Bitcoin Forum

DIFFICULTY DISCUSSION
Updated 24/01/2013

Current Bitcoin Difficulty
4,847,647

Current Trend Arrows
↓ = Next Difficulty is estimated DOWN 1% to 10%
↓ ↓ = Next Difficulty is estimated DOWN 11% to 50%
↓ ↓ ↓ = Next Difficulty is estimated DOWN 50% or more
↑ = Next Difficulty is estimated UP 1% to 10%
↑ ↑ = Next Difficulty is estimated UP 11% to 50%
↑ ↑ ↑ = Next Difficulty is estimated UP 50% or more

Difficulty Graph of the last 2 months
http://bitcoin.sipa.be/speed-small-lin-10k.png

Sources
http://bitcoindifficulty.com/
http://bitcoin.sipa.be/
http://bitcoinwatch.com/

--------------------------------------------------------
O.P.
Looks like the difficulty might be going down? I know it's only a drop of 0.8% but could it be the start of something?

Does everyone think this is because everyone is starting to sell their GPUs, in anticipation of 1/2block.

I personally think (hope) that when the block halving happens difficulty will either go down(a lot), or Bitcoin value will double (*crosses fingers*).

either way, I reckon this is how it will go until our ASICs start rocking up.

I can tell you right now that btc price will not double overnight, the decrease in inflation rate is tiny compared to 10m+ coins outstanding. On the other hand, it'll add up over time, and have an effect on price months later. Difficulty decrease is likely, due to all the people quitting due to being unprofitable. Eventually an equilibrium will be reached again.

I'm curious what percentage of coins are sold vs held vs swapped on markets. There must be some stats on that and some super math people that could guesstimate how much the supply change will change the price. Say 78% is saved, 22% is sold or moved for product?
We already know what the daily supply will be and we have the volume of the markets...

Atm I would assume more people would be holding then selling. Unless the have to dump their btc for some financial reason

Just simple variance would account for a decrease of that estimate level with no decrease in physical hashing capacity.

But I don't think there are many people adding either GPUs or FPGAs so anyone trying to beat the rush by selling their GPUs now could cause a net decrease to the amount of hashing capacity that exists.


The difficulty adjusts right before block 210,000 so if many Thash/s of capacity drop out that first adjust period will be brutal.  Not just for severity (i.e., a shock to many miners who learn that "halving" means half.   As in if your system is bringing in 0.4 BTC before it now brings in 0.2 BTC) but in duration.  The more capacity that drops out means the longer that first adjustment period persists.

Here was my play-by-play prediction made about a month ago:
 - http://bitcointalk.org/index.php?topic=118952.msg1278827#msg1278827

Back when there were mostly solo miners you could get a general feel by seeing where the generated coin hadn't moved.  But with payouts from pools, there's really no data showing how much of that gets cashed out.   With pools that pay out generated coin directly (e.g., Eligius and P2Pool) you can see this again, and perhaps assume these miners are typical and then extrapolate the results for all miners. 

But just because the generated coin moves doesn't mean it is cashed out -- it could simply mean it got moved to a savings wallet, for example.

I was thinking the same thing.I looked around & still have no idea..................

MtGox volume is around 24,500 today....if that can be extrapolated to anything I don't know...............

Some are probably starting to disable GPU mining farm because of bicoin halving in 10 days AND Asic coming in less than a month.

Its just variance. I'm sure it'll be clearly obvious when GPU mining becomes unprofitable.

That made me "lol".

( ͡° ͜ʖ ͡°) Why does everyone think halving will cause the price to go up?

There's a difference between in going up and it doubling.

There are some who mine specifically to acquire coins to make purchases.   If the proceeds of mining bring in less than the electric bill, then to obtain those coins needed to continue making purchases there is increased demand at the markets.  Nobody knows how much of mining is for this purpose. 

And the supply of new coins will definitely be dropping by half.

So these factors normally put upward pressure on the exchange rate.

