Topic: Network just jumped up 100 petahashes!!! (Read 5561 times)

KNC was overtaken by Bitmain and Bitfury long long ago. Bitmain is the largest supplier of the miners in the world.

KnC has had no significant hash rate increase in a LONG time in their in-house pool.
 Dunno if they're having yield issues on their announced "solar" chip, or if they were making bogus claims of "when" AGAIN, or if they're in a major financial bind and can't buy enough of them to matter (one post somewhere on the forums here mentioned their bank "freezing" their account).


 I suppose I should list them as 7) though.

Where are KNC on the map here? Thought they are a major player in this game?

We've seen bigger diff adjustments in the past, just not in the last couple years or so.

 I expect it's a combination of factors:

 1) Bitmain continuing to build S7s as fast as they can, deploying them to their own farm 'till they can sell some of them.
 2) Avalon getting 6 production ramped up, then see #1.
 3) Possibility BitFury has gotten it's new 14nm gear into full production.
 4) Possibility Spondoolies has gotten the SP50 into production.
 5) Possibility BW.com got the B-Eleven into production earlier than expected.
 6) Long shot Innosilicon has gotten the A3 into production months earlier than estimated.



 Generation change is happening in miners, always fuels a high amount of new hashrate for a while.

combination of many things, all at once, more miners, another new company jumping in? and diff error a big %

Most likely Antpool.  They have almost 25% of the entire network.

Anybody got any explanation for this jump ? Its like nothing ive seen in the past. Im quite shocked actually.

942,302.93 TH/s  24 hour high   looks like about 200 peta added in last 24 hours.. some could be variance  but damn  thats a hell of a swing!..



Price   416.95 USD/฿
Hashrate   942.30 PH/s
Activity   120 TX/min
Unconfirmed Txs   1,836
Fee Rate   0.00015504 BTC

https://chain.so/BTC

you can see here the expected growth has no signs of stopping soon either.. even the 14 day growth is headed to 1 % per day!!

i believe your right.. 1 exa easy  just how much more from there is the question?

Best Regards
d57heinz

913,340.18 TH/s new 24 hr high reported by wallet
diff at this speed 127,592,148,631.51000000

what a day today

best regards
d57heinz

make that 927,194.71 TH/s

every body need to mine before the hafing into 6 months or soo in this case the Hses growing its normal but will come to less gorwth after the halfing

We will easily push a 1EH network within the next month at this rate.

Wonder if we hit the 100 difficulty on this difficulty adjustment...

869,363.67 TH/s   for the 24 hour high.. diff at that speed once adjusted.. 121,448,702,034.58000000

just keeps creeping up and up

http://bitcoin.sipa.be/speed-lin-10k.png  500 block avg
http://bitcoin.sipa.be/speed-lin-2k.png    200 block avg

new high    858,525.92 TH/s

 


when the network steadies there it will be at diff  119,934,685,196.55000000..


well have to wait it out to even think of buying any hw...  i learned a hard lesson with the s1 as i chased not only diff but the price dropped aswell.. Sucks to be squeezed so hard so fast

best regards
d57heinz

i agree completely.. I think this push is to drop off all the old gear.   And to preorder now i think is a bit risky to say the least.. miners will be 20 % less by end of jan or maybe more if diff keeps up like this..

Best Regards
d57heinz

well price is over 410 and does support 100 diff and we are in the middle of a huge amount of hash.

 So 100 by new years will happen and if you look at this price offered for cloud s-7's


https://bitcointalksearch.org/topic/lee-group1150-the-group-buy-of-hosted-bitmain-s7-473t-online-on-27th-jan-1283287


I can see more on line.  By the end of Jan  say a diff of 120.

A diff of 120 with a price  of 400 usd  starts killing off some gear.  S-3's   lose money at a diff of 110 and 10 cent power

s-5's lose money with a price of 400 usd a coin  a diff of 160 and 10 cent power

So maybe we push up near a diff of 120 kill off all gear  worse then .75 watts a gh  then slowly rise. At a certain point  the older gear drops off and the diff flattens.

i predict diff over 100 possibly by end of month at the earliest.. latest would be  mid jan..   We already past 660 peta.    Source bitcoinwisdom.   so to go on the method that seem to work for me.. the highest 24 hour peak that i seen was 740 petahashes.. So i expect within a couple more diff changes that we be around 105 bil diff.  Current 24 hr hashrate is.. 740,930.10 TH/s ..

