Bitcoin Forum

Since reward halving might come before ASICs, how much do you think difficulty would drop in November if thats the case?

Meaning how many of the GPU miners will have to stop if their revenue halves with current price/difficulty?

EDIT: Ok, wrong forum, I reposted this in Hardware!

Gazifilion GPU miners.

It would knock out everyone GPU mining except for people who don't pay for electricity (colledge students, people who live with their parents, and people who leave em plugged in at work)

This would probably knock out 95% of GPU miners that mine >2GHs.

Anybody paying less than $0.10 per KW/h would either still be making decent coin or just breaking even.

Not really.
I mine with a bunch of 7970s, ~3.2Gh/s @ 790W at the wall on my main rig.

I pay £0.0959 (~$0.155) for power, and I'd be profitable at much lower reward-per-share than we have currently.

You aren't the average miner though. The average miner will earn substantially less than you and even at the same energy costs may find it unprofitable.

OTOH, I think most northern hemisphere GPU miners will continue for a while yet - GPUs heat up a room quite nicely.

Why would you want to mine on 2xHD7970 at 1200Mhash, if you can sell them for $700 and buy an ASIC that will mine at 30Ghash  ??? 

I think I will continue mining with my watercooled 5870s as long as they manage to break even - after all, I can't heat the bathwater with ASICs.

Yeah, I'm probably not, but what he was saying is that the reward halving would knock out most >2Gh/s miners, when those of us with that sort of GPU hashrate typically have fairly efficient setups. I think we're more likely to see a lot of the smaller (<1Gh/s) miners give up.

Also, I live in scotland and my heating bill over the winter is awful, so with any luck the ASICs will get delayed a couple of months so I can have effectively free heating all winter  :D

You could with enough of them  ;)
Multi-terahash under-floor heating system anyone?  :D

Yeah, take into account the heating element.  I had my main rig in the lounge room at the tail end of winter, the MRS was hogging it most of the time :P

My I recon your average pro miner has maybe 2-5GH/s worth of radeon 5850-5970s, the reward half would probably kick em right out.

Regarding heating, if you sell the GPU and buy ASIC, you will be able to afford even more electricity and heat your house even better :)

Difficulty is not going to drop at all.  We've seen in the past that most miners keep mining even after it's no longer profitable in the hopes it's just a glitch.  Add in the fact that on ASIC minirig coming online will more than make up for 300 5GH/s GPU miners going offline, we're not going to see anything drop.

This depends on wether or not (or to what extend) the reward drop will affect exchange rate.

I think most are underestimating the effect of x% of 3600 BTC (x being the percentage of mining reward having to be sold for paying power bills in fiat) suddenly missing from the supply-side on exchanges.

Related speculation:

http://betsofbitco.in/item?id=673

http://betsofbitco.in/search?q=difficulty

At present prices, pre reward-halving, I make ~$4.18/day. After reward halving, I will make ~$1.37/day with just my GPUs. The price would have to dip under $10/coin after the reward halving and stay there in order for me to lose money. As others have stated, GPU mining does keep a room warm. It also helps provide a steady drone/hum so I can sleep at night.

I also have plenty of FPGA and I'll keep those going for quite some time, too.

Like molecular, I have suspicions that the block-reward halving seems to have not been priced in for most of the people purchasing ASIC at this point. They would have been better served purchasing coin, letting ASIC mature, letting the block reward halving kick in, then selling coin to buy stable and reduced-price ASIC miners.

I think the first wave ASIC miners are going to get screwed as the price spikes cause of exponential difficulty increases, then falls as the older miners start collecting on our mine-and-hold strategy by selling our piled coin. I'm pretty sure the global fiat economic atmosphere post-election is going to be unfavorable to everyone, and the ASIC manufacturers are going to be lowering prices no matter what to compensate for this. With a low price/difficulty ratio, it might be like the August 2011 GPU scramble, where Europeans and Californians sold their GPUs in masses. Only this time it'll be all the ASIC miners trying to compete for lower prices with the companies that originally sold them the devices.

Just a guess, I obviously could be wrong, but this is my assessment from the GPU craziness of last spring/summer.

It's currently at -3.4% since the last change ...

Is it 100% that higher difficulty = higher $ price? I agree that higher price = higher difficulty!

I'm thinking about voltmodding my FPGAs to squeze some extra juice out of them before ASIC hits. Just reasoning if that would make sense: now, after reward halvation or not at all... basically I'm lazy/scared to mess the chips up...

About fiat, isn't that why we all have bitcoins? So isn't that a null game: deflation in fiat + scared of fiat implosion = prices are stable for everything concerning bitcoin? Just like gold?

