Bitcoin Forum

Let's take a look at what's coming online soon:

Who	Hash Rate	Notes
100TH	70-100	Initial rollout of 70TH at the end of July followed by 30TH
Avalon Batch #3	~50TH	600 units @ ~82GH = 50TH
Avalon Trade-in	20TH?	Unknown. Guesstimate 20th...should be more
ASICMINER new Blade+USB	15TH?	No idea???
ASICMINER solo mining	15TH?	No idea???
BFL	30TH	Let's give them the benefit of the doubt and say they ship 500 singles worth of orders for the month. They claim ability to ship 3-400 units combined models a DAY

Lower bound: 70+50+20+15+15+30 = 180TH
More likely: 100+50+20+15+15+30 = 210TH


If someone has better guesstimates about what ASICMINER is going to do, I'd be interested in those numbers.


The bloodbath coming in the fall is going to be even worse :P

In my spreadsheet I calculated 180T of new hashing power in August. So our numbers are very close. I didn't contribute so much to 100TH and ASICMiner sales (blades and USB), but something more to Avalon chips sales (DIY ppl are very commited).

Yeah I left out the chip sales. I don't have great confidence in Avalon's ability to deliver on time. It will also take some time for a significant amount of chips to come online, so I feel like late August would be the soonest we'll see an impact from Avalon chip sales.

Avalon started delivering chips this week.

Ah, interesting. I see they've arrived in Russia. Glad I chose not to order from batch #3.

How many TH do you think we'll see from those chips by the end of August? I'm thinking 50TH max. Most will probably come online in Sept.

The amount of Avalon chips in the order was between 500K and 2.1M (the pictures of the stacks of boxes is open to interpretation). If each of those chips can do 280MH/s, then that order is between 140TH/s and 588TH/s.

Stock Avalons run fairly easily at 340-350MHz, so it's probably at least 15% more. I was basing the 50TH August guess on 500K chips.

Starting slow in late august, and then in september gradually up to 150-200TH (if the sales were about 500-750 thousands chips).
They can grow faster then 100TH because there are so many assemblers.

Of course, nothing may change in August except a lot of people getting pissed off.  ::) It is difficult to predict the future and entails a certain amount of "belief".

100TH should come online almost immediately after they get their PCBs (assuming no PCB flaws). 100TH is only 250 of the 400GH units and they expect to have those deployed in a day.

After August it's a whole other story. Bitfury, KNC, BFL, Avalon chips, several other ASICs projects...

At some point things will need to plateau though. There's only ~108K BTC generated a month and at $100/BTC that's only $10.8M to be mined a month. I see that as the limiting factor on how fast difficulty will grow. That plus power consumption. ASICMINER is probably hitting a limit with their solo mining due to power concerns since they are currently still on 130nm (pure speculation).

At a price of $100/BTC I see the difficulty increase slowing down in 12-18 months. Obviously, as the price of USD/BTC goes up, so will the diff.

It's much more difficult to predict what will happen in the fall, but from now until the end of August, we have a fairly good idea of what is coming. Avalon, BFL, AM and Bitfury all have working hardware.

there is a difference between having working hardware and being able to deliver in quantity. AM seems to be the only one that has exhibited any extensive production capability.

Avalon have fixed their production issues and finished delivery of batch #2. Batch #3 is soon to follow...production should not be a problem. However, refunds from batch #3 might be significant, but then again Avalon can just self mine with refunded units.

BFL was only given credit for 30TH worth of product when they have actually sold hundreds of TH, so your point is taken w.r.t. to BFL.

100TH is an unknown here in terms of ability to execute. However, their hardware footprint is much smaller than Avalon and BFL, so they do not need to produce in the same quantity. Plus, they are self mining so they do not need to deal with customer deliveries. From what I know, they already have working prototype boards which they used to slightly improve their design prior to sending the designs to the assembler (so at least some bugs have been worked out).

What about BitFury devices not destined for 100TH? I remember seeing at least 60 units of Megabigpower 400GH/s August delivery kits available for sale at one point.

Well, Although there are many chips, there are some defects during the connection.

Even adding 200Thash ir even worse, they cannot stand long.

This is a legitimate concern, however, megabigpower has stated August delivery only guarantees shipping/delivery by August 31st. That said, it's very possible they will ship earlier. Personally, I think they will start shipping soon after the 100TH deployment...which might be even be early August. Since I actually have an order with them I probably biased myself into discounting delivery until Aug 31st for earnings projections.

This would be a prime reason to switch to alts :D. As for august, I think that 200 TH is possible

BTCGarden claims they'll put 200-300THash online in september, check it out: https://bitcointalk.org/index.php?topic=213172.msg2761067#msg2761067

I hesitate to guess at what will happen after August, but it is going to be insane...probably over 1PH in November.


I haven't been following BTCGarden too closely, but are they still doing 130nm? The power needed to deploy 200-300TH of 130nm chips is not trivial. I'd really be surprised if they had no delays after September...every other ASICs producer has had delays after getting chips.

I am projecting around 600 TH/s by the end of Aug.

https://i.imgur.com/aL4YmGc.png

What's the projection using the 30 day average?

The 30 day average will be at around 400 TH/s at the end of the month.  The 7 day average seems to be the best statistical indicator though.

They've had trouble shipping their chips.  They haven't had trouble getting their chips, as far as we know.


Eh, there's no reason to think the hashrate growth can be modeled with a simple exponential function anymore. It's all about how many companies decide to make chips, and how many they make. There's no way to predict that.

Yeah, you wouldn't think that there was such a good statistical fit but the growth rate has been pretty predictable since April.

That would a good percentage up.  But with the evidence, I think that's what we'll see.  Total hashrate right now is around 270 from the webpage I checked.  We could double that up if your figures are low.

I see this happening eventually, and depending on how much the Asics are costing should effect whether they decide to deploy more Asics.  Let's say I was thinking of deploying 10 TH/s of mining devices.  I could calculate how much I expect to earn based on the current total hashrate, the bitcoin exchange rate, how much production cost, and then enter in estimate of how much hash rate will increase in future to determine how fast profit will drop off.  Say BTC price is $100, 108k BTC/ month available.  At 270 TH/s total hash rate if I add my 10 TH/s I will have 3.5%, and expect 12.6 BTC, $1300 per day or so.  If hash rate goes up 20x to 5400 TH/s, my 10 TH/s will only mine maybe 0.6 BTC, not so much anymore.  Electricity cost and production cost becomes a bigger factor and profitability down a lot.  Unless BTC goes up the plateau will show up when there's no longer profit.

Half way through august and half way to +200TH =P

The 100TH mine has been delayed a few days, so this run up has probably been attributed to Avalon B3 and BFL. Maybe those missing avalon chips are hashing somewhere too.

it seems the exponential fit of 7 days average is toooo slow ;D
it hit 500TH/s and it's not 8/19th yet!!
is there a simple tool online to create these graphs?

I am showing the 7 day average right now is 433 TH/s.

7 Day average is now 525.72.  Its looking like the projection was pretty close...

The exponential graph is spot on prediction so far.

Seems like a 100TH kick in the last 24 hrs. Anyone know where it came from?

Maybe just a luck. Wait a bit more

Looks like the estimate might be a little low.

bitcoincharts.com = Network total   672.653 Thash/s

That's the rate over a very short period which could include some luck.  We will have to see it it holds.  Currently showing 702 TH/s.  7 Day average is still about 550 TH/s.

If it holds at 670'ish TH, the stage is set for 100 Million difficulty.