Dec 31 to Jan 12 Diff thread picks closed.

[Mikestang]

This will be the highest prize given as I will stop  new rollovers at max of .2 or .3 btc

I'm glad you're doing this, I would vote for 0.2 max to keep the funds alive as long as possible.

This exponential network growth is truly amazing.  10% of 10% of 10% really adds up fast.

[philipma1957]

This will be the highest prize given as I will stop  new rollovers at max of .2 or .3 btc

I'm glad you're doing this, I would vote for 0.2 max to keep the funds alive as long as possible.

This exponential network growth is truly amazing.  10% of 10% of 10% really adds up fast.

I have .9 btc to give away so 9 more adjustments and we are done.  

(unless my mining picks up) ----  I am so fucked mining on mmpool.org  Stuck more then 10 btc. I  truly need us to find  a block on that pool.


BTW https://bitcoinwisdom.com/bitcoin/difficulty  --- (+8.26%)  way off  they use the last 500 blocks from the prior jump so that number is really high.

Bitcoin Difficulty:   103,880,340,815
Estimated Next Difficulty:   112,464,469,621 (+8.26%) way high
Adjust time:   After 1668 Blocks, About 11.7 days
Hashrate(?):   722,213,465 GH/s
Block Generation Time(?):   
1 block: 10.1 minutes
3 blocks: 30.2 minutes
6 blocks: 1.0 hours
Updated:   18:10 (1.7 minutes ago)


Real time below

http://bitcoincharts.com/


Blocks   391453
Total BTC   15.036M
 
Difficulty   103880340815
Estimated   103106281319 in 1667 blks   we are negative  0.745%  at this moment.
 
Network total   769163.947 Thash/s
Blocks/hour   6.21 / 580 s


my early guess is  + 8.26   and -0.745 =  7.515/2 or   +3.7575%     not bad if I am correct still early only 450 block made.

[adaseb]

Someone just turned the tap back on... In the last 3 Hours we have gone from a running rate of -6% to -3%.

Rich

Today will end up as flat for the day at negative or 0%. Don't know about tomorrow or for the rest of the period however.

[philipma1957]

Someone just turned the tap back on... In the last 3 Hours we have gone from a running rate of -6% to -3%.

Rich

Today will end up as flat for the day at negative or 0%. Don't know about tomorrow or for the rest of the period however.

yep as I believe at least 50 maybe 100 ph is offline at this moment.  so we could do 144 blocks or 134 blocks or 170 blocks on sunday.

The diff picks are a different game then the have been as power on power off appears to be real and to be practiced.

I am still not quite sure as to the why of it.

[VirosaGITS]

Someone just turned the tap back on... In the last 3 Hours we have gone from a running rate of -6% to -3%.

Rich

Today will end up as flat for the day at negative or 0%. Don't know about tomorrow or for the rest of the period however.

yep as I believe at least 50 maybe 100 ph is offline at this moment.  so we could do 144 blocks or 134 blocks or 170 blocks on sunday.

The diff picks are a different game then the have been as power on power off appears to be real and to be practiced.

I am still not quite sure as to the why of it.

I'm glad they're having problem with their center, but it must still sting to have so much hardware not mining. Thats a pretty hefty loss per day.

[philipma1957]

Someone just turned the tap back on... In the last 3 Hours we have gone from a running rate of -6% to -3%.

Rich

Today will end up as flat for the day at negative or 0%. Don't know about tomorrow or for the rest of the period however.

yep as I believe at least 50 maybe 100 ph is offline at this moment.  so we could do 144 blocks or 134 blocks or 170 blocks on sunday.

The diff picks are a different game then the have been as power on power off appears to be real and to be practiced.

I am still not quite sure as to the why of it.

I'm glad they're having problem with their center, but it must still sting to have so much hardware not mining. Thats a pretty hefty loss per day.

well remember when you have 20 % of the network and move to 30% of the network  you are mining against your self.


Here is what I mean  20% of 100 = 20   to go to 30%     30 of 110 is not high enough.  it is only 27.27%   


 so if you have 20 of 100 and 20% 

you need 34.3  of 114.3  to be 30%


so to move up to 30% of the blocks  in a diff adjustment  which is  604.8 blocks  vs   20% of the blocks 403.2

you need to raise your hash by 71.5%    say 20 to 34.3   yet your blocks go up 50%      403.2 + 201.6  is 50%

so at  a certain point increase of hash is not as good as you think.