Best Regards
d57heinz

edit   new 24 hour high 748,122.41 TH/s

Make that  786,558.35 TH/s   Damn.. when is enough enough!!!  at what point do we say thats enough hash to secure the network. Come on now.. at some point id say its overkill..

Bitmaintech will sell s-7's as long as people will buy them. 

It will be 100ph by march 1st.

The op has gotten it right  faster then I did.  Builders of asics are going to sell and sell and sell.  And sell more.  diff will go over 100 by march 1st.

The worry needs to be price..  as only one thing will stop the selling of new gear +  huge price drops to under 200 usd .

The hash rate will keep on rising as Bitmain has plan to deliver 40PH/s in the next two months. Bitmain usually delivers.

I did look at new network highs for the last year and in 19 of 26 cases .  when we hit new highs at a peak rate our average was there in under 2 months.

so a diff of 92 bill by 7 more weeks which would be 4 adjustments should be around 70% chance .

   Frankly I failed to see this idea of yours, but it looks as accurate as any other I have seen or thought of.
I could tell you a really long boring story about why I tried to find what I was missing about your idea.  but the short version is when I was dating my wife in 1985 I opened my mouth about a game of Trivial Pursuit  and sounded like an ass.  I was proven wrong in under 30 seconds.  Ever since then I force myself to step back and look at what I said or wrote.
     As for the huge up and down seesaw moves they mean something important but like you I am not sure of the mechanics behind it.

Price   377.21 USD/฿
Hashrate   714.45 PH/s
Activity   125 TX/min
Unconfirmed Txs   4,923
Fee Rate   0.00002170 BTC
Fee paid by unconfirmed transactions for every 1,000 bytes of size   https://chain.so/BTC
 

New all time high being reported by the wallet (from what ive seen) 714,452.51 TH/s  .. I believe its a 8 hour avg  honestly idk.(id appreciate if anyone can elaborate on how bitcoin-cli getnetworkhashps arrives at its calculation?)..  ive been watching this spike since early 2014.. When it would spike to a new all time high according to the wallet it would be at that speed or damn close within a couple diff jumps.. What im looking at is if you look at the hash since day one.. you can see the obvious climb when we had asics come on the scene.. All that with 2 watt a gh gear.. With s1 and then soon after release of 1 watt per ghs gear we hit around 350 ph if i remember correctly.. Then we had s5 release which was .5 watt per ghs  but the net hashrate never doubled.  Now we are at .26 watts per ghs and some companies claiming less than that.. If they only replaced what they were currently running(thats all the big farms) then we should be seeing 700 peta at minimum and possible up to over 1 exahash if they all get to .25 watts per ghs..  I have felt this huge jump is LONG overdue imo.. Im not here to start a rival thread .. i feel the one you all do is just fine.. I just didnt appreciate being jumped on for speculating in a speculation subforum.. But anyway.. we are looking at another 10 % jump this time.. i feel that if manufactures run diff up too fast it will leave to dismal sales.. But i am not ruling out the theory of ghost farms .. hell could be testing batches .. or could be as simple as a ddos on a huge pool.. i dont know whats causing these huge swings..  anyway not here to step on anyones toes..

big jump showing here too.. have to wait cpl more hours to see it on chart  https://bitcoinwisdom.com/bitcoin/difficulty


Best regards
Doug
d57heinz

Not sure about 10% for the next 3 or 4 diff increases, but I'd guess close to that for each - depends in part on when the B-Eleven finally shows up, and what Bitfury and Spondoolies manage to get built too not just Avalon and Bitmain.

All I know right now is that we are in for another 3 or 4 10% jumps at least. Bitmain and Avalon have a few more batches that are shipping out right now and are about to go online in force all next month. With the recent price rise people will turn on old gear as well. I don't know where we will go from there but I think we will see a massive price drop in gear and then another round of 10% or more jumps as that gear ships.

As a miner I really hope I'm wrong, but I don't think I am. If I could take back my last purchases I probably would because with the 10% or more diff increases the ROI is DOA. 

I use to think this way and in fact got into a fight with op accusing him of spreading FUD.  

