Google Doc with exponential fit, lead time, and linear rate target

[ujka]

well, asicminer predicts they'll push out another 800-1000TH by the end of this year.

They will push whatever they have to (or can) to stay in that slice of 25% to 33% of network hashrate. Remember, they said they'll never 'own' more then 1/3 of the network.

[1715048911]

1715048911

[1715048911]

1715048911

[ujka]

well, asicminer predicts they'll push out another 800-1000TH by the end of this year.

And forgot to ask - where did you come across these numbers?

[jspielberg]

One report for the asicminer to deploy 1PH/s by years end:
https://bitcointalk.org/index.php?topic=246375.msg2612413#msg2612413

In other news:
I have updated the spreadsheet with a global hash rate since I think people can model for themselves a bit better based on what vendors are announcing as shipping/shipped.
I made the exponential value a global so you can tweak to follow the current difficulty grows to do "what-if" growth scenarios.

I have reset the exponent to a more optimistic 0.013825, which has been tracking actually difficulty more closely... however I believe the curve will steepen as Avalon, BFL, 100TH/Bitfury, Metabank, Avalon Chip DIY/clones, etc. start ramping up.

Feel free to make a copy and tweak as you like.

[SilentSonicBoom]

Interesting speculation indeed. Let's see how soon or how late hashrate is delivered by the major players in the next six months.

[Polyatomic]

Difficulty is 21.3 million now . Excellent Spreadsheat @ jspielberg.

[jspielberg]

Difficulty is 21.3 million now . Excellent Spreadsheat @ jspielberg.

I think actuals will outstrip projected in the next few retargets, but with the more optimistic growth curve exponential, it is tracking now fairly closely.

[joeventura]

Yes and if you do the math you will quickly realize that three things will happen if these numbers are true.

Using 300GH as an example for how much hashing power these statements are based on:

1. In March of 2014 (maybe April) the Diff level will be so high that 300Gh will yield between 3 and 4 BTC a month
2. By Nov 2014 you will need 1TH of equipment to make 1 BTC a month
3. Bitcoins are done by the end of 2014 for the average person


Just my opinion I may be wrong and the price of BTC may go up 30 fold then everything I just said is CRAP   

[jspielberg]

Updated spreadsheet in OP with pristine values from wolfram. The results I think are a little to optimistic, but they are what they are.  You can still tweak your delivery date, hash rate and exponential rate.

[markyminer]

Yes and if you do the math you will quickly realize that three things will happen if these numbers are true.

Using 300GH as an example for how much hashing power these statements are based on:

2. By Nov 2014 you will need 1TH of equipment to make 1 BTC a month
3. Bitcoins are done by the end of 2014 for the average person

You are forgetting it's not about the money and ability to add ever increasing amounts of hashrates. What matters is BTC extraction rate and difficultly vs energy use and cost.

So basically it's about who is mining with the most power efficient ASIC's. The big players will be forced to stop long before "the average" person when their BTC extracted per g/hash is less in value than the cost of the electricity to extract it.

Remember few people get "free" power, and no commercial businesses do.

M

[jspielberg]

Looks like the next difficulty retarget on August 3rd is surprisingly close.
Spreadsheet: 36,578,251
Allchains.info: 36,689,312
https://docs.google.com/spreadsheet/ccc?key=0Auya3iRE6az1dDc0UVgwMU52YVpTazVjSHByOGNiWHc#gid=8

The next retarget is likely to forecast way low with Avalon B3 completing, and 100TH mine coming online.

[Takeshi_Kovacs]

Yes and if you do the math you will quickly realize that three things will happen if these numbers are true.

Using 300GH as an example for how much hashing power these statements are based on:

2. By Nov 2014 you will need 1TH of equipment to make 1 BTC a month
3. Bitcoins are done by the end of 2014 for the average person

You are forgetting it's not about the money and ability to add ever increasing amounts of hashrates. What matters is BTC extraction rate and difficultly vs energy use and cost.

So basically it's about who is mining with the most power efficient ASIC's. The big players will be forced to stop long before "the average" person when their BTC extracted per g/hash is less in value than the cost of the electricity to extract it.

Remember few people get "free" power, and no commercial businesses do.

