a large amount of the network hashrate very cheaply with the possibility of being able to potentially attack the network. 

Yes, that scenario is exactly what I'm talking about. I worry less about an attack when the price is stable and profitable for miners. When it becomes unprofitable and dropping or stabilizes at an unprofitable price point that creates an opportunity to exploit the weakness of the network.

At some point in the future I expect businesses like BitPay to take over mining in a big way. When this happens the network will be more centralized between a few businesses but safer because their profit will really exist in their primary business model. Mining, for them, will only be a sidebar necessary to secure their business investment. They will have no incentive to do anything but keep the network safe. Today the network is unstable because it's controlled by large pools/farms that have no other method of funding the mining operation except the mining. When they switch-off then anyone with enough power can switch-on and ultimately control the network. It would only take a small window of opportunity to do this.

Services like bitpay probably should get into the mining business (if they are not already). Doing so would greatly reduce their risk of being the victim of double spend attacks and would allow them to even accept 0 fee transactions.
I would however make a counter point to your concern, that is similar to a recent argument as to why PoW is superior to PoS. If someone were to buy up a lot of mining equipment cheaply and were to attack the network then they would lose out on their investment on the mining equipment (this may not matter if it was cheap enough) but also on future mining revenue. As it stands now miners will only be able to make roughly 1% (it is probably closer to 0.75%) of the total revenue they would mine over the course of their useful life. This percentage is probably lower when the difficulty has recently taken a huge drop. I think it is fair to say that any major attack on the network would likely be discovered within 48 hours of it starting, so an attacker would potentially lose out on ~98% of their potential mining revenue by attacking the network (this assumes that an attack were to cause confidence in bitcoin to be wiped out and have it's value go to essentially zero).

[QuestionAuthority]

I like to think that independent miners would keep running to secure the network but simple economics might keep that from happening. I see a possibility that large mining farms have so concentrated the mining power that if they were to all shut down one could switch on just long enough to do some damage. The concept of logical self interest isn't all encompassing and can't for tell every possible motivation for an attack.

I doubt that any hashrate drop will be so disruptive as to create any attack risk. Even if it drops by half you would still need a massive 150 PH/s to do a 51% and where would it come from?

I'm sure the network will be fine, diff drops, some miners switch off, perhaps exchange rate goes up a bit if coins are mined 10-20% slower until next adjustment, in other words it should be able to balance itself out.

I think he does make a valid point. As it stands right now (prior to the massive price drop) both bitmain and SP are having trouble selling their equipment, as evidenced by their massive discounts, group buys, expanding into reversible payment methods, ect.

The current selloff and any potential future selloff will only exaggerate this.

This will result in someone potentially being able to purchase a large amount of the network hashrate very cheaply with the possibility of being able to potentially attack the network. 

Yes, that scenario is exactly what I'm talking about. I worry less about an attack when the price is stable and profitable for miners. When it becomes unprofitable and dropping or stabilizes at an unprofitable price point that creates an opportunity to exploit the weakness of the network.

At some point in the future I expect businesses like BitPay to take over mining in a big way. When this happens the network will be more centralized between a few businesses but safer because their profit will really exist in their primary business model. Mining, for them, will only be a sidebar necessary to secure their business investment. They will have no incentive to do anything but keep the network safe. Today the network is unstable because it's controlled by large pools/farms that have no other method of funding the mining operation except the mining. When they switch-off then anyone with enough power can switch-on and ultimately control the network. It would only take a small window of opportunity to do this.

Services like bitpay probably should get into the mining business (if they are not already). Doing so would greatly reduce their risk of being the victim of double spend attacks and would allow them to even accept 0 fee transactions.
I would however make a counter point to your concern, that is similar to a recent argument as to why PoW is superior to PoS. If someone were to buy up a lot of mining equipment cheaply and were to attack the network then they would lose out on their investment on the mining equipment (this may not matter if it was cheap enough) but also on future mining revenue. As it stands now miners will only be able to make roughly 1% (it is probably closer to 0.75%) of the total revenue they would mine over the course of their useful life. This percentage is probably lower when the difficulty has recently taken a huge drop. I think it is fair to say that any major attack on the network would likely be discovered within 48 hours of it starting, so an attacker would potentially lose out on ~98% of their potential mining revenue by attacking the network (this assumes that an attack were to cause confidence in bitcoin to be wiped out and have it's value go to essentially zero).

