Diff thread Nov 11 to Nov 24 Set up stage-picks are now closed!

[HerbPean]

It's been pretty much the same drop at the end of almost every diff since a few months back ... the last 2-3 days .... we see a big drop in hashrate.

[Mikestang]

It's pretty consistent to be a coincidence, too.  I wonder if the larger mines do this on purpose so that the next adjustment is slightly lower, thereby making their operations a bit more profitable for the next adjustment period.  Or is that just a paranoid delusional idea? 

[philipma1957]

don't know but today we are up with a strong number.


384485 (Main Chain)   2015-11-20 16:13:14      16 hours and 13 min = 97 blocks

384365 (Main Chain) 2015-11-20 00:10:17


but we did   121  not 97  this is too big not to be a power gear power down gear cause.

it is over 24.7 % but yesterday at    131 vs 144 is  -9.02%


that is really not variance  that is power on power off.   It is a flux of over 120ph

[notlist3d]

It's pretty consistent to be a coincidence, too.  I wonder if the larger mines do this on purpose so that the next adjustment is slightly lower, thereby making their operations a bit more profitable for the next adjustment period.  Or is that just a paranoid delusional idea? 

I don't know for the operation to shutdown for 3.5-4 day's with a lot of hash that seems like a lot of lost profit.  For most operations I think they would just leave it on.

Only way I could see it happening is if it's a maker of miners.  If a big one wanted difficulty not to show huge on change as it might effect sales. So they would have a reason if sales of miners go down for some big changes.

But I don't see anyone shutting down honestly for this amount of days to effect difficulty.   But I could always be wrong this is speculating.

[skuser]

It's pretty consistent to be a coincidence, too.  I wonder if the larger mines do this on purpose so that the next adjustment is slightly lower, thereby making their operations a bit more profitable for the next adjustment period.  Or is that just a paranoid delusional idea? 

It's pretty consistent because it is variance. Shutting off miners in order to lower next adjustment will actually make your operations less profitable unless it causes difficulty change going really negative. Only way to make most money is to mine with everything you have without disruptions. Exception is selling the miners to public when the ROI starts to be boring

Nice view of variance can be seen on this chart. It happens over and over again and it's being denied by public over and over again
http://bitcoin.sipa.be/speed-lin-2k.png

[philipma1957]

It's pretty consistent because it is variance. Shutting off miners in order to lower next adjustment will actually make your operations less profitable unless it causes difficulty change going really negative. Only way to make most money is to mine with everything you have without disruptions. Exception is selling the miners to public when the ROI starts to be boring

Nice view of variance can be seen on this chart. It happens over and over again and it's being denied by public over and over again
http://bitcoin.sipa.be/speed-lin-2k.png

Not true if government subsidies are involved and if you don't think China's government is involved you could be correct.  But if you think China's government is not involved

or that the are not capable of direct power subsidy to bitmaintech or avalon  look at this

http://www.businessinsider.com/60-minutes-chinas-ghost-cities-2013-3


there is enough power in the cities above to provide for all the mining equipment in the world.


I can think that  5 guys in five of the cities above shunt power to mines in china for next to 0 cost.  the catch is they can not do it 24/7/365

So they cut a deal with an asic builder  and the builder has mines that turn on and off. 

How hard is it for this to be the reason vs variance?  Or better  yet both  my case example of power on and power off.  Plus your idea of variance.

[Jamphone]

How hard is it for this to be the reason vs variance?  Or better  yet both  my case example of power on and power off.  Plus your idea of variance.

Phil, you'll remember that I was extremely skeptical about the PHashes that were sitting around offline and able to be turned on if the price jumped. I plainly refused to believe there were tens of MWs sitting around that could be switched on, but then we had a quick run up here. (I had figured there would be less than 5M.W

We saw 100PH (on the 3.5 day average) come online very quickly in only eight or nine days. The most efficient equipment, and anything below .7w/gh, was already online, maybe higher. That means we saw a spike of around 70+ MW if you attribute it all to old gear. There was probably a fair bit less due to variance and a lot of new equipment, but that still means 40MW or 50MW of old equipment was thrown back online and now appears to be coming off. Even Bitmain can't crank out 20PH in a week.