  I now am thinking if you look at a 1 day peak in terms of peta hash  most of the time that peak ends up the new norm in under 60 days.
Not always but more then 65% of the time. So if we hit 606ph as a one peak 606ph will be the norm by mid January pretty much ⅔ of the time  that will be the case.

So peaks have tremendous future value as a predictor. But not so much current value. After reviewing all of op's posts  I think he has a good method for telling us the network ph 60 days from now.

Peak numbers mean NOTHING, there is a very large random luck variation inherent to Bitcoin mining.

 It's like saying "I can hit 27mph in a sprint when the wind gusts hard enough at my back, so I can run a Marathon in under an hour!"

The point is, you can't look at such a small sample size and make any kind of useful determinations from it.  For example, if you look at just the previous 2 blocks (as of this post), it took the network 20 minutes.  So according to that, the total network hash rate is exactly on target for the current difficulty.  However, if you had made the same comparison about an hour ago, you would have seen that the network found 2 blocks within seconds of each other, so the hash rate was astronomical.  This is precisely why the network waits 2016 blocks to adjust... and even with that many blocks, it's still pretty much a SWAG.

If you only look at the last 6 block, it doesn't represent the reality at all, it may be variance or it may be gear. As for now, we are pushing new gear into the network but saying out loud we just it 660 Peta is pretty much a very long shot to be 100% sure someone just throw 200 peta into the network and doesn't keep it online for x reasons.

It's like looking one feet away from you to predict the future ... He may be right, but it's a long long long shot. He should solo mine ;-)

yeah  when diff  jumps used to be 10% or more I guy would piggyback on my thread a bit.    but when diff  jumps got lower  he stopped as over under was too easy.

I know I knocked heads with op with diff on my threads last week but in a calmer moment I went back and read everything we said.  I decided I was wrong and misread him.

I think his point is peak numbers have more meaning then you may believe and once a peak is hit  the network will reach that as  a norm in the future.

He could be correct and maybe his 606 hashmark or around 92 diff will come faster then I would have thought.

Wasn't there a guy running a long-term betting site?  You could bet on BTC price/difficulty at some point in the future... I swear I remember that.

Would be kind of fun... the long-term speculation thread.  Where do you think difficulty and price will be by year end?  By the end of Q1 next year?  Predict the date/time of the halving... all kinds of fun things you can speculate and bet on .

he should do one.  we can have a rivalry.

His style of prediction is vastly different then mine. Fresh thinking could jazz us up a bit.

 But I like the idea of a rival thread.

 Here is my Jan 31st prediction Diff is under 92 bill if price is under 375 usd.

@ op I really do think you could do one  and I don't mean this post to be sarcastic or negative.

Maybe your peak number has more value then people think.

It would be interesting to see how long it takes the diff to catch the peak.

looking back at the winter of 2014, we can say that btc price is more important than cold weather!

You're kidding yourself if you think weather doesn't play in even for the big boys - chinamine is a perfect example of this.

You're STILL looking at peak rates - the long term rate is still well under 500, though the increases since your original post might push to 500 as a non-peak average in another couple weeks to a month.

 We are NOWHERE NEAR your claimed 606 as anything but an instantanious PEAK rate.


 There is already a recurring thread predicting hashrates, no need to start another one.



 Winter has very little to no effect on major farms - they MIGHT be able to push their hardware a little harder, but it's NOT a big difference as they have serious cooling efforts already in place

.

yeah,  we have reached 1000000000peta atm!

Maybe Hashnest is testing their new Antminer S8

Northern hemisphere is heading towards winter and so the minimal activity from home miners in the region will increase.  I'm not sure if the weather really alters plans for the big mining farms though.

well so its been some time since this thread.  I got bashed for it being variance.. But here we are.just like i predicted.. Over 500 peta hash for the network.. Now we are pushing 600 peta.. Anyone care to start a thread where hashrate will be next month.. I predict 580 peta.  Hashrate   606.32 PH/s  this is where we are at atm.  So you can expect just like i thought last month.. When you see this spike expect within two to three diff changes it will be at or past this in network hashrate..  maybe i start a thread that involves predicting network hashrate

Best Regards
d57heinz

alright i see what your saying.. Ill sit tight and watch for a few weeks more and see where we stand.  Thanks for the input guys.. Ive got a lot more to learn it seems..  Thanks again

