Diff thread sept 4th. to sept 17th Picks are closed.

[philipma1957]

New thread. Just setting it up.  last thread is here  jumps was 4.98%


https://bitcointalk.org/index.php?topic=1159269.msg12325943#msg12325943


Picks start at block 300.

So we cracked 400 ph this is  about 380,000 s-5's  or around  84,000 s-7's

So for us to jump 5% for a second time  in a row  4200 s-7's need to go on line.



372961 was the starting block for this adjustment.  I could be off 1 block here.

[valkir]

+5% and right now +7% ouch that's hard! 

[notlist3d]

+5% and right now +7% ouch that's hard! 

I think some machines are kicking on.  I was afraid of this.  Last week was high and not fun for us.

Now this possibly even higher looks like someone is making something.   I wonder if it truly is S7's if so they sure know how to make the units quickly.

[Dexter770221]

....
So for us to jump 5% for a second time  in a row  4200 s-7's need to go on line.
.....

Very probable that it will happen.
But more interesting times ahead. In about two months we should see more +%.
http://www.coindesk.com/bitfury-completion-16nm-bitcoin-mining-asic/

[jonnybravo0311]

Well, the conspiracy theorist in me says the increased hash is directly from Bitmain turning on those S7s in their data centers.  We were certainly right that the S5+ was nothing more than them liquidating their remaining stock of BM1384 chips.  Imagine the same form factor with the BM1385 chips.  That'd be about 14.6TH/s for 3630W.  Pretty sweet.

Alright, so this upcoming difficulty adjustment is likely going to follow the 5% trend according to my tea leaves and chicken entrails.  Also, Marty McFly paid me a visit earlier and told me so.  I tried to steal the DeLorean, but no such luck - he said he knew I was going to try.  Stupid time travelers... think they know everything.

[mwizard]

There is another possibility other than the Bitmain S7 causing the increase in network hash rate.

It is 6 months since KNC announced the tapeout of their 16 nm chips.  KNC stated in June that they had the 16nm physical chips and that they worked.  Is it possible the hashrate increase is coming from these new chips?  How would we know.  KNC were the first with 20nm Bitcoin miners so they have a track record of leading bitcoin technology.

See for example http://www.kncminer.com/blog/newsarchive
http://www.coindesk.com/kncminer-deploys-next-generation-16nm-bitcoin-asic/

Based on the KNC 0.07 Joules/Gigahash figures a typical 1,600 Watt 16nm miner would run at potentially up to 20 Terrahash/s.  This is better than 3 times the performance of an S7.

Just a thought on what is coming , or may even be here now.

[philipma1957]

next two jumps are important.  two 5% jumps plus the one we had would be a trend that the bigger builders all have better chips and are all retooling. If this occurs we will be at 450ph by  oct 1st and we will then see if the big builders decide to slow growth down again like they did last year.

Or and here is the or do they all say f it and try to out build each other.   I believe they will get us at 60-65 diff and 450-500 ph and then go flat.

Should be fun to watch.

[Mikestang]

It's going to be a wild winter, I wonder when we'll see a negative diff again?

[alh]

It's going to be a wild winter, I wonder when we'll see a negative diff again?

The rainbow colored Unicorn will personally deliver the notice to you..... 

Actually probably sooner than that, but not a lot.

[mwizard]

next two jumps are important.  two 5% jumps plus the one we had would be a trend that the bigger builders all have better chips and are all retooling. If this occurs we will be at 450ph by  oct 1st and we will then see if the big builders decide to slow growth down again like they did last year.

Or and here is the or do they all say f it and try to out build each other.   I believe they will get us at 60-65 diff and 450-500 ph and then go flat.

I suspect it is much more likely we will get to 100 diff and possibly 1,000 ph before growth goes flat next year.

Both KNC and Bitfury are expecting 16 nm chips to be at 0.07 Joules/Gigahash.  This means miners with at least 3 times better electrical efficiency than current 28nm miners.  

I know there is considerable daily statistical variation but over the last few days the KNCMiners pool seems to have grown considerable.  It will be interesting to see how KNCminers and Bitfury change over the next few months.  See
https://data.bitcoinity.org/bitcoin/hashrate/30d?c=m&t=a

Another period of change is coming whether we like it or not.




  

[VirosaGITS]

I hope not. 1k PH would be insane growth. I can't imagine it happening at all.

If it happened it would really suck because i very much doubt BTC would go up to 500$.

[philipma1957]

next two jumps are important.  two 5% jumps plus the one we had would be a trend that the bigger builders all have better chips and are all retooling. If this occurs we will be at 450ph by  oct 1st and we will then see if the big builders decide to slow growth down again like they did last year.