I'll admit it, I threw 10kW of old gear on there (1.1w/gh), and actually haven't taken it off yet since it's still above break even for me at this price, although we'll see what difficulty brings.

So, there wasn't a ton of gear ready to just be switched on, but there was plenty of capacity and machines near enough by that people through on a lot of extra online in just a few days.

[philipma1957]

How hard is it for this to be the reason vs variance?  Or better  yet both  my case example of power on and power off.  Plus your idea of variance.

Phil, you'll remember that I was extremely skeptical about the PHashes that were sitting around offline and able to be turned on if the price jumped. I plainly refused to believe there were tens of MWs sitting around that could be switched on, but then we had a quick run up here. (I had figured there would be less than 5M.W

We saw 100PH (on the 3.5 day average) come online very quickly in only eight or nine days. The most efficient equipment, and anything below .7w/gh, was already online, maybe higher. That means we saw a spike of around 70+ MW if you attribute it all to old gear. There was probably a fair bit less due to variance and a lot of new equipment, but that still means 40MW or 50MW of old equipment was thrown back online and now appears to be coming off. Even Bitmain can't crank out 20PH in a week.

I'll admit it, I threw 10kW of old gear on there (1.1w/gh), and actually haven't taken it off yet since it's still above break even for me at this price, although we'll see what difficulty brings.

So, there wasn't a ton of gear ready to just be switched on, but there was plenty of capacity and machines near enough by that people through on a lot of extra online in just a few days.

good point a lot of older gear turned on and off makes sense too.

[UfoRia]

good point a lot of older gear turned on and off makes sense too.

I know people at work that turned on old gpu machines when it spiked. I don't see someone sitting on 100ph idling waiting for the market rise, I see a shitload of people see the price spike and go to their attics/basements , dust their old friends off and fire them up. I would lean towards that as the spike to be honest, so I agree with phil.

I bet if the price spikes to $400 with resistance that there will be a huge spike again that lasts longer.

Fun times


Ufo

[philipma1957]

okay day 10

http://btc.blockr.io/charts

11-11-2015   169 blocks  this is a plus 25   now the first 77 blocks were on the last adjustment  So  on a part day I guess +14

11-12-2015  158 blocks this is plus 14    total is +28
 
11-13-2015   161 blocks  this is plus 17 total is +45

11-14-2015   146 + 2                               +47

11-15-2015   156 is +12                           +59    

 11-16-2015  172 is +28                           +87

11-17-2015   146 is +2                             +89   and plus 2 looks nice

11-18-2015   156 is +12                           +101

11-19-2015   131 is -13                            +88  same drop about the same time last diff adj

11-20-2015   169 blocks or + 25               back up to a very big  113 over.

1525 blocks made should be  1412    which is +8.00%


https://bitcoinwisdom.com/bitcoin/difficulty (+7.73%)

Bitcoin Difficulty:   65,848,255,180
Estimated Next Difficulty:   70,935,427,608 (+7.73%)
Adjust time:   After 491 Blocks, About 3.1 days
Hashrate(?):   499,662,832 GH/s
Block Generation Time(?):   
1 block: 9.2 minutes
3 blocks: 27.6 minutes
6 blocks: 55.3 minutes
Updated:   23:10 (7.0 minutes ago)



my number is higher then bitcoinwisdom's     and at 491 blocks we are near 3 days.

[adaseb]

I don't know.

I really doubt that people sit and wait by the computer and whenever BTC goes over a certain level they immediately power on-power off their gear.

I also don't believe mega farms shut down gear at the end of the difficulty cycle to get a smaller difficulty, they would lose more money that way.

I know that power is cheaper on certain hours of the day so that could be a theory but that would mean then during certain hours the hashrate increases not a certain day off the week.


I really think that a new ASIC chip is out there in operation and its very buggy so they ran it for a while and then it either burnt out or causes too many hardware errors and they are tweaking setting to get it to work properly.