100%? No. But I do think there is pretty strong evidence that new miners hoard/speculate on coin, thus decreasing supply. When I first got my set of GPUs the price of BTC was $14 and rising. It hit $30, or an ROI of around/under one month on my GPUs, but I still wasn't selling. I wasn't paying attention to my ROI and I should have known better.

I have to imagine that if demand stays the same and supply from miners is decreasing because new miners typically hoard, the price will, in fact, increase. I've said this before, but a drug addict that wants $100 worth of Silk Road goods doesn't care if he gets 10 BTC at $10 or 1 BTC for $100. (or 100 BTC for $1, for that matter). It seems natural that the price will increase because new ASIC miners won't sell. Shorter supply means higher prices. At least, for a while.

Will there be such a thing as new ASIC miners?  Will people who have never mined before want to invest money in ASIC?  When GPU mining was profitable, anyone could do it, people got interested and upgraded.  I think the advent of ASIC makes it a niche sport now.

Why??

Its gonne be even easier dude!!!!!!!!!!!!! No high end PC or vid cards to buy & maintain.No issues with vid drivers,OC'in,drawing too much current from your PSU OR your outlet.

ANYONE can buy one & hook up a USB cable,DL the miner software,join a pool & boom your mining on an old P.O.S. PC..............

Hell,my mom is 72 years old & she could do it  8)

This is true, however, I anticipate that after ASIC's are about and hashing publicly, the difficulty will rise so high that owning a Jalepeno will be like using a X750 or X770 ATi card.

Also keep in mind that it is relatively easy to get a graphics card in much of the world, but acquire a system ASIC is very different, especially if you do not live in USA.

You are forced to buy from abroad, adding paperwork, waiting times and tariffs (very big in some cases).

In addition, BFL, BTCFPGA and Avalon (I think also is based in USA but produced in China) were U.S. companies and these devices are classified under the "U.S. Bureau of Industry and Security export control", so they can simply refuse sale by your country of origin.

Although jalapenos and single SC look cheap, with the rise of difficulty, maybe not be more efficient than the same investment in GPU today.

Edit:
13 years old gamers can't mining any more.

China also has this type of control, although it may ASIC not be listed.

http://www.kslaw.com/Library/publication/China's%20Export%20Controls%20and%20Encryption%20Regulations.pdf

I agree  :(

But,like with GPU's now,you still need a specific amount of hashing power to make anything (around 3gh/s seems to be ok for me now).

So,maybe 30-90 gh/s in the future to make what I'm making now.We won't know until feb-march when the ASIC's saturate the mining market.

+ the price of BTC should go up a bit after the halving.How much?? Only time will tell............gonna be interesting to watch at any rate  ;)

Ahh,didn't see that point of view........I'm an American & I only see things from my side the world  :D

Dare I state the obvious ....

Is this in any way unexpected?

The price performance of the devices is of course relative - but the price point certainly suggests buying a Jalepeno and mining with it in the near future will be no better than buying a cheap GFX card now and mining with it.

The $1k ASIC devices may even (in the not too distant future) be no better than mining with a high end (but certainly cheaper) GFX card is now.

Difficulty will adjust and we'll all be back where we started ... except our pockets will have emptied themselves into the ASIC manufacturer's pockets. Certainly not a criticism of the manufacturers - just a simple fact that's how it works.

Shovel-making sure is a sweet business to be in during a gold rush.

Agreed.  Its interesting to see the people who are buying 2 or 3 of these devices think they are gonna make millions.. they aren't thinking about the difficulty increase

Most americans do :P

On a long enough timeline you're correct (of course) - I believe they'll be slightly better return than the same money spent on a video card initially. I project the difficulty will stabalize at a point where an SC single makes between 3 and 4 times what an fpga single makes right now.

Something to recall is the ASICs will have a much lower power consumption and that (at least in the USA) should drop operating costs for most miners. I don't know about everywhere else, but here we have a percentage based rate... in my area it's 11 cent up to 625kwh then 13 cent @ 110% and 15 cent @ 120% then jumps to 35 cent over 130%. Which frankly is a moot point for me since I've offset most of my electric with solar panels. But that will affect some folks at least.

Of course in another year, instead of 5x or 10x difficulty we'll see 20x or 30x and then it's a race 40x as peoples initial re-investments get paid off and everyone starts scaling up.

At 40x difficulty - an FPGA single would take a loss for mining if paying 13 cent per kwh. And SC singles will still be hitting ROI in under a year.