Or and here is the or do they all say f it and try to out build each other.   I believe they will get us at 60-65 diff and 450-500 ph and then go flat.

I suspect it is much more likely we will get to 100 diff and possibly 1,000 ph before growth goes flat next year.

Both KNC and Bitfury are expecting 16 nm chips to be at 0.07 Joules/Gigahash.  This means miners with at least 3 times better electrical efficiency than current 28nm miners.  

I know there is considerable daily statistical variation but over the last few days the KNCMiners pool seems to have grown considerable.  It will be interesting to see how KNCminers and Bitfury change over the next few months.  See
https://data.bitcoinity.org/bitcoin/hashrate/30d?c=m&t=a

Another period of change is coming whether we like it or not.




  

We are not going to get to 100 diff and  1000 ph .  before the ½ .

 unless price jacks upwards   quite a bit.

[mwizard]

We are not going to get to 100 diff and  1000 ph .  before the ½ .

 unless price jacks upwards   quite a bit.

We are already close to a 60 diff and likely to go over 60 at the next adjustment.  I don't see any reason why there cannot be growth to 100 by the middle of next year even at the current Bitcoin price.  There are 23 adjustments until then.

Mining, and the push for faster miners,  will continue before and after the halving.

[philipma1957]

We are not going to get to 100 diff and  1000 ph .  before the ½ .

 unless price jacks upwards   quite a bit.

We are already close to a 60 diff and likely to go over 60 at the next adjustment.  I don't see any reason why there cannot be growth to 100 by the middle of next year even at the current Bitcoin price.  There are 23 adjustments until then.

Mining, and the push for faster miners,  will continue before and after the halving.

you don't understand why a builder with 10% of the hash loses money by expanding to 15% of the hash you think he makes more  money if he does that.

the problem is 4-6 guys all have 10 to 15% of the hash each.  they all can build bp and then the other 4-6 guys match them.

So what hey have been doing is swapping  gear with out really expand hash by a ton.

they have s-3's at .75watts  they have s-5s at .50watts  now they have s-7s at .25watts

so they ⅓ their power   so many think they  will 3x their hash .  if the top 6 builders all do this they all gain nothing and they spent money to make the gear.

so they try to ⅓ their power  and 1.5x or 1.2x their hash.

if the top 6 builders do this.  they keep the same share of the network and they spend less on power.

This is what has stopped the crazy growth the last year.

unless your gear is god like say .01 watts a gh and really cheap to build expansion of hash matching your efficiency gain is a loser economically.

so s-5 at .5 and s-7 at .25  for example does not pay to 2x the hash power
 if the other builders can match you.  but if antminer ½ the power and 1.25x the hash   and are matched by the other builders  antminer makes more money.

this is very simple math.

[mwizard]

you don't understand why a builder with 10% of the hash loses money by expanding to 15% of the hash you think he makes more  money if he does that.

the problem is 4-6 guys all have 10 to 15% of the hash each.  they all can build bp and then the other 4-6 guys match them.

So what hey have been doing is swapping  gear with out really expand hash by a ton.

they have s-3's at .75watts  they have s-5s at .50watts  now they have s-7s at .25watts

so they ⅓ their power   so many think they  will 3x their hash .  if the top 6 builders all do this they all gain nothing and they spent money to make the gear.

so they try to ⅓ their power  and 1.5x or 1.2x their hash.

if the top 6 builders do this.  they keep the same share of the network and they spend less on power.

This is what has stopped the crazy growth the last year.

unless your gear is god like say .01 watts a gh and really cheap to build expansion of hash matching your efficiency gain is a loser economically.

so s-5 at .5 and s-7 at .25  for example does not pay to 2x the hash power
 if the other builders can match you.  but if antminer ½ the power and 1.25x the hash   and are matched by the other builders  antminer makes more money.

this is very simple math.

I believe there are several errors in your analysis.  You have assumed everyone pays a similar amount for their miners, has similar costs and does not try to drive others out of business.

1) Not all miners pay the same for their miners.  The mining hardware with the highest efficiency is not likely to be sold to the public.  It is more economic for the manufacturer to keep it to themselves and mine themselves. This is what KNC and Bitfury do.  They manufacture their own cheap hardware and do not sell to to the public.  It is in their interest to increase their hash rate as it increases their profits.

2) For similar reasons miners in areas with the cheapest electricity will try to maximize their hash rates.  Anyone who pays more than about $0.05 per kWh is likely to be ultimately driven out of the mining business.

3) Finally you have assumed people would co-operate if they all had the same costs.  But people are not always that rational in an unregulated market.  See the Tragedy of the Commons.  https://en.wikipedia.org/wiki/Tragedy_of_the_commons

The entire mining game is a zero sum game.  There is only a fixed amount of money available from Bitcoin mining.  

