Bitcoin Forum

after hitting ~4500TH is now at ~3250TH, according to the graph on link. This is a major drop. Are the T machines failing or what is going on? There's always a drop prior to a diff adjustment but this seems to be a bit over expectations. Just wondering

http://bitcoin.sipa.be/

yes, I noted it too some hours ago. really strange.

It looks like a big datacenter switched off its miners.

more then 1 datacenter, we're talking of ~ 1000 1T mchines. Maybe KNC is shipping the machines they have been mining with?   ;D The actual current hasrate is 3657.70 according to http://bitcoinwatch.com/

ghash.io passed quickly from 736.42 Th/s to 933.69 Th/s

really strange...

Probably some big datacenters are experiencing temporary difficulties ?

ghash has a jumpy hashrate, never understood why, maybe their hash calculation formula. bitcoinwatch seems to be right on the money when it comes to hash calculation and if you notice the graph the decline didn't start today, it has been going on for a few days, so no last minute turn off to manipulate the diff

anyway, I think that this "strange" 1000TH computational speed reduction, probably will reduce next difficulty value..

we are now at 510929738 difficulty

When the difficulty goes up it makes some alt coins more profitable for a while so lots of miners switch over. Then the alt coins difficulties go up and everything gets back to normal.

yay !!!!

#Conspiracy #ItIsAllPartOfThePlan #SellMoreMiningHardware
Don't mind me.
I noticed that too. Something is going on......... it's a trap, lol. On the serious note, there is something going on, I just don't know what!

Edit:
I suppose some mining equipment is no longer viable to run. I hope whosoever equipment that was, they broke even at the least.

My humble guess so far is simply some lag somewhere in the reporting of found blocks that confused blockchain.info and maybe a few other places (perhaps there were a couple orphaned blocks here and there?) combined with a bit longer time to find the first 510,929,738.02 diff block making it all look very slow suddenly to those not in front of the miner.

lets see where it goes now, if it jumps to 4.5 it was manipulation, if not, 1000 T machines are somewhere on a plane or they are breaking apart

is it possible they are mining some sha-256 altcoin like TRC?

Luck and/or DDOS attack on some big pools and drop is visible. It's happend before and will be in the future...

If there is a drop in hashing power/miner, the difficulty will decrease and therefore it would be a wise idea to buy hardware.
At this time and age wouldn't it be cost effective for manipulation by a bitcoin hardware seller? What I mean is a organized DDOS in order to make it look like there is less miners, thus people will think it will be easier to mine and will buy hardware?
Then when they shipout the hardware they stop the DDOS.

I hope this is not the case.

doesn't look like it's dropping anytime soon
next estimate is 755745693

where are you getting that estimate from? that would imply a 50% hash rate raise or close to 1800 Tera hashes, according to what i'm seeing so far the next hash rate will be 513. I know this is not final but i would expect around a 25% increase to 625 million (900 Tera hashes added to the network)

That is a pretty intriguing thought.

Or perhaps a trojan, or bot that goes around hacking into mining machines,

either diverting payouts or attempting to permanently damage the board.

That would bring the hashrate down, and the companies wouldn't need

to do a thing, letting human nature take its course.

Realistically, its only a matter of time now.


How many people left their log and pass admin? A certain percentage will.

Hell, I did til just now. ::)

Please remember/note that there is a lot of noise in the hashrate reading.  Also note that the estimated next difficulty is not that accurate right after an adjustment.

The best charts I have seen for this sort of speculation are found here:

http://bitcoinwisdom.com/bitcoin/difficulty

Note the second graph down.  It shows graphically how much the next difficulty estimate varies over each difficulty adjustment cycle.

Also note:

Bitcoin Difficulty: 510,929,738

Estimated Next Difficulty: 656,156,896 (+28.42%)

I believe this to be the best estimate at this time.  

Just temporary, it is going up again... Next generation mining equipment being shipped everyday... HashFast <----

the reason for this post is not if or why the network hash is going up. It will and my guess is around 25% for the next diff. The question is where did those 1000T go. They were on the network so they are somewhere. And if you notice they didn't drop all at once, they dripped slowly. Burn test by a manufacturer? ASIC going online and breaking down? BFL datacenter went up in flames? ;D Well, there are no news on any major datacenter fire in the past 2 weeks, at least in the google english news. The only 2 i can find in the past month are the Gigenet and the NSA fires

What I understand is the network hashrate also have luck factored in. I.e. your miners might be doing 50 GH/s. But when you go to your pool it says 30 GH/s - 80 GH/s. If your lucky you might submit more shares and etc....