A better example is 20% to 40%      or 403.2 blocks to 806.4 blocks   

  just double the hash from 20 of 100 to 40 of 120 nope

try 20 of 100 to     53.3 of 133.3      that is 2.65 x 1 of your hash to 2x your blocks


             so  you start to compete against your own hashrate.

[flikflak]

@philipma1957

Sure, but thats always the case - even with <0.5% of the network and moving it to ~1% (like kano did). Now for Bitfury every generated block before the halving counts and I can't belive they are slowing it down for a reason. They have to compete with other pools like antpool. And if your competitor can deliver 18% more hashrate (from 170 PH/s to 200 PH/s), then you have to match it - if you can't, you are losing the "game" in the long run. I don't think that the 1st gen of 16nm chips is much more efficient than current 28nm chips with 0.25 J/GH @ the wall.

[philipma1957]

@philipma1957

Sure, but thats always the case - even with <0.5% of the network and moving it to ~1% (like kano did). Now for Bitfury every generated block before the halving counts and I can't belive they are slowing it down for a reason. They have to compete with other pools like antpool. And if your competitor can deliver 18% more hashrate (from 170 PH/s to 200 PH/s), then you have to match it - if you can't, you are losing the "game" in the long run. I don't think that the 1st gen of 16nm chips is much more efficient than current 28nm chips with 0.25 J/GH @ the wall.

you math is wrong if you think that moving from .5 % to 1% is the same as moving from 20% to 40%.


When you have 1/ 200  of the network you would earn 10.08 blocks  out of 2016  that is .5%

so to go to 1 %  you need.  2.03/201.03 (rounded number)     so gear wise to 2x you block share    you need to 2.03  x  your gear a small penalty


but when you are 20% of the blocks  if you want 40% of the blocks or 2x you blocks you need 2.65 x the gear. A big penalty.


so it becomes more expensive to 2x your share at 20% to 40%  then .5% to 1%

So bitfury has much more difficulty in increasing its share of the network. Then kano pool does.

They need to weigh how much gear to turn on since the gear does not earn more blocks. With out a extra cost.



2.03 x your gear at .5%  makes your blocks go 2 x to 1%   kano.is  the extra price in gear is small

2.65 x your gear  at 20% makes your blocks go 2 x to 40% bitfury  the extra price in gear is bigger.




Does anyone disagree with this math?  If you do disagree  you are going to have to learn it is the right math or live with a false idea.  

 It is true and the biggest reason companies do not go much over 30% of the network as a farmer/miner.  The extra hash does not increase blocks on a linear basis.

to go from 20% of the blocks to 50%   is a factor of 2.5  so  

20/100   is 20%    It needs to become 80/160 to be 50%

so your gear is 20  and you raised it to 80    you 4x your gear  and only 2x your coins

So if bitfury is truly 225ph  of 725ph  that is 31% of the network  or  around 625 blocks an adjustment.

if they can go to 300ph    it would be   300 of 800ph or 37.5   percent     that is about   756 blocks

so  they add 75ph to get 131 blocks more

or they let 225ph get  625 block now  simple math says   if 225 ph earns 625 blocks now  they are earning 208.33 blocks for each 75ph piece they already have.

adding the next 75ph earns   only 131 blocks  which is 77 blocks lower.

They do not have to add that 75 ph in reserve.  They will wait one or two jumps let the network grow just a bit and simply slowly add the 75ph in a bit at a time keeping at 30% or 29% or less.

Since growing past 30% become really fucking costly.

[flikflak]

Well, I dont disagree. I just meant that:

...so  you start to compete against your own hashrate.

this is always the case. Even growing from 0.5% to 1% of the networkspeed. Those additional 0.5% are only 0.4975% worth then . This goes on and on as long new or better gear hits the network.

[philipma1957]

Well, I dont disagree. I just meant that:

...so  you start to compete against your own hashrate.

this is always the case. Even growing from 0.5% to 1% of the networkspeed. Those additional 0.5% are only 0.4975% worth then . This goes on and on as long new or better gear hits the network.

Yeah it can be hard to be fully clear on the net without crazy long posts.

You are correct that  from .5 to 1 you pay penalties

[flikflak]

Your numbers are much more impressive than my humble ones - I have to admit that . To push a pool from 20% to 40% takes a lot more than just double the pool speed, because the networkspeed is rising with it.