Well, being late to the scene, I'm considering buying a little sc unit from butterfly labs but I have my concerns about profitability (in part because no deliveries have been made yet and no effect can be seen yet).
However if it is still profiltable after a few months I would buy an asic unit and start mining as a first time miner.

(I have a laptop with internal video card for the moment, i even tried cpu mining a few months ago hehe..)

Think on this a bit...BFL has already said that they have enough orders to bring the network up about 10x from it's current 20 TH to about 200 TH, which will cut your profitability to 1/10th of where it is today.  Then you factor in the drop in the reward from 50 to 25 BTC in the near future, and that halves your profitability even further.  So, that's a cut of about 1/20th of what you are making today.

Of course this assumes the $/BTC price stays steady, and that 200 TH numbers is really where things end up.  If that is the case, if you are mining with the 4.5 GH that a Jalepeno is supposed to do, you'd make just under 1.5 BTC/day.  After the increase in network size and halving of reward, that 4.5 GH is going to produce about .075 BTC/day, or about 90 cents in USD at today's rate of $12/BTC.

Granted, it's probably still profitable using only about .1 kilowatts/day (or 3/month), but you really going to need to go big or go home when things really ramp up.  We're not even considering orders from other ASIC vendors getting into the hashing pool.  Even if every piece of equipment that's mining today stops and only the new BFL orders are mining, that's about 180 TH which knocks your profitability down from today's rates to about 1/18th...

So I'll be making 1/20th of what I would be making today, on a per GH/s basis. But I've increased my GH/s 60x, so I'm still coming out 3x ahead.  ;)

Absolutely...my point was that anyone thinking they'll be able to just continue on with what they're doing now is going to be sorely mistaken.  While there might be a few months of flux in the difficulty rate and hashing numbers that might still allow someone to survive, you're going to need to jump on the ASIC bandwagon.  Figure you're going to need 20x the hashing power of what you're doing now to match what you're making now.  That's likely easy to do if you go ASIC, but someone mining a couple of GH now isn't going to be able just swap in a Jalepeno or two thinking they tripled their hash rate and they'll be alright.  If you got 2 GH now, you'll need at least 40 GH to match your current profits.

Never too late to start mining.Just get what you can afford & reinvest the mined BTC in another device & so on.Eventually you will get to a more profitable point.

Thats my thinking for my situation,but,don't put in what you can't afford to tie up for at least 6 months to a year.

I like your thinking Meatball,thanks for sharing  8)

Never too late to start mining.Just get what you can afford & reinvest the mined BTC in another device & so on.Eventually you will get to a more profitable point.

Thats my thinking for my situation,but,don't put in what you can't afford to tie up for at least 6 months to a year.

I like your thinking Meatball,thanks for sharing  8)
[/quote]

Exactly personally I re-invest about 20% of what I mine into new hardware. . . seems to keep my income increasing so far.

Definitely agree with most of you. Have to reinvest a significant amount to keep reaping benefits. I have to do it piece by piece, but I continue to add until I notice I have surpassed my goals. What I like about the reward halving is the fact that I know I have 4 years till I have to worry about it again. Therefore I will probably try and accumulate as much hardware as I can over the next 12 to 18 months(depending upon the performance increase on next generation products). After that I hope I can slow down on purchasing and use what I've accumulated for the remainder of the reward period(pretty much between 2.5 and 3 years).

Did a quick off the cuff calculation.  We're currently on block # 202,616 (and likely beyond that by the time you read this) and we move to the 25 BTC Reward at block # 210,000 which is a mere 7,384 blocks from now.  Bitcoin charts currently shows we're doing about 7 blocks/hour.  So, at that rate we're just a smidge under 44 days till the switch to 25 BTC.  ASICs are supposed to start shipping in 20-30 days, so at most, the first ASIC's hashing are going to see the 50 BTC reward for maybe 2-3 weeks, and considering that once they start cranking the blocks are going to come fast and furious for a round or two while the difficulty ramps up, so my guess is those 2-3 weeks get compressed to under a week.

So, just for grins, I'm throwing a guesstimate out there that we'll see the switch to 25 BTC right around November 15th.

Nope - it's 6 blocks an hour.

It doesn't matter what it is at the moment - it matters what it will average over the next 7,000 blocks.
As soon as it hits 2016 blocks, it resets it back to make 10 minutes a block.
With a 10% gain over 2 weeks that will only drop 1.4 days (i.e. 302 hrs instead of 336 hours)
But it will reset after that.
It's quite unlikely to keep going at a constant 10% increase every day for the next 7,000 blocks.