On the flipside, there have been a lot of speculators, miners and others holding onto coins waiting for the big payoff.  With there being less than 10 days to the halving and no steady rise (last weeks bump can mostly be attributed to the WordPress announcement) it could be that $12 is all we see.

From my casual observations though, I think there are a lot more people who would buy here but are playing chicken ... waiting until the last minute to commit, hoping the exchange rate drops again giving a lower entry point.  But even if there is a spike up as panic buying sets in, there are still plenty will be thrilled to cash out at $14, $15, if it gets there.  So it looks really unlikely halving will result in a doubling of the exchange rate.  But a 20% rise from $11.75 in the days leading up to the having?   Sure, it could happen.

And now it's a slight increase

Bitcoin Difficulty
3,368,767

Next difficulty (estimate):
3,371,490

Nearly virtually the same

There is no logical reason to expect the price to go up.

Many people post nonsense suggesting that the price will go up.

Widespread nonsense is usually sufficient to provoke a market reaction.

Nonsense is particularly effective when market participants are poorly informed.

This is how I think it'll go...

block reward halves

miner receives less money

Miner asks for more $$ per bitcoin mined.

Now I don't think the ask price for 1BTC will double right away, but I think that it will trend upwards.

A man can hope and dream can't he?

 :P

I wouldn't disable anything yet and wait for the actula halfing and/or ASICs to start pumping the difficulty up.

Unless you want to sell your GPUs and think that the price drop AFTER is more than the BTC you can mine BEFORE.

Sadly, it doesn't work that way.  The only thing that drives up price is increased demand vs. supply.  In the coming months there's going to be a lot more supply than demand as GPU miners liquidate their final BTC pools and ASIC miners start dumping everything they can to get some payback on their investments which will drive the BTC price down.  I wouldn't be surprised in the least if you see the BTC/$ price go half or lower where it is.

The only possibility that could drive up demand is if GPU miners that stop mining buy BTC in lieu of mining.  Honestly, I don't see that happening.  Let's be realistic, most miners are in it to make a quick buck.  They're not in it to help the BTC economy, because they they think Cryptocurrency is the wave of the future, or they want to hoard in case a BTC will be worth $1000.  They do it to turn a quick profit and make "free" money for hardly any effort on PC equipment they already had or picked up on the cheap.  I highly doubt that any GPU miner that isn't migrating to ASIC's will say, "Well, I'm not going to mine anymore, but I'll put all the money I would have spent on power for my GPU's into buying BTC every month."

People can kid themselves all they want, but 90% of the people are mining bitcoins because of greed, pure and simple.

FTFY

The bitcoin price is determined by the flow of people selling coins versus the flow of people buying coins. The amount of coins issued can effect this flow, but the number of coins minted certain has an effect. There's only 61K bitcoins between the current price and $14. With 3600 coins no longer being minted each day, this should tilt the flow to eat through that in a few weeks.

This is speculative though because we don't know how many of the minted coins actually go to markets now and how many are squirreled away and saved. And then after the halving, with miners save the same amount by percentage, or will they sell off the same amount nominally and thus have a lower saving percentage. For example,  starting with saving at 80% and selling at 20%, this goes to saving 60% selling 40%, which is still 10 bitcoins per block being sold, just less being saved relatively.

Which means those miners don't hold many unsold coins



Which they normally sell anyway, so that isn't an upward change in the quantity of coins headed to market for sale.


There would have to be something major to happen to see a drop anywhere near that.  GPU miners being unable to compete and powering down doesn't really impact the exchange rate either way, from what I can tell.

Lots of good answers here.
Maybe I'll make a highly speculative chart where the monthly volumes of all the markets are added up and I'll assume 20% of the daily BTC allotment gets thrown in there. Then take half out and see how the total volume changes.

It might cause it to go up, as people who were previously mining to support an SR habit will now need to source their coins from the market.

Yes, well a price rise would provide supporting evidence for a very strong form of Keynesian theory. One that even I don't believe is valid.

http://bitcoinmagazine.net/where-is-bitcoin-a-look-at-some-analytical-data/

Read this article today I must say I am impressed enough to share it here because it seems pretty relevant to the topic.