Best Regards
d57heinz

Because their block got orphaned. Actual block 374219 that ended up in the main chain is https://blockchain.info/block-index/991377/00000000000000000e8a1c90a62c1dbed6d730ffde34ff981dbd9ab3ee9dd5e9 (by F2Pool)

Click on the first transaction, then click on "Show scripts & coinbase" (if not already shown):


Not sure where you get the notion that they're turning things on and off; their block interval is fairly consistent (remember variance).
https://www.blocktrail.com/BTC/pool/btchina/1

https://blockchain.info/tx/91e7cfb8dc8ace823cc0b7557f562a3a52166545a6b4e2b0083768c56d4956b6

now if we can just figure out who this is and why they have a double spend .hmmm

looks to be btcchina pool

Best Regards
d57heinz

i understand.. well we will see.. https://blockchain.info/address/1MG3LEnxt4qBewi3BCy5Q6M2jCqgiDdtTC  here is the miner turning on and off..

Best Regards
d57heinz

Bitcoin hashrates are a vague estimation by the speed between block. There is something called luck or varience. Varience allows blocks to be mined either longer than 10 minutes or shorter than 10 minutes, it would never be a consistent 10 minute. Hence, Bitcoin difficulty is determined by the average speed. Another possibility that cannot be ruled out is the testing of shipping ASIC miners. Antminer S7 has been a thing and they might just be testing the units before shipping hence the sudden increase in speed.

BTC and "get rich in a month pump & dump altcoin scams" are very different, but clearly you're allowing your experiences with one of them to jade your outlook on the other.

You say you were a BTC believer, but that you invested in bs altcoins, have lost money on BTC and scams, and yet point the finger at "the greedy ones" for fucking the valuation up.  Why did you do all those things if not out of greed? People are going to do what they do, it is human nature.  Hardware manufacturers are going to produce products that sell, people will buy products they believe will make them money, and a small minority are going to try to take advantage of others for profit.  These are the constants in life, you just need to be aware and prepared for them. The ones that fall for the altcoin and other scams are the ones that typically are greedy, and allow that emotion to cloud their judgement.

Most of us are not trying to lure people in to an immature currency, personally all I do is educate people and let them make their own mind up.  Those that invest with the expectation that they will make xxx% return on their investment are the ones that end up sour and disappointed.  Those that stay in it for the long term will be the ones that end up alright, but to think we are going to see over-night millionaires the way we did in late 2013, is pure fantasy. The long term trend of bitcoin is upwards, and IMO the valuation at present is "healthy" from a market point of view.  The halving will definitely be interesting, and I look forward to see what it will bring to the market.

My point is that you're making a thread talking about giant swings in hash rate and claiming something that simply isn't true.  The network has not added 100PH/s of gear.  Look, I can take an incredibly small point in time (say between two blocks) that are found seconds apart and claim the network jumped to 3EH/s.  Nobody in their right mind is going to believe that for a second.  All you've done is try to make it seem a bit more plausible by using a "relatively" innocuous number... enough to be scary, but not enough to be unbelievable.  I simply showed with actual numbers that your facts are misleading.

Is the network adding hash rate?  Trends say it is.  Are we going to see a 25% increase this adjustment?  Not a chance.

Im not in the least bit stressed. Moreover i feel some duty to warn others as i was not.  Those that have elite status should be airing more on the side of caution when encouraging more people full into a system they know damn good and well is immature at best.. So many are being burnt that BTC is becoming the game of hot potato.  dont be crying when the music stops..


