Bitcoin Forum

So the bitcoin block reward just dropped from 50btc to 25btc. I thought the hash rate would drop pretty quickly as many mining rigs became instantly unprofitable. But it looks like the hash rate is actually *increasing* and is sitting at around 28 thash / sec.

Is there a lag because a lot of people don't realize we've hit halving day? Or maybe a lot of miners don't pay for their electricity and it's all about sunk costs?

ASIC testing?
A bunch of people who used to mine but stopped noticed the publicity surrounding the halving decided to start up again?
The NSA is trying to mount at 51% attack?

Apart from the rules of bitcoin there is absolutely nothing predictable about it.

If a 51% attack were underway, it would not be contributing to the network hashrate.  The attacker would be extending the blockchain faster than the entire legitimate network in secret and attack by revealing a longer chain.

I suspect the current jump is simply variance.  Some miners will move to alt coins and some GPU miners with expensive electricity will power their rigs off for good.

Looks to me like simple time lag. People turned on some extra hashing power to finish off the last few 50 btc blocks and haven't got around to turning them off yet.

Many miners are probably waiting to see what happens to price. Last fall it also took a long time for miners to drop out, but many eventually did. If there's any chance that the price does go up then miners want to keep mining so as to be able to sell more at the higher price. When hope fades of a higher price, and miners decide they can no longer invest, then they will start shutting down. It takes time as miners don't all decide this at once.

Combine this with the expectation that soon they'll be shutting down GPUs due to ASICs and I think many will keep mining until that finally kicks them in the nuts.

Forum users are a small small minority of the hashing network it often seems, surprising as that may be.

I expect a delay of a week to a month for results to become truly apparent, as people either notice their bitcoin wallets slimming/their pool rewards not being what they once were or they get their next power bill and realize "hey this shit isn't paying itself off so well anymore".

I also expect that the "big boys" (> 100GHash/sec) had to be supremely efficient in the first place to even get that large, and are still making significant enough bucks after the halving that it is worth it to keep going (people with 4GHash/sec might think making just $1-$2 / day is annoying, but multiply that by 25+ and you're still a happy camper).

I was hoping to see some decrease instead of a huge increase in hashing power though, for a graceful decline in hashing power, rather than a big drop that might incite panic.

I havn't seen a huge number of people here in the forums that have closed up GPU mining just yet. Most people seem to be sticking it out, and seeing where it is going. For me, it still learns a (slight) profit, so why not? When the ASICs hit is the other side of the coin, and I'm assuming a lot of GPU miners will shut down them. Of course, the network won't notice them leaving, but that's besides the point.  ;)

I was expecting to come back today to see a shitstorm in difficulty and price! Good job guys, you average out to much better speculators than I am. I'm amazed how smoothly this went.

If you don't have cheap or free electricity, you might as well stop mining now, you can't compete with those people with cheap/free electricity.

It appears that many has free or cheep electricity. For me, its two 5870's on free electricity and one 7970 downclocked to draw 200W totaly with hole rig at home. At home i ~ 0, because i have to pay for electricity anyway. This is just slightly higher than without 24/7 PC running on 200W.

I'm still running my GPUs because they provide heat during the winter. Otherwise I would have to run electric heat. I figure I'm breaking even when considering the cost of electric heat vs. the cost of miners - value of coins mined. Come next spring I am shutting everything down and selling most of the equipment for a final cash out. Maybe some ASICs will actually exist in real people's hands by then.

That has been my observation also.  ;)

I can verify this from first hand knowledge. 

FPGAs have no reason to stop as free electricity also:)
but i wonder where the big LTC hashrate increase came from if not from BTC people?  ???

Whats up with the fluxuation of the network hashrate??? From 23TH-31TH & back,what a swing !!!!

Can we expect swings like this or even more at higher network hashrates??

I've seen it vary by 2TH or so,but WOW  :o

Big in LTC is minor in BTC, ie if 1% of BTC miners go to LTC it might be for example 10% (using Stetson–Harrison method) or more increase in LTC hashrate.