Thoughts?

The main reason why I wouldn't be surprised to see it drop is that I would think you're going to see a lot supply than demand and GPU miners sell off their remaining BTC and ASIC miners dump as much as they get to get payback for their ASIC's. 

But of course, we all know that trying to predict what the BTC/$ will do is about the same as trying to predict which direction a herd of cats would go.

My thoughts are that Vitalik is a very poor writer, which may not be his fault, he is russian or something? But the endless self-referencing is a bit annoying.

Sometimes articles prompt me to skip to the end based on their writing style, and this was one of them:


That's the gist of what I got from that article as its conclusion. "We dunno, but it seems like it should be ok".

I'm not sure what you are referring to on that.  Do you mean links to other articles that he authored?

I do indeed. I don't mind linking to other articles as an information resource about other topics (though linking 10 articles in one paper can be a little self-aggrandizing), however the stylistic choice is somewhat misleading. Perhaps an overabundance of Wikipedia-mindedness, but I expect that a hyperlink of a term (such as Bitcoins Savings and Trust) will lead to a very impartial explanation of specifically what that thing is/was, not another similarly styled wandering treatise on some event. He also uses his own writings as a 'reference' to back up the argument he is making in the current article.  It is poor form in my opinion to build an argument, then use that argument as evidence, without at least mentioning that he is citing himself.

I prefer the magazine style choice of: "You can read more about XYZ in other articles I have written about XYZ <hyperlink>here</hyperlink>" (or similar to that) if choosing to link to oneself in ones own article while discussing a topic, so that I clearly know what type of information they are presenting. Also I think it helps the author from over-linking, which can make the article itself hard to follow, as you split off into tangent after tangent.

Hope that makes my position clearer.

Ok, but what do you think about the data?

I think that halving will cause the price to go up yes. Not like a price-shock overnight, but on the longer term; absolutely. And I would argue it's quite basic economics. By halving the bounty, inflation will drop in half. In 2011 the inflation was roughly 60%. In 2012 the inflation was roughly 25%. In 2013 the inflation will be about 15%.
The rough numbers are like this:

Year   BTC@end   multiplier year-over-year
2011   5000000   
2012   8000000   1.6
2013   10000000   1.25
2014   11500000   1.15

Someone who has the exact numbers please correct me.

The supply is decreasing on a technical level because there are fewer BTC produced.

When it comes to demand, I have a hard time seeing what would put the brakes on there. BTC remains stable, and it's possible to move in and out of the market with increasing ease. The currency becomes more and more popular, both among merchants and customers. As the mining revenues drop, mining will become less and less the center part of the bitcoin economy; they will in a few weeks drop to just 25% of the economy. So the miners will gradually lose their opportunity to sell large quantities of BTC, since their revenue goes down, at least in terms of BTC.

I might have overlooked some flows of money in and out, but I think the concept of looking at how the money flows is quite basic.

I am still looking for reasons that Bitcoin would drop. I think one such thing would be if a government banned the currency, I think that would send the price spiraling down. Apart from that, I cannot see any clear indicator, but please enlighten me if you see other hurdles for Bitcoin.

I agree witht he above statement ... but to go half value? I don't think we have enough selling pressure to achieve that.

Well, difficulty is at 3438909 right now. Anyone want to make any predictions on what it will be come the next retarget at 211680?

Way down to the low 2Ms?
Something like 2.8M?
Down a bit to 3.2M?
Staying the same?
Small increase?
Big increase as ASICs hit?

I'd guess it will end up somewhere around 3.2M as some GPU hardware is turned off, but there won't be a huge change as some miners will be slow to wind down and FPGAs probably make up a significant percentage of the network at this point.

Yeah I don't think ANYONE thought it'd go in THAT direction

o.0

Everything speculation of course, but logically I would suspect the price to increase. All I know is I dont believe the time to sell is now. Im hanging on to them all for now. Theres too many unknown factors out there right now. But hell. This market is so damn volatile no one can truly know.