M

Don't forget though that businesses generally pay less for their electricity than individuals. I would expect that private miners who pay for power will be forced out by electricity costs long before the same will happen to businesses.

[ScaryHash]

It's now August 4th, and difficulty is 36.7 million.

Looks like you're pretty much spot on.

Good job. Hard to argue with good math.

[jspielberg]

It's now August 4th, and difficulty is 36.7 million.

Looks like you're pretty much spot on.

Good job. Hard to argue with good math.

I think the predictions based on previous growth rate is going to break down as we enter a steeper curve in the near future.  New technology projects like bitfury 100TH and avalon B3 come online in pulses.  This model takes no expected future events into account.  It is purely the exponential model of the difficulty growth since entering the "Asic Era" which I set as starting in early Feb.

My guess is that this model will be low for the next 2 months, slowly catch up to the actuals, and then be overly pessimistic in the early part of 2014.

[superresistant]

It's now August 4th, and difficulty is 36.7 million.

Looks like you're pretty much spot on.

Good job. Hard to argue with good math.

I think the predictions based on previous growth rate is going to break down as we enter a steeper curve in the near future.  New technology projects like bitfury 100TH and avalon B3 come online in pulses.  This model takes no expected future events into account.  It is purely the exponential model of the difficulty growth since entering the "Asic Era" which I set as starting in early Feb.

My guess is that this model will be low for the next 2 months, slowly catch up to the actuals, and then be overly pessimistic in the early part of 2014.

So I'll ROI my KNC shares in about 1 month... nice.

[voxelot]

Update

8/27/2013 spreadsheet prediction 62,333,209

Current difficulty 65,750,060

Still close but predictions are starting to get behind reality a bit.   

[jspielberg]

Update

8/27/2013 spreadsheet prediction 62,333,209

Current difficulty 65,750,060

Still close but predictions are starting to get behind reality a bit.   

I just updated the spreadsheet with the latest projection.
The wolfram alpha data for the exponential portion of the graph can be found at:
http://www.wolframalpha.com/input/?i=exponential+fit+%7B%7B13%2C+3651012%7D%2C%7B37%2C+4847647%7D%2C%7B96%2C+11187257%7D%2C+%7B120%2C+15605633%7D%2C%7B156%2C26162876%7D%2C%7B167%2C31256960%7D%2C%7B189%2C50810340%7D%2C%7B200%2C+65750060%7D%2C%7B210%2C85000000%7D%7D

I also added a daily linear difficulty  since the exponential won't grow forever (the red box).
Once the daily difficulty growth hits the threshhold difficulty will grow linearly at that rate rather than exponentially.

The link is still:
https://docs.google.com/spreadsheet/ccc?key=0Auya3iRE6az1dDc0UVgwMU52YVpTazVjSHByOGNiWHc&usp=sharing

I have put in a daily linear rate of 40,000,000 which seems high, but we currently haven't seen wide scale asic deliveries yet and we are already at 1,500,000 daily difficulty growth.
Feel free to copy and edit as you like.

[Diabolicus]

1 billion by christmas ... wee!

[greghawk]

Thank you for this jspielberg !

So am I correct, that if I order one KNC Jupiter right now and it´s really shipped in November, I´m not getting my ROI, right?
That means I could buy 10 or 100 of them (if I had a millon dollar) and still lose?

So what´s the deal here?

[Its About Sharing]

Thank you for this jspielberg !

So am I correct, that if I order one KNC Jupiter right now and it´s really shipped in November, I´m not getting my ROI, right?
That means I could buy 10 or 100 of them (if I had a millon dollar) and still lose?

So what´s the deal here?

KNC has already told some Jupiter orders that they will ship with 500 Ghash/s and not 400. Not sure if that changes your ROI but still a nice gesture (especially compared to BFL - "But For Loss")

I'm curious though, any other sites for good mining profitiability calculators? They are all over the place.

[ujka]

So am I correct, that if I order one KNC Jupiter right now and it´s really shipped in November, I´m not getting my ROI, right?
That means I could buy 10 or 100 of them (if I had a millon dollar) and still lose?

So what´s the deal here?

You lose, KNC wins. That's the deal. Buy more, lose more.