Well you are both entitled to your opinion. Mine has been stated.. Lets see where we are at in two months.. Lets see if this is really variance or just huge farms gettting things swapped out..
 just sit back while it all goes to shit  just keep buying out the bag holders/feeding the hw manufactures.. "Everything will be just fine" is not something i take very seriously. I think that mentality is what put our countries in the state they're in ... NO?.. Tell me this  how would bitcoin be doing if we didnt have a steady injection of 1 million usd every day(where do you think that is coming from?).  What would its value be then? this is turning out what looks to be a good way to transfer the wealth of those that "believe" into the ones that understand the system.. Please explain how everything is ok for those that continue to invest and it constantly goes down  in value.. I guess we are getting close to the point where the system whether it be btc or the many other alts that enough people have been burnt now that they are Starting to see the light so to speak.. Personally i expect the halving to be the nail in the coffin as many more will get burnt thinking they will double there money. when in reality they got there doubling with the efficiency of the miners that were just released... Now that we have short sells involved with no checks and balances.. It will get shorted all way to 0 once the big boys acquire enough coin..  I was once a true btc beleiber but having been burnt by so many scams and bs alt coins. its time for a reality check here.. Bottom line at this point before you start mining you best learn to read charts as this will inevitably be pushed much lower.. Im only stating my opinion and take it fwiw. I dont claim to have full understanding.. hell maybe i dont understand it at all...   on that note would love for someone to clear me up on this.. Is btc turning into a ponzi in disquise? Is this why satoshi bailed?.. I feel the great thing to come of all this is the blockchain.. The greedy ones will fuck it up for being a currency i can guarantee that!.

Best Regards
d57heinz

Best advice of the day ;-)

Sit back, relax and take a deep breath.  Since the last difficulty adjustment, there have been 1547 blocks found.  It has taken 889775 seconds between those blocks.  This means an average of 575.16 seconds per block.  The bitcoin network states it should be 600 seconds per block.

If we do some math, we can see that the average hash rate of the network is as follows (using the shortened formula, which is accurate enough for this exercise):

Worrying as you are over 100PH/s swings is causing you too much stress.

even higher still.. wtf

Price   227.81 USD/฿
Hashrate   522.28 PH/s
Activity   76 TX/min
Unconfirmed Txs   541
Fee Rate   0.00017055 BTC
https://chain.so/BTC

https://blockchain.info/charts/hash-rate

Finally broke 500 peta for first time on blockchain chart.. if you look at that chart seems once we hit a top we will be at that speed in less than two months .. happens many times in the past..

Best Regards
d57heinz

Price   230.48 USD/฿
Hashrate   513.93 PH/s
Activity   55 TX/min
Unconfirmed Txs   98
Fee Rate   0.00014839 BTC

 They must of decided that dropping the price of btc was a little less noticeable/easier lol.  Well you can bet on them reaching 500 peta.. That was just a short glimpse of whats to come this winter(maybe sooner).. i know damn good and well its not just variance. and besides something got shut down as we are lower now in global hashrate than we were at last diff adjustment. If you think you can gain one up on them at this stage of the game your dead wrong. Unless you have free power. I guarantee in 1 years time there will be a new more efficient miner to come. then of course everyone will have to have that otherwise your stuck again.(being squeezed out). Its a vicious game.we are losing more and more as they push roi to the extreme at one year. We all know at this point you best be dedicated as you will be spending your hard earned money to keep this afloat.. I wouldnt doubt that bitmain doesnt take half the money of the miner and keep it in btc just to keep the price of btc afloat a little longer..  Have anyone else notice right about the time the miners go on sale they price go sideways or up or they know that they have finished installing all the 16 nm and its time to dump the old on us before the diff skyrockets.    its no doubt they want you to think its a good investment.."hell bitcoin is rising" youll think," diff is steady last 6 months", Then after sales it seems to crash hard.or diff jumps up .  they pulled the trick one to many times for me not to notice..  Looking like a good time to pull out and sell off what i have and just sit on the bitcoins. If they ever do anything then great.. But i cant keep throwing money at these hw manufactures and then hope to get it back ..  At this rate of return.. its almost making the banks looks like a good option again.. It was only a matter of time that the greedy ones fucked it up for the 99 % of us.

Best Regards
d57heinz

with that thing they can mine with profit, up to at least $50 per bitcoin, this mean that the hashrate will jump much higher, stay tuned

i'm wonreding what will happen if the efficiency keep growing like crazy but bitcoin price remain there, i guess, even more centralization...

Hash Rate    390,730,448.58 GH/s https://blockchain.info/stats
it's called variance

Battlecruiser OPERATIONAL

And by battlecruiser I mean S7 Bitmain miner.

well look at this chart 168 blocks  vs 144 or  about  16.7% but only for 1 day have to see how long it lasts this high


http://btc.blockr.io/charts

I understand that it needs to be averaged out. All i was stating was that it jumped 100 peta.. if you look at the graph in the posst before yours you can clearly see it.. 380peta to well over 480.. That is all i was pointing out.  Will it hold above 465 thats left to be seen.. But so far i havent seen it drop below that.. Mostly hovering around 475 average.. So i think your right on cue with the 5 % each diff increase..