Thanks for showing this - I've never thought about it until you mentioned it.  

[lolxxxx]

reserved.

[Quickseller]

From the looks of it, the difficulty might actually decrease a little bit this period.

This would probably be a nice change for those who are mining after 3 of the last 4 were double digit increases and the single digit increase was over 8.7%.

I am curious to know what the projects are for the difficulty change for the year, especially considering the reward halving that is coming soon.

[d57heinz]

Well, I dont disagree. I just meant that:

...so  you start to compete against your own hashrate.

this is always the case. Even growing from 0.5% to 1% of the networkspeed. Those additional 0.5% are only 0.4975% worth then . This goes on and on as long new or better gear hits the network.

Yeah it can be hard to be fully clear on the net without crazy long posts.

You are correct that  from .5 to 1 you pay penalties

i agree with this completely.. But there is also the factor of being able to produce asics at cost vs us who have to pay retail.. So the difficulty for manufacturers to increase their share is not nearly as difficult as it is for us.. We are at the short end of the stick on this one

Best Regards
d57heinz

[philipma1957]

Well, I dont disagree. I just meant that:

...so  you start to compete against your own hashrate.

this is always the case. Even growing from 0.5% to 1% of the networkspeed. Those additional 0.5% are only 0.4975% worth then . This goes on and on as long new or better gear hits the network.

Yeah it can be hard to be fully clear on the net without crazy long posts.

You are correct that  from .5 to 1 you pay penalties

i agree with this completely.. But there is also the factor of being able to produce asics at cost vs us who have to pay retail.. So the difficulty for manufacturers to increase their share is not nearly as difficult as it is for us.. We are at the short end of the stick on this one

Best Regards
d57heinz

yes they make them cheaper then we buy them.  the only saving grace is power and cooling come into factor as cost.

if you go back a look at all the super players since asics came out.

there were 3

asic miner friedcat
knc
bitfury

all were self mining with more then 20% but not much more then 30%

they all faced the rising cost I am mining against myself issue.

so all of them never really got much over 30% even if they could.  the gain in actual fiat is not as much as one would think.

so right now bitfury is going to slowly  add to match its 25-30%  share.

If knc and others do a big farm it may be in early feb. I hope for 2 good jumps here.

[notlist3d]

Well, I dont disagree. I just meant that:

...so  you start to compete against your own hashrate.

this is always the case. Even growing from 0.5% to 1% of the networkspeed. Those additional 0.5% are only 0.4975% worth then . This goes on and on as long new or better gear hits the network.

Yeah it can be hard to be fully clear on the net without crazy long posts.

You are correct that  from .5 to 1 you pay penalties

i agree with this completely.. But there is also the factor of being able to produce asics at cost vs us who have to pay retail.. So the difficulty for manufacturers to increase their share is not nearly as difficult as it is for us.. We are at the short end of the stick on this one

Best Regards
d57heinz

yes they make them cheaper then we buy them.  the only saving grace is power and cooling come into factor as cost.

if you go back a look at all the super players since asics came out.

there were 3

asic miner friedcat
knc
bitfury

all were self mining with more then 20% but not much more then 30%

they all faced the rising cost I am mining against myself issue.

so all of them never really got much over 30% even if they could.  the gain in actual fiat is not as much as one would think.

so right now bitfury is going to slowly  add to match its 25-30%  share.

If knc and others do a big farm it may be in early feb. I hope for 2 good jumps here.

I would also add Innosilicon to major players their A1 they sold ton's of chips.  There were lot's of 1T and eventually higher TH  models.   I still suspect in some low priced data centers in china you would find these.

Also bitmain does not make it as noticeable with having a big pool.  But I think they have to be on list it's hard to tell how much of antpool is internal, we likely will never know.

[adaseb]

Looks like today will end up +2-3% or so. Still nothing to worry about yet

[Mikestang]

Looks like today will end up +2-3% or so. Still nothing to worry about yet

Yea, except that 2-3% today is the same as 10% about 8 months ago, and that's just crazy!

[adaseb]

Nice

Today 144 block were solved and 144 as expected which means +0%

About time we get a break from these difficulty increases.

[notlist3d]

Nice

Today 144 block were solved and 144 as expected which means +0%

About time we get a break from these difficulty increases.

Love to have news like this much better what I thought this week would be with day's like this.  But very good day with that.

Now only if price of BTC could rise from 430's seems like we are kinda stuck there.