It will drop, most miners aren't the best business man, so the hashrate drop will be delayed till the bills come in.

Part of it is mining for principles and some is speculation that the price will shoot up. There are plenty of people mining at a loss right now but most will only realize it when they actually have to pay for their power which isn't for another month or so.

You hit the nail on the head.

Fact: Many Americans are bad at math (look at the popularity of the Lottery!)
Fact: Many Americans live paycheck-to-paycheck. This shows an inability to budget, live below one's means, insufficient attention to financial detail, etc.
Fact: Many Americans have to LEARN how to be frugal from websites, etc. It isn't something in their bones, or something they grew up with.
Fact: America has much cheaper electricity than, say, Europe -- so a good % of miners are going to be in America right now
Fact: Americans don't read books. Should I assume that they studiously read over their electric bill and understand it?
http://stephenslighthouse.com/2011/04/28/who-reads-books/
Fact: Young Americans are notorious for NOT being frugal. In fact, they are known for wasting their money on all manner of things ($18 for a cotton T-shirt with something specific printed on it?)

Yes, I could see young Americans running their Bitcoin miners just because they can afford it, and they "like" Bitcoin.  If they were losing money, many of them wouldn't even know it!

In fact, because men (let's face it -- the majority of miners) run on ego, I'd also expect them to brag online about how much they're making right now after the block halving. Who's going to verify it anyhow? They could be losing $5 a day and they could still fire up their browser and type in, "I am profitable all the way down to $5/Bitcoin!". Who would be the wiser?

Some drop already observed ~24TH now, interesting, how low will it drop?

The network hashrate is only an estimate based on the number of solved blocks in a given time period and also on Difficulty. Variance is expected, and a change in the short term average hashrate doesn't necessarily reflect a change in the number of hashes submitted to the network.


That is not true. The network hashrate has been quite predictable. I've correlated changes in difficulty with changes in MTGOXUS$ exchange rate and written models that predict the hashrate reasonably well most of the time. I posted results here:

http://organofcorti.blogspot.com/2012/11/101-price-drives-difficulty.html
http://organofcorti.blogspot.com/2012/11/102-forecasting-network-hashrate.html
http://organofcorti.blogspot.com/2012/11/103-canaries-coal-mines-and-black-swans.html
http://organofcorti.blogspot.com/2012/12/104-long-range-forecasts-of-network.html


These models will probably break down while the ASIC hashrate is being added. Block reward halving I'm not sure about, but my guess is that in the short term there won't be a significant effect on the network hashrate or difficulty.

The hashrate of the network is already beginning a decline as people shut off their rigs.

You can't tell yet. The network hashrate you see in various places is an estimate based on the number of blocks solved by the network in a given time period and on the network mining Difficulty. The shorter the time period over which this average occurs, the greater the possible error.

If you see huge swings in the network hashrate over short periods of time. it's more likely these are due to variance rather than a change in the number of hashes being submitted to the network.

it seems to me its more accurate to look at the larger pools listed hashrates, than the average based on block times listed on some. I see no decrease in Eclipse, or Bitminter. Maybe a slight decrease with deepbit. I'm not familiar with slush's average rate though. It will probably be at least a week until we see any decrease

Using block solve times and network difficulty for a week results in a fairly accurate estimation of the network hashrate over the week. Hourly averages will have a much larger variance from the actual (unknowable) network hashrate.

only 22.79Th/s now

only a 4% estimated increase right now for the next difficulty reset.

EDIT: I should also add that if the BTC/USD price went up by 4% as well, the price would go from $12.30 to $12.79. We haven't seen that yet.
So the BTC price isn't even keeping up with the post-halving diff increase -- nevermind making up for the halving.

I think the hashrate will be below 20 Thash within a week. Where will we be within 3 weeks?

I bet it will dip below 20 but be above that in 3 weeks. A slight price increase and a slight difficulty drop will go a long way imo.

That's why I've been keeping my fingers crossed!

Responses bolded.