Its funny how the hash rate dropped in the days immediately before the diff. change, and then immediately after the diff. change the hash rate jumps right back to pre-dropping levels. It's more than variance. Someone is  'trying' to game the diff. changes.

From https://en.bitcoin.it/wiki/Difficulty


Since the difficulty is adjusted based on the entire 2016 block period, reducing the hash rate during the last few days would have little effect on the actual adjustment value.  The only thing turning off your miner(s) during the last few days does is to reduce your own income at the lower difficulty - which makes no sense.

This ^^^^^^^

No conspiracy here, just some miners switch to other coins when they are more profitable to mine.

Might not be dropping, but it certainly ain't going up right now, not for the last week or so.

maybe this is where the GH went?

http://blogs.wsj.com/moneybeat/2013/11/07/judge-orders-coinlab-to-pay-up-in-bitcoin/

"CoinLab Chief Executive Peter Vessenes has said that the lawsuit was a contributing factor in last week’s bankruptcy filing of CoinLab’s mining unit, which is called Alydian Inc. Mr. Vessenes is also chairman of the Bitcoin Foundation, a nonprofit trade group that promotes the virtual currency.

Mr. Vessenes noted in an email on Thursday that CoinLab itself has no mining operations, but “of course we will comply (with the judge’s order) as well as we’re able.” It isn’t clear if Alydian is still generating bitcoin while it is under bankruptcy protection."

good find.  Either someone shutdown their mining operations (coinlab?), terahash machines are breaking left and right or tera hash machines are on their way to customers. Time will tell

thanks!  and my other thought was knc unplugging and packing up their second batch, which someone already mentioned.  anyone know what hardware Alydian used?  Might we see it hit the market?

Yes, I know and understand this. Hence the emphasis on 'trying'.

Looks like that news may have been the result.

Miners that are unsuccessful at overclocking?

A little early to tell but the current estimate is for "only" a 19% hike at the next adjustment:

http://bitcoinwisdom.com/bitcoin/difficulty

It looks to me that the 504 and 1008 block hash rate plots are very telling for the last 3 or 4 difficulty jumps.  I haven't done enough research to say definitely, nor do I care enough to spend the time, but perhaps someone more in tune with the large scale ASIC market will know...

It would seem that big ASIC batches are being tested used on the customer's dime up to the point of a difficulty jump, then boxed up and shipped out.  Meaning the manufacturers get the benefit of the difficulty not yet accounting for the massive increase in hashing power and the customer gets screwed out of the advantage of being the first on the network with new hardware.

I can't seem to explain otherwise the clear pattern of large jumps in hashrate, a big drop around the difficulty jump, then a slow ramp up (accounting for difference in shipping to customers world-wide) in hashrate.  Call me a skeptic, but heck if I was running one of these mining hardware companies, it would be awful hard to resist the temptation to explain away a one or two week delay in shipping with some BS while taking advantage of the relative easy difficulty.

I'm sure if someone was so inclined, they could do the legwork and either confirm or deny a correlation between rapid hash-rate drop offs and large batch ASIC shipments.  Perhaps even some blockchain sluething to correlate that mining income with a small group of addresses...

I think it could also be some of the early asic hardware being shut down....basically usb erupters are useless, Japalpeno's are just like 25 usb erupters so they are useless as well, do the gen 1 ASICMiner blades still make anything on a daily basis?

I haven't had my usb erupter plugged in for quite a while...it just doesn't matter anymore.
It was only making a fraction of a cent per day...maybe the electricity isn't much either but it isn't worth me thinking about it anymore.
My Saturn is generating a decent amount each day and it eclipses anything put out by the USB or Jalapeno.

There have to be others shutting down their old ASIC hardware as well...and nobody can possibly be running gpu rigs anymore(?)

Also, I think that the hardware market is just saturated.... nothing you can buy makes a profit...and no one has any tolerance for mining at a loss.
 

You forget that there's no point in shutting down your mining rigs completely unless it's broken of course. When your revenue gets close to electricity costs, you're better off selling your rig to someone with cheaper/free electricity. This way hashing power just changes hands, while overall impact on the network hashrate is negligible.

That won't work in practice as most miners have failover pools set in their configs.
Even if all pools are DDoS'ed, there's still p2pool, which is quite difficult to shut down.

But there can be other kind of manipulation.
Say some mining hardware manufacturer has a lot of mining power already deployed and wants to manipulate network hasrate.
It can manipulate network hasrate to a some extent by diverting some of its hashrate to PPS pools while performing block withholding attack.
So they are getting paid while the number of actually found blocks drops.
There are at least 2 pools offering PPS - BTCGuild and gigavps' private pool.
While the fees are significant, it's a lot cheaper than just turning off yor mining operations.

http://eligius.st/~wizkid057/newstats/topcontributors.php

Look at the bottom 2/3 of that list... I don't think a lot of people are shutting down miners.