... and to show that point, over the last 7 days (since the last change) it has actually averaged -1.33%

A little OT but:

Until ASICs come online, yes - 6 blocks an hour. After they come online, and before they reach a steady state, the Network hashrate could rise significantly within a single difficulty period. This would mean blocks being solved much faster than usual, a much shorter difficulty period than usual, and probably an increased number of orphaned blocks.

Well, unless someone else has some inside information that I don't know about :)

The first public ASICs will be the week after 22 November from bASIC - Tom's specifically stated in his thread that week and I'd be surprised if he wasn't quite accurate in his statement (give or take a week)

As for BFL - still no sign of a reliable estimate update ... but that gives them only 6 weeks to equal or better Tom.
Yes I said 6 weeks, not 4-6 weeks :)

I would be surprised to see any REAL amount of asic devices hit the streets before Thanksgiving.

All I know is: when ASICs start being delivered, there will be a lot of thanks-giving.

I am just wondering if it will be as big a boon as people hope/claim being it is so close to halving. There will still only be 7200/2 coins available no matter what ASIC you have. Maybe if there were deliveries in October but that seems unlikely.

I was under the impression that difficulty can only increase by 4x per adjustment... so if we're scaling up to 10x or 20x difficulty, it could take awhile before the difficulty scales up to where it should be.

With the halving and one 4x difficulty adjustment from current levels 120GH/s should only produce around 5BTC a day if my math is right. Not really burning it up.

If the network hashrate changes significantly, the difficulty period will be very short. Scaling up won't take very long.

Not long in the sense of 'time taken' but if it's limited to 4x per adjustment that means 3 adjustments to get to between 40x and 50x.

During that time, the folks with asics will take the majority of the coin generated... that's what 6048 blocks? that's 302400 btc going mostly to the people with the new asics. or aprox. 3.6 million USD. If that happens in a day, we're going to see massive de-valuing at all exchanges and a steady climb back to pre-asic rates.

Suppose it is aprox 200 th/s after asics. running your 60gh/s single will bring in aproximate 1000 USD - which meets ROI on upgrades =P

Other thread tastes BFL now think shipping cold slip by a couple of weeks, meaning mid November. Add to that at least a week for international orders and its unlikely any non US residents will be mining on BFL ASIC gear until December at best!

I haven't been following the other offers so couldn't comment on them!

i think a lot of gpuminers will switch to litecoin mining, some already have

Problem is lets say you move 23GH/s worth of mining power from BTC to Litecoin and your aggressive in your settings you may make almost 4000 Litecoins a day at present difficulty give or take a few hundred.

Those that have been mining Litecoin's are going to try to cash in on the recent popularity. So you have to hope the speculators that opt to use BTC to purchase their speculative share of Litecoins are more then those moving their mining infrastructure over to  Litecoin.


I can see speculative pressure moving Litecoin value up to the mid 0.0100 when valued against BTC. However if more "miners" join the network the returns will diminish quickly as there is no use for Litecoins apart from mine and dump at this time.

To sum up I'll point out the obvious. Unless more speculators buy up the pre-existing Litecoin available on the market. The popularity of moving mining infrastructure to Litecoin will in turn put pressure on the present value of Litecoin as early adopters choose to cash in their earnings and moving them to BTC and other potential fiat currencies such as USD.

Well… not quite. It's only linear for sane changes— though it does have quartic convergence even in crazy cases.

I beg to differ. If BFL delivers, that's exactly what will happen.

There's a thread where people publish the orders they placed with bfl. These amount to 37TH. This would account for 61% of total network hashing power already if added to the network right now. Of course that list is incomplete, so you can probably at least triple that to 111 TH (82%).

Why are you guys still wildly guessing? Josh already said indirectly that they'd sold ~200 TH as of a few days ago. And if anything, I'd expect him to understate that figure, as a higher number discourages additional sales. I believe that by the end of the year, the network will be composed 90% of ASICs.

Careful how you quote molecular. That's not me you quoted, that's firefop. Mind editing your post?

yep, I should be more careful. Sorry 'bout that. Edited.

Maybe we're not as up-to-date as you are. Thanks for the info.

90%?  Ah, if they actually start shipping in volume, I bet it's closer to 100%.  The difficulty is going to skyrocket and within a month or two anyone doing anything other than ASIC will probably find out they're spending a dollar to make a nickel.