What if they want the value to go to zero? The current thinking is one dimensional. There are more reasons to attack the network than just to double spend and profit is only one reason for action. A government, a competitor (Visa/MasterCard) or several of them working together could kill Bitcoin easly and spend almost nothing to do it. They wouldn't even need the resources they threw at Silk Road. If I were going to attempt to destroy Bitcoin I would use your mining equipment to do it. I would host mining equipment somewhere electricty is cheap then get people to ship me their equipment. I would run it honestly until the farm was a large percentage of the network and execute my plan. Miners switching off just speeds up that process.

[Rum152]

What if they want the value to go to zero? The current thinking is one dimensional. There are more reasons to attack the network than just to double spend and profit is only one reason for action. A government, a competitor (Visa/MasterCard) or several of them working together could kill Bitcoin easly and spend almost nothing to do it. They wouldn't even need the resources they threw at Silk Road.

I don't think the government would need the value the cost of  mining equipment to become very cheap to attack the network. They have a near unlimited amount of resources and bitcoin does not pose any serious threat to the US government/currency (the same is true for other major governments). Bitcoin poses to help the economy (and in tern governments) of smaller/weaker countries (eg 3rd world countries).

A payment processor like Visa/MC (or even Western Union) may have somewhat of an incentive to attempt to destroy bitcoin however they would probably also not need to wait for the price of mining equipment to fall (for example Visa reported a profit of $1.07 billion in the 3 months ending September 30 2014) plus them doing so would likely be considered anti-competative and could potentially result in criminal/civil action against them by the DOJ (it isn't like the DOJ hasn't ever used it's own interpretation of laws against major companies before). A best case scenario would be that their reputation would be greatly harmed.

You might argue that they could buy up the mining equipment without revealing their identities, however just like it is difficult to "mix" 10,000 BTC, it is difficult to buy up PHs worth of mining equipment.

If I were going to attempt to destroy Bitcoin I would use your mining equipment to do it. I would host mining equipment somewhere electricty is cheap then get people to ship me their equipment. I would run it honestly until the farm was a large percentage of the network and execute my plan. Miners switching off just speeds up that process.

This could be a threat. I would need to think about this one for a little bit. Although there are very few hosting providers that control large amounts of equipment like that.

[philipma1957]

well for now

Price 206 usd

 https://bitcoinwisdom.com/bitcoin/difficulty


Bitcoin Difficulty:   43,971,662,056
Estimated Next Difficulty:   43,363,458,449 (-1.38%)
Adjust time:   After 1142 Blocks, About 8.1 days
Hashrate(?):   321,684,441 GH/s
Block Generation Time(?):   
1 block: 10.2 minutes
3 blocks: 30.6 minutes
6 blocks: 1.0 hours
Updated:   18:55 (4.3 minutes ago)

****************************************
****************************************

http://www.bitcoincharts.com/   (-3.12%)



Blocks   339562
Total BTC   13.739M
 
Difficulty   43971662056
Estimated   42598199361 in 1142 blks
 
Network total   366207.958 Thash/s
Blocks/hour   6.98 / 516 s


So I see  big farms with a slew of s-3's sticking it out.

here is why :

  A 1 ph farm of s-3's with net op costs of 4.1  cents a kwatt  earns  1598 usd a day over op costs.
  A 1 ph farm of s-3's with net op costs of 5.1 cents a kwatt   earns 1411 usd  a day  over op costs.
  A 1 ph farm of s-3's with net op costs of 6.1 cents a kwatt   earns 1224  usd  a day over op costs.
 A 1 ph farm  of s-3's with net op costs of 7.1 cents a kwatt   earns  1037 usd  a day over op costs.
 A 1 ph farm of s-3's  with net op costs of 8.1 cents a kwatt   earns   849 usd   a day over op costs.
 A 1 ph  farm of s-3's with net op costs of 9.1 cents a kwatt   earns  662  usd  a day over op costs.



So in every case above they most  likely will mine and cash coins.

 I am not so sure they upgrades to s-5's
I am not so sure they sell off the gear.

And at the price of 206 usd a coin they are still earning over op costs.