+1

Huh, hundreds of Th/s switched to other coins? Show me these coins I want to see them being mined to death.

I don't think so, seems to me Bitcoin is the most profitable to mine.....

The overall decrease is exponential growth is fairly easy to explain. Most of the orders currently being shipped were ordered when difficulty was 5-10 times lower. Back then it was easy to fall for the "im gonna be rich" mentality. Now, there is no miner on the market that even if delivered today will make bank. Very few new orders are being placed for miners, and now that KNC has shipped very little new hash speed is hitting the market. BFL, Avalon, ASICMiner are done with new hardware. It's just not profitable enough even at cost.


The next jump will be cointerra. I predicted months ago the plateau would be around 4-5 Phash. We are there.

KNC ships the next batch (probably >2 PH/s conservatively and maybe closer to 2.5 PH/s realistically) in 6 days so calling it a plateau is a little premature.
Then HF batch one is mid Dec a little smaller maybe 0.5 PH/s.
The HF batch two is unknown since the first batch got pushed back (will it be a month late? will it ship right after batch 1?) but it looks to be larger maybe 2 PH/s max.

The rest should be the usually slow BFL dribble and small blocks by the other three (ASICMiner, Avalon, Bitfury) but my guess is they are not going to be significant compared to the existing hashrate and the three events above.

Cointerra is saying "on track for late Dec" but given holidays (vacation time) and the late tapeout, my guess is "late Dec" is code word for "oops at the last minute we just missed it and we are looking at mid January". Regardless of it being late Dec or early January we aren't going to be sitting @ 4 (or even 5) PH/s for that long.

Thank you for this link! The charts here put things in a very nice visual relationship.

It'd seriously wager that it's mostly BFL orders too. For a long time they were the only company taking unlimited ASIC orders. And if you look at their shipping update blog, they were cranking them out like no tomorrow for a while, but have recently slowed down shipments.

Apparently most people think there is some big meter attached to the network that measures the global hash. This is not true at all.

When you go to websites that tell you the global hashrates, they are giving you the information based upon a calculation. The look at the difficulty (a known number) and the time between solved blocks (a known number) and they calculate an approximation of the hash rate required to perform that. They average these numbers over several timeframes to smooth out the curves and give us reasonable approximations... but these are of course just approximations and can be skewed by large difficulty increases.

The global hash does not drop when difficulty increases, it is just a temporary lag where calculations are not reflecting reality quite so closely.

One of the contributing factors to this apparent loss of hash rates we are experiencing is the huge volume of transactions recently. The number of transactions is increasing faster than the actual hash rate. Additionally, as the global hash increases, it becomes more difficult to ship or manufacture enough hash to make a percentage increase.  You cannot have exponential hash increases forever, because if you could we would need the entire energy output of the sun by 2016 to increase the global hash by another 50%. It is slowing because it has to. 40% or 50% increases every two weeks simply cannot continue. There just is not enough energy available. I saw somewhere a chart where someone projected that if we kept increasing the hash rates by 50% every 2 weeks and had 28nm chips that by 2015 we would be using the entire energy output of the united states just for hashing.

^
That obviously

What? That's the most nonsensical thing I've heard all week. How is the hash rate affected by the transaction volume?

I only bother asking because everything else you said was spot on.

you must've not read the post a few up from yours that says people are shutting down asics then.  why on earth would people be unplugging asics that earn more than they consume in electricity/cooling costs?  they wouldn't, unless they don't like money.

I believe that monkee's comment was a comparing of the nonsense in the thread.  JakeGold says "most nonsensical thing ever...",  monkee then says "If you think that is nonsense well what about this..." and discusses what he thinks is even more nonsensical.

Yup it makes sense now.  I feel stupid (but I blame it on a lack of caffeine as part of a balanced breakfast) . :(

sorry, i considered quoting from both posts but i decided to be lazy :)

*edit* thanks for explaining BurtW.  sometimes i do need a translator, haha

Hash rates will not be dropping, I am still expecting a 15% increase next time, then another 10-15% increases each time for the rest of the year. Come end of Jan and feb we can expect 40-50% increases.

Moral of the story for small to medium players, buy and hold bitcoins. All ur mining rigs will become expensive coffee tables within a year.

Looks like normal service has resumed.

Hash rate might only drop temporary due to not profitable and miner off their ASIC.. but right now the price of BTC went up so I guess they turn back on immediately..