You're assuming that there will be steady or increased demand for BTC. 

Is Bitcoin Charts Network hash rate correct? It currently shows Network total   20.997 Thash/s, a couple days ago it was like a little over 29 Thash/s. Should the difficulty be going down? 

It's just an estimate based on the time to find a block, so it doesn't take into account the effects of variance. You'd expect it to bounce around quite a bit, and it does.
Still, I wouldn't be surprised if hash rate went up to try and find 210000 and it is now dropping off. Many GPU rigs are now unprofitable.

The drop we are seeing is starting to look more significant than the variance over the last 3 months: http://bitcoin.sipa.be/speed-lin-10k.png We need 2-3 more days of data to be sure.

Hey, did you ever get ahold of that special hashing algorithm you were going on about all those months ago?

?

I wouldn't be surprised if we see total network hashrate dip below 20 TH in the next month or two (assuming ASIC's aren't in the mix).  I for one have pulled the plug on my miners.  Even with the hardware long payed off and a decent power rate of .10 / Kwh, I'm paying $9 in power costs to make $10 in BTC.  Just not worth the effort/aggravation.  I think in the next week or two, more people will see that and only folks will really low power rates or FPGA's are going to stick it out.

Checkout the green (3 day estimate) drop!  Biggest drop all year!

http://bitcoin.sipa.be/speed-small-lin.png

The difficulty estimate is still...
3,438,909
Next difficulty (estimate): 3,461,345

...higher?

It's an average over the whole 2016 blocks, and it was higher over the first part (especially right before the halving). Give it a few more days and it will probably end up slightly lower.

3,438,909
Next difficulty (estimate): 3,435,354

Difficulty is going to lower definatly by the looks of the graph and the estimate.

I reckon people are seeing the spreadsheets and earnings estimates, and have decided to call it quits.

Current difficulty 3,438,909

Next difficulty (estimate): 3,372,281


Starting to go down faster!

2886714.88092081 !

http://blockexplorer.com/q/estimate

That link returns: 3405980.70883142 for me.

Now it is:
3423585.80041288

Maybe we need a Block difficulty watch thread...

Certainly isn't tanking.

Done. Thread renamed. I'll adjust my first post to include a few more details soon

I don't know about that link. I would use Bitcoin Watch (http://bitcoinwatch.com/) instead.  Right now the future difficulty is still trending downward: 3,280,731.

It's not tanking, but from what happened last year, this isn't surprising. Miners are slow to throw in the towel.

Ain't that the truth

Wonder how long this slide will continue, before the new FPGAs and ASICs start to outweigh the leaving GPU miners.

Bookmarked! Thank you and nice to see the difficulty going down some.

Whoah wtf? 3.9 million BTC sent in last 24 hours?!?!?!?! That's pretty healthy.

I'd be betting 90-95% of that is Bitcoin Laundry and Blockchain.info Laundry

You mean SatoshiDice? :)

new estimate 3,160,030.... and we're up to 13.50ish

only 4 hours later and the new estimate is

  3,151,234 @ 11.45PM UTC 15/12

http://www.bitcoinx.com/ (http://www.bitcoinx.com/) estimates 2,940,164

and http://blockchained.com/ (http://blockchained.com/) estimates 2,949,803

Sweet. Just a quick idea you should add in the title of the thread a little +/^ or -/v or something to indicate the movement of difficulty. This could be like the MtGox wall thread and live forever...

Thread title changed

got any other good ideas?  :)

Looks like about 3 TH/s has been removed from mining bitcoin.  :o

You mean 0.3GH ?

3TH/s is 3000GH/s.

I was having a retard moment.  My Avatar Picture is evidence of it :P


I've been doing some projections and this is what I think the network will do...