[philipma1957]

I don't know.

I really doubt that people sit and wait by the computer and whenever BTC goes over a certain level they immediately power on-power off their gear.

I also don't believe mega farms shut down gear at the end of the difficulty cycle to get a smaller difficulty, they would lose more money that way.

I know that power is cheaper on certain hours of the day so that could be a theory but that would mean then during certain hours the hashrate increases not a certain day off the week.


I really think that a new ASIC chip is out there in operation and its very buggy so they ran it for a while and then it either burnt out or causes too many hardware errors and they are tweaking setting to get it to work properly.

I like my China ghost city idea best. If an s-5 is 50% power and 50% profit.  and i am bitmaintech having a ghost city s-5 10ph farm and a 5ph s-3 ghost city mine.

seems like it is not far fetched.  power turns on  when ever it can. since the gear was paid off why not move it to a ghost city and run it when power is available.

[mavericklm]

Estimated Next Difficulty:    70,819,346,021 (+7.55%)
Adjust time:    After 480 Blocks, About 3.1 days
Hashrate(?):    506,027,398 GH/s

better than +1x%!!! 

[notlist3d]

Estimated Next Difficulty:    70,819,346,021 (+7.55%)
Adjust time:    After 480 Blocks, About 3.1 days
Hashrate(?):    506,027,398 GH/s

better than +1x%!!! 

Ok... I need .06 of everyone to turn of miners so I can be right.  Which I never am but it's still fun to speculate.

We do have some miners that have went out.  So it will be interesting if it goes up some more once plugged in, or if they are already plugged in.   But I do agree not seeing 1X change is nice just not to see.

[notlist3d]

I don't know.

I really doubt that people sit and wait by the computer and whenever BTC goes over a certain level they immediately power on-power off their gear.

I also don't believe mega farms shut down gear at the end of the difficulty cycle to get a smaller difficulty, they would lose more money that way.

I know that power is cheaper on certain hours of the day so that could be a theory but that would mean then during certain hours the hashrate increases not a certain day off the week.


I really think that a new ASIC chip is out there in operation and its very buggy so they ran it for a while and then it either burnt out or causes too many hardware errors and they are tweaking setting to get it to work properly.

I like my China ghost city idea best. If an s-5 is 50% power and 50% profit.  and i am bitmaintech having a ghost city s-5 10ph farm and a 5ph s-3 ghost city mine.

seems like it is not far fetched.  power turns on  when ever it can. since the gear was paid off why not move it to a ghost city and run it when power is available.

Just saw a article about this on digg: http://priceonomics.com/surveying-the-ghost-cities-of-china/

There are a lot more ghost cities then I would have guessed.   I guess it makes sense they would have a good amount of unused power in each of them.   So would be interesting if anything is happening in them with mining.

[philipma1957]

I don't know.

I really doubt that people sit and wait by the computer and whenever BTC goes over a certain level they immediately power on-power off their gear.

I also don't believe mega farms shut down gear at the end of the difficulty cycle to get a smaller difficulty, they would lose more money that way.

I know that power is cheaper on certain hours of the day so that could be a theory but that would mean then during certain hours the hashrate increases not a certain day off the week.


I really think that a new ASIC chip is out there in operation and its very buggy so they ran it for a while and then it either burnt out or causes too many hardware errors and they are tweaking setting to get it to work properly.

I like my China ghost city idea best. If an s-5 is 50% power and 50% profit.  and i am bitmaintech having a ghost city s-5 10ph farm and a 5ph s-3 ghost city mine.

seems like it is not far fetched.  power turns on  when ever it can. since the gear was paid off why not move it to a ghost city and run it when power is available.

Just saw a article about this on digg: http://priceonomics.com/surveying-the-ghost-cities-of-china/

There are a lot more ghost cities then I would have guessed.   I guess it makes sense they would have a good amount of unused power in each of them.   So would be interesting if anything is happening in them with mining.

yeah it is a great way to use old gear.

I remember a guy in china grabbed up a ton of s-3's  dirt cheap last spring and was running a 2 or 3 ph farm with them. He said he could use the s-3's because he had 2 cent power.