I think the X6500 is the most power efficient miner on the market today, at 16.4W for 400 MH/s. But when the block reward is 25 BTC and the network hashrate is 500 TH, which may happen by the end of this year, you'll need to be paying less than $0.09/kwh to break even. I'm assuming $12/BTC. And even if your power is free, you'll be making a piddly $1/month. Is it worth the time to keep it running? And what's the value of a mining device that makes you $1/month? $10? Maybe $20? The device may be worth more for scrap or recycling at that point. I believe most rational GPU and FPGA owners will quit by then.

A monetary system that's affected by profit.
Wow that's gotta be so fucking unexpected ...

What most people here are forgetting is that the price of Bitcoins will probably go up due to supply and demand.  Less supply with same demand = Higher price. Now a lot of people will sell some of their coins to pay for ASIC's which will probably keep the price down for a while, but as soon as people start hording again the price should go up. I could be wrong, but that's my opinion.

Currently, it seems network hashrate stabilizes around 2 TH per $1 market price, for example currently 22 TH and $12/BTC. When the market was around $5 for a while, hashrate stabilized around 12 TH. After the new hardware is out in mass and the block reward is halved, I expect the new normal to be 40 TH per $1 market price. At $12, that's ~500 TH. If the market goes up to $20, expect hashing to increase to ~800 TH within a couple months. Hashing increases do seem to lag price increases though, so while the market is going up, mining is more profitable (if you don't compare to the profit of simply buying and holding the currency).

We should do a poll to see what percentage actually buy BTC and hold for value.  Mining makes more sense initially to me since you have actual hardware.

I saw mining with GPUs as a safer bet than buying and holding bitcoins, even if it often didn't make much more profit than holding bitcoins. Because even during the crash from $32 to $2, difficulty dropped so I was able to profitably mine most of the time, and for a short time when I couldn't, I simply shut off the GPUs and didn't lose. And if the price had kept plummeting, I'd have sold the GPUs to gamers without losing nearly as much as an investment in the currency would have. In other words, while the price was going up, mining was profitable, and while the price was going down, mining was either profitable or neutral.

Of course ASICs will change that, and miners will be more exposed to the risk of currency failure.

Yeah, that's for sure.  Back when I was big into GPU mining I made about half my initial investment back through selling BTC and got back about 60% of my original cost on gear reselling everything, so I made out ahead a bit.

Now with ASIC's, there's basically little resale value unless you can find another miner willing to take your gear.  I'm sure it'll be a lot harder to resell the specialized ASIC's than the general GPU/motherboards.

I don't really mine for the money, so I will just keep on mining :p
then again, I cant mine atm as I fried my mobo... :/

Yup most of these ASICs will have a close to zero resale value if the BTC market crashes. I think BFL mentioned that they had to lock down the chip so it cannot perform generic brute-force on SHA-256 encrypted data (because of export limitations).

Care to elaborate on this?

I'm not sure what you would like me to elaborate on. Bitcoin block generation is based upon the SHA-256 encryption algorithm. The ASICs and any mining equipment is performing this same function to produce hashes and check to see if it is a winning hash. SHA-256 is also used to encrypt data and passwords...

SHA-256 is a one-way hashing algorithm, not encryption algorithm.

If you encrypt N bytes, you get N bytes out. And it is possible to decrypt these bytes back to the original value.
If you hash N bytes, you always get a fixed-size hash as output. And it is not possible to compute the original value from the hash (well, unless you bruteforce), hence the name "one-way hash".

Good point. Thanks for clarifying mrb.

What do you guys think is up with the Bitcoin rise? We just hit 25,000 Ghash, up from 20,000 Ghash just 2 months ago. I'm not making as much coins now :(

I think that's FPGAs, but the gold rush of bitcoin is soon to be over. I will miss the crazy rigs and shared experience of something truly innovative. Now it's proving times for BTC; either it goes into stable growth (=only investors use it), skyhigh (=it reaches the sheeple and usage explodes, this would of course really be a problem because of diskspace (atleast for me I'm on SSD)) or bitcoin will be seen as a blip on the history of currencies as prices collapse and only a few ASIC fanboys keep mining.

I'm just not convinced (difficulty = price) is 100% solid.

The only buyers today are speculators, as time go dire they will need to selloff (like everyone will sell everything) to pay for rent and food. Only those that survive without selling might see BTCs true value at the end of the current civilizations currencies.

But back on subject; I think we will indeed experience the difficulty drop I was talking about in the first post. ASIC won't be delivered before block reward halving, and prices are going down... So all those GPUs falling off must show, only ~4000 blocks left now, or what do you think?

Looks like nice steady growth. Mining is profitable, so miners are increasing capacity and new miners are coming on board.

http://bitcoin.sipa.be/speed-small-lin-10k.png