[suchmoon]

After ~1000 blocks in this cycle (halfway point) we have this:

Estimated Next Difficulty:   43,313,520,359 (-1.50%)
Block Generation Time (Average of last 504 blocks):   
1 block: 10.2 minutes
3 blocks: 30.5 minutes
6 blocks: 1.0 hours

So the most recent 504 blocks were running at 10.2 minutes average, or ~2% below target. The first ~500 blocks of this cycle were ~4% below, making the first half about 3% below target. This doesn't sound like much, but if you look at past data you would see that this is one of the slowest first-half performances, so there is a good chance we'll see a decrease this cycle. It likely won't be double digits though as I was hoping for:

To end up at -10%: hashrate would have to decrease immediately by ~16% from the current level and stay there until the end of the cycle.
To end up at + (diff increase): hashrate would have to increase immediately by ~5%+ from the current level and stay there until the end of the cycle.

Unless I messed up my napkin math, in which case shame on you for reading this far

After nearly 500 blocks done in this cycle the average 504-block time is 10.4 minutes, or ~4% below target.

It looks like there is a lot of hope still in the mining business that BTC will rise. I don't believe that only less than 10% of hashpower would become unprofitable with this recent exchange rate drop. Some are probably just trying to hold out until their next electric bill

[philipma1957]

slight drop

https://bitcoinwisdom.com/bitcoin/difficulty

Bitcoin Difficulty:   43,971,662,056
Estimated Next Difficulty:   43,095,517,516   (-1.99%)
Adjust time:   After 1005 Blocks, About 7.2 days
Hashrate(?):   316,383,090 GH/s
Block Generation Time(?):   
1 block: 10.2 minutes
3 blocks: 30.8 minutes
6 blocks: 1.0 hours
Updated:   18:5 (3.2 minutes ago)

http://www.bitcoincharts.com/


Blocks   339699
Total BTC   13.742M
 
Difficulty   43971662056
Estimated   42529538155 in 1005 blks  (-3.27%)
 
Network total   316900.229 Thash/s
Blocks/hour   6.04 / 596 s


I figure drops this small may be revealing that a lot of data centers are still making money.
I did expect  -6 to 9% for this due to price drop in coins.

[davejh]

I figure drops this small may be revealing that a lot of data centers are still making money.
I did expect  -6 to 9% for this due to price drop in coins.

I'm not sure that DCs are actually making money, just that the alternatives to continuing to run would be to lose even more money. If you have a contract for power and facilities and have to pay for both whether you use them or not then you probably have to just keep paying and hope that the coin price improves. Similarly if you're supporting cloud mining then those miners probably have contracts that have to be honoured too.

My guess is that any reduction will be gradual, based on contracts expiring, but even if several DCs with 1 PH/s of capacity went offline the impact wouldn't really be that noticeable.

[philipma1957]

I figure drops this small may be revealing that a lot of data centers are still making money.
I did expect  -6 to 9% for this due to price drop in coins.

I'm not sure that DCs are actually making money, just that the alternatives to continuing to run would be to lose even more money. If you have a contract for power and facilities and have to pay for both whether you use them or not then you probably have to just keep paying and hope that the coin price improves. Similarly if you're supporting cloud mining then those miners probably have contracts that have to be honoured too.

My guess is that any reduction will be gradual, based on contracts expiring, but even if several DCs with 1 PH/s of capacity went offline the impact wouldn't really be that noticeable.

I used 1ph farms for simple math.  Any farm running at .8 watts per gh and under 10 cents is not going to turn off.

S-3's and s-4's do .7 to .8 watts   they will continue to run at this price.  I would say if they can maintain diff under 45xxxx they can run on and on and on.

We may see diff between 40-45  and price about 200 until the spring as it does not kill off bigger miners with <10 cents a kwatt and under .8 watts a gh.

It does destroy just about all miners in europe or usa that have 15 +cents a watt.

[adaseb]

So I guess when prices go to $100 or so would be the safest time to buy mining ASICs?

[suchmoon]

So I guess when prices go to $100 or so would be the safest time to buy mining ASICs?

If you are talking about profitability, at any BTC exchange rate the factors determining the purchase decision are still the same:

- price of the miner (if it costs $300 per TH/s then at $100 it's 3BTC - steep, but see below)
- difficulty and your expectation of its future direction
- power cost ratio with the BTC exchange rate and your expectation of its future direction, unless your utility bills you in BTC, which would greatly simplify this part

But if it's a hobby, just buy an S5 or an SP20 now, makes no big difference really. It's mostly a lottery ticket anyway.