Why would people still mine at profitability close to zero? makes no sense to me. My 5970's I bought a little over a year ago when Newegg got them back in stock for that promo code. I paid 500 for the first one and 400 for the second. now they are worth $220 at the most that's $460 I lost, or about 51%. So in reality $40 a month from approx 1400m/hash was breaking even. Add to that the approximately $30 a month in electricity at 0.08 a kwh. $70 a month was breaking even. Now most months I made between over $100 a g/hash net so I did profit. But at this difficulty with the block halving seems to me everyone but people with free power are losing money

when you figure in power costs, profits were cut substantially more than 50%. The only reasons I can see we haven't seen a huge decrease in hash rate is either people still think the price is going to double, or they think the used market for video cards has bottomed out and they want to make a few more bucks towards paying off their equipment. Well I can almost guarantee we haven't even begun to see the fire sale on gpu's yet

Many miners treat it the same as buying bitcoins. It makes sense to mine down to breakeven anyway. I'm just slightly above breakeven so I'm continuing just because it supports Bitcoin and because it's a somewhat decent gamble that the price will go up.

Below breakeven there is still some advantage for some people. They should just go buy bitcoins but if they don't want to give out ID info, or would have to pay a fee, or just don't have ready access to a market (like me in a distant location) - then I can see them still mining.

For me the halving was useful in that it prompted me to evaluate my efficiency and try to improve it. I should have done this a year ago as it turns out I've been burning more electricity than I had to.

I mined for a year 930MH/s @ 560W, net profit in the $40-$100 /month range.
Before tweaking I was at the low end of about $44/mo.
After tweaking things I found 750MH/s @ 375W netted me the same $44/mo.
Also the fans hardly run and the noise is lower on hot days here. Big improvement.

In the first case I would be losing money now due to higher electric costs but in the second I'm around $4 net above water. Not that I care about making $4/mo. but I'll just accumulate bitcoins until it actually drops below as I've already got everything setup and running.

Next I'm going to explore running P2Pool as that's more beneficial for the network.

cause people are hoping that the price will skyrocket some day, making up for the fact that they may even have lost some bucks here and there.

diff will go down, cause of block halving and ppl not making enough to break even.
price will go up, because of supply and demand (less miners)
diff will go back up again. cause miners fired up there rigs again since the price went up.
its the way of the world  :)

I already have hardware, my house heats up with electircity (which is cheap). Why not mine until summer when no heating needed anymore?

Mining until coinotron offers loyality points to redeem.

Well, by reducing the lifespan of your hardware then it is no longer a good deal.

still making a profit, not much but still a profit. Besides it's -10 C outside and I've got to heat the house somehow :p

I see that 50btc.com is down from 5500 to 4300... and btcguild also seems to have lost a TH as well.

Plus heating with electricity usually means a two way heat exchanger which have an "efficiency" of excess of 100%. Cases of classical electrical heating are rare.

Still running my GPUs for a little profit.. BUT, I appreciate that the temperature of my entertainment room in the basement is 22 Celcuis, instead of less than 15 without mining ;)

proportionately according to the pie chart at bitcoinwatch.com "unknown" has been getting pretty massive, so new hashing devices whatever they may be are still getting turned on, big farms that are solo mining? hopefully not a single user trying to interfere with network operation? banks? the majority of blocks are being relayed by a node somewhere in europe, 93.174.55.27 , seems to be relaying 10 or so blocks a day pretty consistently ...

I believe that's 50btc.com. They consistently don't get identified but if you go look at the block they claim solving on their web site they show up there. eg. 210682, 210650 are both on 50btc.com solved list and have that IP according to blockchain.info.

isnt p2pool a big part off the "unknown"?

ASIC devices are mining for a long time now. doubt its just me being pesimistic. so far it allways has been +2% increase , then like -1,2% and so on , it was allways going up and down, but last months it increases like mad without stopping.
asics are mining, thats why they are still not shipped anywhere.

imho ofc

People that need to support their SR Shopping?  Plus you don't have to deal with an exchange.

I'd rather build a bitcoin miner than go through MtGox

They are usually counted, but it seems not at the moment. In any case that is just 2 blocks per day on average, not a huge amount.