The best way to maximize your profits is to not share your competitive advantage and to keep your costs below your competitors.  Ideally you would like most of your competitors to give up and leave the market.

So I expect the battle between the miners to continue and the hash rate to continue to increase for some time yet.  



 

[VirosaGITS]

you don't understand why a builder with 10% of the hash loses money by expanding to 15% of the hash you think he makes more  money if he does that.

the problem is 4-6 guys all have 10 to 15% of the hash each.  they all can build bp and then the other 4-6 guys match them.

So what hey have been doing is swapping  gear with out really expand hash by a ton.

they have s-3's at .75watts  they have s-5s at .50watts  now they have s-7s at .25watts

so they ⅓ their power   so many think they  will 3x their hash .  if the top 6 builders all do this they all gain nothing and they spent money to make the gear.

so they try to ⅓ their power  and 1.5x or 1.2x their hash.

if the top 6 builders do this.  they keep the same share of the network and they spend less on power.

This is what has stopped the crazy growth the last year.

unless your gear is god like say .01 watts a gh and really cheap to build expansion of hash matching your efficiency gain is a loser economically.

so s-5 at .5 and s-7 at .25  for example does not pay to 2x the hash power
 if the other builders can match you.  but if antminer ½ the power and 1.25x the hash   and are matched by the other builders  antminer makes more money.

this is very simple math.

I believe there are several errors in your analysis.  You have assumed everyone pays a similar amount for their miners, has similar costs and does not try to drive others out of business.

1) Not all miners pay the same for their miners.  The mining hardware with the highest efficiency is not likely to be sold to the public.  It is more economic for the manufacturer to keep it to themselves and mine themselves. This is what KNC and Bitfury do.  They manufacture their own cheap hardware and do not sell to to the public.  It is in their interest to increase their hash rate as it increases their profits.

2) For similar reasons miners in areas with the cheapest electricity will try to maximize their hash rates.  Anyone who pays more than about $0.05 per kWh is likely to be ultimately driven out of the mining business.

3) Finally you have assumed people would co-operate if they all had the same costs.  But people are not always that rational in an unregulated market.  See the Tragedy of the Commons.  https://en.wikipedia.org/wiki/Tragedy_of_the_commons

The entire mining game is a zero sum game.  There is only a fixed amount of money available from Bitcoin mining.  

The best way to maximize your profits is to not share your competitive advantage and to keep your costs below your competitors.  Ideally you would like most of your competitors to give up and leave the market.

So I expect the battle between the miners to continue and the hash rate to continue to increase for some time yet.  

I think there is a flaw in your counter analysis, because in reality, if a side push for tripling their hash, the other sides will keep the pace and in the end they all undermined themselves. And this is not speaking of the production limitations that simply cannot be overcomed in such a small time span. So no, we won't see 1000PH in a few months.

It's not that easy to produce 2 billions chip at the technological edge. The more you try to get in a small time span, the higher to production cost. Exponentially.

[notlist3d]

you don't understand why a builder with 10% of the hash loses money by expanding to 15% of the hash you think he makes more  money if he does that.

the problem is 4-6 guys all have 10 to 15% of the hash each.  they all can build bp and then the other 4-6 guys match them.

So what hey have been doing is swapping  gear with out really expand hash by a ton.

they have s-3's at .75watts  they have s-5s at .50watts  now they have s-7s at .25watts

so they ⅓ their power   so many think they  will 3x their hash .  if the top 6 builders all do this they all gain nothing and they spent money to make the gear.

so they try to ⅓ their power  and 1.5x or 1.2x their hash.

if the top 6 builders do this.  they keep the same share of the network and they spend less on power.

This is what has stopped the crazy growth the last year.

unless your gear is god like say .01 watts a gh and really cheap to build expansion of hash matching your efficiency gain is a loser economically.

so s-5 at .5 and s-7 at .25  for example does not pay to 2x the hash power
 if the other builders can match you.  but if antminer ½ the power and 1.25x the hash   and are matched by the other builders  antminer makes more money.

this is very simple math.

I believe there are several errors in your analysis.  You have assumed everyone pays a similar amount for their miners, has similar costs and does not try to drive others out of business.

1) Not all miners pay the same for their miners.  The mining hardware with the highest efficiency is not likely to be sold to the public.  It is more economic for the manufacturer to keep it to themselves and mine themselves. This is what KNC and Bitfury do.  They manufacture their own cheap hardware and do not sell to to the public.  It is in their interest to increase their hash rate as it increases their profits.