JANUARY 1st 2012
Network will bounce between 17TH - 22TH

FEBRUARY 1st 2012
Network will bounce between 14TH - 18TH

MARCH 1st 2012
Network will bounce between 12TH - 17TH

APRIL 1ST 2012
Network will stabilise between 11TH - 16TH

The data is based on these variables staying...
ASICs still being on the horizon. 
While people are waiting for ASICs they won't create much new power.
Power costs force GPUs out.
BITCOIN value continues to be stable or rise slowly ($0.50 -$1.00 USD per month)


I for one can't wait until I'll be earning the same amount on March 1st as I was on Nov 1st. 

What yas think?

You have network hashrate going way too low, IMO. Even with a 2MH/J total system efficiency, existing GPU hardware will be barely profitable at the current (3.37M) difficulty at US$0.15/kWh. At 16TH/s mining would be quite profitable even for inefficient GPU miners.

I honestly don't see the network dropping much past 18-20TH/s or so. Too much of the network is made up of FPGAs and GPU users with low power costs and/or who would be heating their homes anyway. If the price of BTCrises, network hashrate will probably follow as people turn rigs back on.

Of course, it's probably all moot as someone should ship an ASIC by the end of January.

Those are some bold predictions...of the past  ;D  More retard moment?  I can prescribe some Aricept or Namenda...

https://bitcointalk.org/index.php?topic=128032.0;topicseen

According to this poll approx 65% should keep GPUs etc on so about 2/3rds of peak hash seems reasonable.
Not too far from that now.

wow network hashrate only 21.61 Terahashs/s, then asics then ... BOOOOOOOOOOOOM ;D ;D

Bitcoin Difficulty

3,370,182
Next difficulty (estimate): 2,988,542

Under 3 million!

Hash rate is shooting up.

It'll fall back down.  After recovering from the DDoS, BTC Guild started shooting out blocks roughly twice what it's hash rate should produce (which means a ~3.3 TH/s pool is pushing out over 5 TH/s worth of blocks in the last 24 hours), which means a single pool is inflating the 24 hour estimate by over 2 TH/s alone.

Ack, wasn't aware that could happen. What is the reasoning behind that burst in blocks?

Just an extremely good run of luck.  It's why looking at daily network hash rate charts you see such massive swings.  It's not that tons of people are turning on/off, it's more about overall network luck than anything.

2,990,000 est... that's a good move.

Yep and a few days left before the change, so it could go even lower. Nice.

New Difficulty update
2,979,637

Next difficulty (estimate):  2,978,049

That's looking pretty stable there.

Once everyone has stopped eating and parting hopefully more GPUs will come offline

Dude u mean online ...stop smoking crack...lol

Nope, I meant offline.  I'm guessing that after everyone has recovered from running amuck at Xmas & New Year, they'll realise how much money they've spent.  Then they'll take apart their GPUs and sell em on ebay.

Now if you'll excuse me, my crack pipe is getting cold...

o.0

Bitcoin Difficulty
2,979,637

Next difficulty (estimate): 3,229,121

Looks like the equipment is coming back on. Or ASICS?

No, these are now "diff hoppers", people on the brink of profitability who wait for the diff to drop, then mine until the diff rises, then stop. They did so numerous times on the LTC chain which changes more frequently, but they're doing it now to the BTC chain.

Up up up and down
Turn turn turnaround
Round round roundabout
And over again
Gun gun son of a gun
You are the only one
Makes any difference what I say

http://youtu.be/3OlxKVa7BHA

Love that song

What happened to the concept that btc price would increase massively after the reward halving? I see a price that’s pretty close to what it was at a 50btc reward.

Yeah I know. Maybe once all these holidays are over people will change. By the end of January I recon the price will go up 25% or the difficulty will go down 25%

I think everyone is holding their breath about ASICs, that will change the game. I think people are trying to squeeze the last bit of profit out of their rigs before the ASICs start arriving close to the end of Jan/begginning of Feb. My GPU rigs can stay afloat until 4.5mil difficulty, and my FPGAs at about 45mil diff, where I'll be getting off the mining train. I'll stick around in BTC, but no more mining for me...

Gonna update the title to reflect the new difficulty?