Best Regards
d57heinz

Nope, not that big of an actual jump - looking at "short term peak" rates is VERY misleading as they routinely bounce around a lot.

 Got to average over at least a full day to get a meaningfull look at the network hash rate.

 With that said, the ballpark 1.6% average diff increase of the first half of this year is obviously history, looking like 4-5% or so on average is going to be the norm for a while - so much for RoI on anything without VERY cheap electric if you haven't had it for a few moons or longer already.

Here you have nice chart showing computed hashrate within various timeframes:


And more timeframes are available there too http://bitcoin.sipa.be

Looks like blockchain is finally catching up
https://blockchain.info/charts/hash-rate

Best Regards
d57heinz

That is also on my mind aswell.. I ran s1 last year one in each room to heat my house and two other households.. Now Im sure that this year it will have caught on even more.. I will run s3's for heat.. So im doubtful for those living in colder climates if they will even shut off being as it will offset heating costs.. I live in central illinois so heating season is very close.. I think we see some big gains then a possible falling off big time near spring 2016..  Then you have the halving to consider.. Its getting very complicated thats for sure:)..  Thanks for all your insight guys..

Best Regards
d57heinz

I really don't want this 3 day's in  a row. Summer mining was so nice 2 percent almost every week.  No huge jumps.

Winter mining will get interesting if this continues.  We might see some big jumps.

3 days in a row and we will know for sure.

Did a little more research  and zeroblock confirms the steep rise aswell.. Im starting to wonder if the backlog of transactions impairs bitcoin core ability to accurately measure the hashrate.. OR we have a shitload hash coming on.. Im hoping its the first one 

Block Time 8.42 Min
Next Difficulty 6.3 Days
Hash Rate 484,165,991 GH/


https://zeroblock.com/

Best Regards
d57heinz

Phillipma.. Thank you.. I knew you kept close tabs on the situation.. Thanks again for chiming in on this..

well she is back to 470 again.. i think ill keep an eye on this for a while.. something is def going on..

Best Regards
d57heinz

I'll show you 24463!!!

Based exclusively on blocks 373857 and 373858

*reads thread* You don't say

show me 500!!!

https://blockchain.info/charts/hash-rate

any graph or something concrete?

variance...

the real time number above is about 10 hours old.



https://blockchain.info/blocks


373785 (Main Chain)   2015-09-10 00:03:00

373879 (Main Chain)   2015-09-10 14:09:00



in the last 14 hours  107 blocks were made   we should make 84 or 85    so this is a   + 22 or + 23 


+ 22 or + 23 means a 167 block day if our luck goes back to normal

if it stays  as high the next 10 hours  like it was for first  14 hours we will do 183 or 184 blocks.

So  go to http://btc.blockr.io/charts   tonight and see if  we stay high .  183 or 184

if we do this 3 days in a row get nervous.

yes is very clear now.. thank you:)

Best Regards
d57heinz

Thank you for taking the time to explain it.  I appreciate it..

Best Regards
d57heinz

getnetworkhashps performs a calculation based on the difficulty and two parameters (number of blocks to average, block index from which to get an estimate).  If unspecified, it'll take the timestamp of the last block it knows about, the timestamp of the first block for the current difficulty, and uses the time difference and the current difficulty to get to an estimate of the network hash rate.

See also: help getnetworkhashps

Over time this number will stabilize more - you can already see that in your latest query you get ~460Ph/s, and not 'over 500Ph/s'

Im just trying to figure this out. because when i put in this on my client  it tells me the exact same number as chain.so .. So does the client itself also take the time between blocks?.. Can someone explain how the bitcoin client calculates the hash of the network?  Is it block by block or does it calculate back x amount of blocks took x amount of time then take the average?

root@d57heinz:~# bitcoin-cli getnetworkhashps
461495597059767552


Price   239.08 USD/฿
Hashrate   461.50 PH/s
Activity   85 TX/min
Unconfirmed Txs   342
Fee Rate   

There's quite a few days in that graph   I was referring to the disproportional nature compared to other pools.

If all pools in the graph stay at the same hash rate, then in that graph of mine they should all basically have horizontal lines; they can vary up and down a bit, but in the large averages (wider lines) this should smooth out pretty well.