It seems to me that this old gear paid off sitting in a ghost city could be fired up pretty easy.


http://natocouncil.ca/ghost-cities-in-china/

just think how many s-3's could be turned on and off in this set of buildings.

It could be done with government approval as a way of testing and maintaining the power grids in those cities.

Hey @ bitmaintech   I know you read the thread....  If you are not doing this you should be.  Feel free to send me some coins for the

idea.   https://blockchain.info/address/164o6jGhpciHC8Lg4XiXyprKfrqZK3tST6

[adaseb]

But the gear would have to have been built first to be used in these ghost towns. To turn 100Ph/s on and off, is like 200,000 Antminer S3's. I don't think that many Antminers in total even exist on the planet.



384638 (Main Chain)   2015-11-21 15:20:06   AntPool   0000000000000000033e5efed37742ec7925ff825853ce21fbd53592c1489620
384532 (Main Chain)   2015-11-21 00:02:42   F2Pool   00000000000000000860c5b9429f78ed94db860c347ec72ac0f7cc3a1b3371b8

106 blocks in ~15.5 hours should be 93 blocks instead.


14% difference




Even in the last 20 minutes, 6 blocks were found.

[notlist3d]

But the gear would have to have been built first to be used in these ghost towns. To turn 100Ph/s on and off, is like 200,000 Antminer S3's. I don't think that many Antminers in total even exist on the planet.



384638 (Main Chain)   2015-11-21 15:20:06   AntPool   0000000000000000033e5efed37742ec7925ff825853ce21fbd53592c1489620
384532 (Main Chain)   2015-11-21 00:02:42   F2Pool   00000000000000000860c5b9429f78ed94db860c347ec72ac0f7cc3a1b3371b8

106 blocks in ~15.5 hours should be 93 blocks instead.


14% difference




Even in the last 20 minutes, 6 blocks were found.

Bitwisdom did go back up:

Bitcoin Difficulty:    65,848,255,180
Estimated Next Difficulty:    71,773,382,950 (+9.00%)
Adjust time:    After 404 Blocks, About 2.5 days
Hashrate(?):    518,513,335 GH/s

Maybe some miners got shipped and are being plugged in currently.  It would not surprise me at all.  I still hope we see less than 10 though.

[philipma1957]

But the gear would have to have been built first to be used in these ghost towns. To turn 100Ph/s on and off, is like 200,000 Antminer S3's. I don't think that many Antminers in total even exist on the planet.



384638 (Main Chain)   2015-11-21 15:20:06   AntPool   0000000000000000033e5efed37742ec7925ff825853ce21fbd53592c1489620
384532 (Main Chain)   2015-11-21 00:02:42   F2Pool   00000000000000000860c5b9429f78ed94db860c347ec72ac0f7cc3a1b3371b8

106 blocks in ~15.5 hours should be 93 blocks instead.


14% difference




Even in the last 20 minutes, 6 blocks were found.

  realistically they don't have 100ph just too much, but 5-20ph they could.

20ph in s-1's were built  50ph in s-3's were built   75ph in s-5's were made.

I don't really think they can turn off and on 100ph.  but  5-20ph in 10 spots in ghost towns  they could

and the gear  would be toss away gear s-1's s-3's  how many in a white  van?   100 s-3's in 1 van = easy.

fit them in a building like the photo  and boom 50th   at the flick of a switch.  the antminer psu can do 4 s-3's at an under clock to 350 watts.  so only 25 psu's   a few gigabyte switches and a router.

keep it under 200 per  location and 1 guy or 2 guys can set it up.

[adaseb]

384532 (Main Chain)   2015-11-21 00:02:42   F2Pool   00000000000000000860c5b9429f78ed94db860c347ec72ac0f7cc3a1b3371b8
384667 (Main Chain)   2015-11-21 18:14:30   Slush   000000000000000001b11ef38a2a4d7e30ad9064529ef50e4a77bd1ddb72f099


110 Expected and 135 Found so 23% TODAY!