[philipma1957]

Tues morning in NJ.


http://www.bitcoincharts.com/

BTC = 210

Blocks   339772
Total BTC   13.744M
 
Difficulty   43971662056
Estimated   42410605778 in 932 blks     This is a (-3.55%) drop
 
Network total   275671.053 Thash/s
Blocks/hour   5.25 / 685 s

Note 5.25 blocks per hour   also note 275 PH estimate.   It would be interesting to see if this continues if it does  the drop in diff will go over  6%.

5.25 block rate is 13% less then normal   average that in with the 3.55 drop so far and it looks like  -8% is possible.

[davejh]

Tues morning in NJ.


http://www.bitcoincharts.com/

BTC = 210

Blocks   339772
Total BTC   13.744M
 
Difficulty   43971662056
Estimated   42410605778 in 932 blks     This is a (-3.55%) drop
 
Network total   275671.053 Thash/s
Blocks/hour   5.25 / 685 s

Note 5.25 blocks per hour   also note 275 PH estimate.   It would be interesting to see if this continues if it does  the drop in diff will go over  6%.

5.25 block rate is 13% less then normal   average that in with the 3.55 drop so far and it looks like  -8% is possible.

the 275 PH/s number looks like it's normal statistical variance so far. On the 13th the daily rate was 229, but on the 14th it was back at 330.

Over on https://bitcoinwisdom.com/bitcoin/difficulty the grey line has been above the red one for pretty-much the whole difficulty level so far so that tends to suggest that we'll stay well above an -7.6% level. The downward trend seems likely though.

[philipma1957]

right now bitcoincharts is predicting much lower then bitcoin wisdom.

http://www.bitcoincharts.com/

Blocks   339788
Total BTC   13.745M
 
Difficulty   43971662056
Estimated   42229565780 in 916 blks
 
Network total   251018.507 Thash/s
Blocks/hour   4.78 / 752 s-------- This is very low number it is -20%

The blocks per hour needs to be watched a lot if it stays under 5 for a day in a row it means a big farm has shut down.

 
It was 5.25     at 932 blocks to go.
it is now 4.78  at 916 blocks to go.
it is now 4.88  at 898 blocks to go.
it is now 5.14 at  880 blocks to go.
it is now 5.26 at  866 blocks to go.   this is a 66 block time period in which the rate has been -10 to  -20
it is now 5.17 at  862 blocks to go.
it is now 5.07 at  857  blocks
it is now 5.06 at  812 blocks
 it is not a long enough time period  to be sure if a major player has turn down  about 13 hours time. I will check it in a while.


If this stays low diff will drop big.     I will check on this a few times today weds and thurs to see if it was variance or a big shut down.

the new numbers dropped us to.
http://www.bitcoincharts.com/

Blocks   339842
Total BTC   13.746M
 
Difficulty   43971662056
Estimated   41869404718 in 862 blks  this is now (-4.78%)
 
Network total   270976.587 Thash/s
Blocks/hour   5.17 / 697 s >>>>>>>>>>>>>>>> not sure how long it stays here but if it stays here we will see a good  drop  come Jan 27 or 28.

[judypug1956]

so from 932 to  812 is 120 blocks and the average is 5 blocks an hour not 6 as it should be.

5/6 =  down 16.67%

  It is now a day .  120 blocks/5 block an hour  = 24 hours  of  -15%

needs to be watched longer to see if this keeps up.

[philipma1957]

so from 932 to  812 is 120 blocks and the average is 5 blocks an hour not 6 as it should be.

5/6 =  down 16.67%

  It is now a day .  120 blocks/5 block an hour  = 24 hours  of  -15%

needs to be watched longer to see if this keeps up.

It was    5.25 at  932 blocks ---    0
it is now 4.78 at  916 blocks ---  16
it is now 4.88 at  898 blocks ---  34
it is now 5.14 at  880 blocks ---  52
it is now 5.26 at  866 blocks ---  66   
it is now 5.17 at  862 blocks ---  70
it is now 5.07 at  857 blocks ---  75
it is now 5.06 at  812 blocks ---  120



I will keep track here since this looks more like a trend then variance.