If all pools go up by 25%, then that wouldn't change - they would still all be horizontal.

So when you see one (or in this case, two) graph line(s) go up quite a bit while all the other ones go slightly down, then whoever's behind those lines are adjusting their hash rate disproportionately compared to the others.  You can see this in the more extreme in a weekly graph:

You can see BitFury's sudden rise quite clearly, while BTCChina's is more of a slow but steady growth.  You can also see BTCGuild shutting down ~10 weeks ago by its sudden plummet to 0 blocks.

And a statistic based on 7 days is probably a good thing.
There's several parties on Twitter that point to blockchain.info or blocktrail.com's graphs for the past 24 hours and saying "look! we're #1".. sure, for that day.. sometimes only just a few hours into a day.  That's just misleading for the overall picture.  So while having day-to-day stats or even block-to-block estimates is a fun exercise, it's not very indicative.

By the way: if you look at that first graph I posted - that's analogous to organofcorti's "Diachronic hashrate distribution".  The same side notes apply there: if all pools stay at the same relative hash rate, then the 'bands' in his graph would keep the same width throughout the time span.  Instead we can see that some of them have gotten wider (implying increased hash rate) while some have gotten more narrow (implying either same or decreased hash rate).

i just found this regarding knc


0. KNC Miner has hashrate increase
KNC miner solved 50% more blocks this week than last - about 50% increase over last week's blocks and outside the 95% confidence interval for blocks solved. The stacked histogram (cake chart) shows this gradual increase most clearly.

1. One unknown this week!
Has a solominer returned?

http://organofcorti.blogspot.com.au/search/label/weekly%20block%20maker%20statistics
I follow this guy but he only gives past weeks performance..

Best Regards
d57heinz

Ahh ok i see.  Thanks for clearing that up.. So 25 % spike is normal.. I have seen my miners jump that high before but for the entire networks miners to jump all at once is something..  Ill keep an eye on it.. Would you mind sharing with me the source of your chart. I would really appreciate that.  Thanks again

Edit.. i mine on slush so i feel that the only reason they are disproportional is because they are recovering from a bad streak over the last couple days.. We was low on blocks for a bit there.. As for kncminer be nice to have some stats on them other than only being able to see there block history..

Best Regards
d57heinz

What jonnybravo0311 meant is that spikes are normal as a result of variance - not as a result of people turning huge clusters of miners on/off.  These spikes are also going to get larger and larger in absolute terms the more the difficulty goes up, but relatively speaking stay about the same.

The only parties climbing disproportionately to the rest right now are KnC and Slush;

Thanks i think i know what you mean by spikes..if they only turn it on for brief amounts of time which it looks like thats what they are doing.. i have seen this many times in the past.. the network will hit 500 peta sooner than you think.. Seems they like to milk it as not to scare off sales..  as for those other sites. they are slightly behind. ill give them a bit to catch up and see where they end up  .. Also by your theory if they are taking the time between blocks shouldnt we be slowing down big time.(global hashrate). since the network is throwing some well over 10 min blocks..

Thanks for the help.. Im still learning. 

Best Regards
d57heinz

I would certainly not trust the stats reported there... especially since no other site I've seen reports anything near the 100PH/s jump you're stating.  For example, bitcoinwisdom.com shows us still around 420PH/s.  Blockchain.info shows 460PH/s.

Remember, the network hash rate is estimated by looking at the time between blocks and current network difficulty.  Seeing spikes is certainly normal and expected variance.

so we have been at approx 400 peta for quite some time.. topping out at 420 at times.. Now we are at over 500 peta in a matter of a day.. This is either bitfury new chips or this is the effect of bitmain announcing selliing of new hw and the other manufacturers are going to keep up.. Here comes the big asic fight between the manufacturers.. Not sure if this is the best time to buy coins or just sit back and watch what unfolds. What does the community think of the 100 peta jump?

Price   238.80 USD/฿
Hashrate   504.20 PH/s
Activity   61 TX/min
Unconfirmed Txs   3,775
Fee Rate   0.00017907 BTC


Sources
https://chain.so/BTC
https://mining.bitcoinaffiliatenetwork.com/index.php?page=dashboard     if you have an account here still i only use this pool to watch global hashrate..

Best Regards
d57heinz