3,249,549

Does anyone know the normal length you have to mine in order to acquire 1 BTC. Also, how many computer's does a "pool" usually consist of? Like 20+, or below 5?

Pools can consist of thousands of different users, and those users may have multiple miners connected.

How long it takes to mine 1 BTC depends on your setup. My setup puts out about 3200 Mhash/s at the current difficulty I mine one bitcoin approximately every 2 days.

To check how much you would make at different hashrates, go here: http://www.alloscomp.com/bitcoin/calculator

Thanks for the info. How much did your hash-rate cost, or basically your set-up to acquire 1BTC per day. I'll be investing in 7 months (I know a really long time away) but I kind of want to get an idea of what mining is right now.

What do you think the top notch/most expensive set-ups could make you per day on one computer?

I'll have 5-6 computers and free power.

You can see what I have in my signature. The whole game will be different in 7-8 months. You won't be able to mine with just a computer, you'll need specialized devices called FPGA miners or ASIC miners. there are plenty of threads in the Hardware subsections.

Bitcoin Difficulty

3,249,550
Next difficulty (estimate): 2,972,949

Difficulty Change in 23 Blocks

SCOURCE : http://bitcoinwatch.com/ (http://bitcoinwatch.com/)

Bitcoin Difficulty

2,968,775

Next difficulty (estimate):
2,977,072

oh crap diff too high, turn it off! turn it OFF!
ok diff dropped, turn it back on...

Bitcoin Difficulty
3,275,465

Next difficulty (estimate):
3,285,916

We've nearly caught back up to before block halving difficulty

o.0

CURRENT DIFFICULTY
3,651,012

Next difficulty (estimate):
3,964,597

Wow, that's nearly a 30% increase in the course of a month

Thats nothing just wait and see what the next 2-3 months look like...

i'll be in still for this next difficulty jump

doubtful for the one after that

up she goes!

Bitcoin Difficulty
4,367,876

Fuck me, there goes a third of my mining income!

0.o

Haha, yeah what's the point of BTC rate going up 2x if difficulty goes up 2x also.

The question is will the next difficulty increase come in under a week if all the Avalon units kick on and ASICMiner gets fully set up.  Either a bad string of luck or some GPU Miners already quitting since 1 day into this period and the estimate is a little down from current difficulty.

As usual, BTC Guild can take a bit of the credit/blame for the inaccurate estimate.  We were running close to -40% luck the last 24 hours, which as far as the network estimates are concerned means around 4 TH/s disappeared after the diff change.  Don't be surprised if it swings the opposite way and you see an 8 TH/s swing in the other direction after another day.

Bitcoin Difficulty
4,847,647

Next difficulty (estimate):
4,969,078

Bloody hell, This jump was over 10% !

Look again  ;D

4,847,647 (+11%) current
6,172,714 (+27.3%) next

in 4d 19h 45m (885 blocks)

6,695,826

Ouch!

Bitcoincharts shows:

Difficulty   6695826
Estimated   7446578 in 821 blks
 
Perhaps due to BFL ASICs finally starting up as well now...

Like what, one unit? ;) Most should still be ASICMINER/Avalon. And maybe old GPU rigs getting started again.

7,673,000 (+14.6%) current

So what are we to expect? how much slower block generation on average?

Difficulty is adjusted so one block is generated every 10 minutes. (If network hashing power doesn't change.)

Right, but isn't increase by 1 mil too big? Though I am new to all this, what do I know :)

Well if the network hashing power goes up by 14%, then so does the difficulty...

New difficulty

7,673,000

As a new hobbyist miner guy I can't believe how fast difficulty has gone up...

Started mining mostly LTC in late March, difficulty was 78ish... and it's looking it'll go to 285 in two days.   From 78 -=> 280 in about two weeks.. ! Pretty insane. Glad I'm just really enjoying it as a hobby and didn't drop too much into hardware..

Look back at the BTC graphs in 03/2011.  Pretty much the same thing as CPU miners switch to GPU Mining and a new boatload of miners piled on.