2) For similar reasons miners in areas with the cheapest electricity will try to maximize their hash rates.  Anyone who pays more than about $0.05 per kWh is likely to be ultimately driven out of the mining business.

3) Finally you have assumed people would co-operate if they all had the same costs.  But people are not always that rational in an unregulated market.  See the Tragedy of the Commons.  https://en.wikipedia.org/wiki/Tragedy_of_the_commons

The entire mining game is a zero sum game.  There is only a fixed amount of money available from Bitcoin mining.  

The best way to maximize your profits is to not share your competitive advantage and to keep your costs below your competitors.  Ideally you would like most of your competitors to give up and leave the market.

So I expect the battle between the miners to continue and the hash rate to continue to increase for some time yet.  

With keeping the company needs to build a data center.  These are not cheap for miners.   The cooling system is much more then standard server data center.

So there is some big costs of keeping them only.  Also if they can make decent sales to public it brings down price to produce and the company could keep miners even cheaper.  So I still see advantages to selling to public as long as they make profit.

[philipma1957]

you don't understand why a builder with 10% of the hash loses money by expanding to 15% of the hash you think he makes more  money if he does that.

the problem is 4-6 guys all have 10 to 15% of the hash each.  they all can build bp and then the other 4-6 guys match them.

So what hey have been doing is swapping  gear with out really expand hash by a ton.

they have s-3's at .75watts  they have s-5s at .50watts  now they have s-7s at .25watts

so they ⅓ their power   so many think they  will 3x their hash .  if the top 6 builders all do this they all gain nothing and they spent money to make the gear.

so they try to ⅓ their power  and 1.5x or 1.2x their hash.

if the top 6 builders do this.  they keep the same share of the network and they spend less on power.

This is what has stopped the crazy growth the last year.

unless your gear is god like say .01 watts a gh and really cheap to build expansion of hash matching your efficiency gain is a loser economically.

so s-5 at .5 and s-7 at .25  for example does not pay to 2x the hash power
 if the other builders can match you.  but if antminer ½ the power and 1.25x the hash   and are matched by the other builders  antminer makes more money.

this is very simple math.

I believe there are several errors in your analysis.  You have assumed everyone pays a similar amount for their miners, has similar costs and does not try to drive others out of business.

1) Not all miners pay the same for their miners.  The mining hardware with the highest efficiency is not likely to be sold to the public.  It is more economic for the manufacturer to keep it to themselves and mine themselves. This is what KNC and Bitfury do.  They manufacture their own cheap hardware and do not sell to to the public.  It is in their interest to increase their hash rate as it increases their profits.

2) For similar reasons miners in areas with the cheapest electricity will try to maximize their hash rates.  Anyone who pays more than about $0.05 per kWh is likely to be ultimately driven out of the mining business.

3) Finally you have assumed people would co-operate if they all had the same costs.  But people are not always that rational in an unregulated market.  See the Tragedy of the Commons.  https://en.wikipedia.org/wiki/Tragedy_of_the_commons

The entire mining game is a zero sum game.  There is only a fixed amount of money available from Bitcoin mining.  

The best way to maximize your profits is to not share your competitive advantage and to keep your costs below your competitors.  Ideally you would like most of your competitors to give up and leave the market.

So I expect the battle between the miners to continue and the hash rate to continue to increase for some time yet.



 

I don't assume that every new miner is equal in cost.--- See my first bold.

There has not been a battle between the building miners for the last year.--- Hash rate has been flat for quite a while.--- This is about builders mining not builders seller as spondoolies and bitmaintech did have a selling war Last DEC/JAN.  That war crashed price of BTC down to 180 for a while

Now Lets be frank I am basing this on Sept/Oct 2014 to July/Aug 2015.

This means Anything after that can be new.  But no builder will attempt to crush another builder at self mining unless they have a god-like chip. With a monster edge. And it needs to be cheap to build and run. And to try to kill your competitors off now if you have the chip of doom is moron just wait for the ½ ing and the summer heat of 2016 then hit them with the chip of death.



So to say mining is an end game is not applicable at all as the reason to ramp your gear up like mad.

[notlist3d]

Night Bitwisdom:

Bitcoin Difficulty:    56,957,648,455
Estimated Next Difficulty:    61,028,387,965 (+7.15%)
Adjust time:    After 1898 Blocks, About 12.9 days
Hashrate(?):    418,800,149 GH/s

So not looking good 2 high weeks in a row.

As far as price around 229 so really no big movement.  Some really needs to change either price of BTC or lower changes.  I think I'm fine through winter no matter most changes.  But if we have these future mining will be much harder for me.

[ezeminer]

Bitcoin got into those special brownies and wants to go to the moon on hashrate and difficulty.