Network Size In Relation To Price

[Jonathan Ryan Owens]

I'm looking for input from everyone. All opinions are welcome.

My opinion, sirs:

Currently, I believe that total network power follows price. I believe most miners are currently underwater, where perhaps 10% of the network is highly profitable (20% above costs monthly) while the rest of the network is either in a zero sum or negative position to some degree.

I also see as a real disconnect between the producers (the miners) and the price (the investors / speculators / users).

The miners are always looking to hold, sell or use. I assume that new, smaller miners try to hold, but with at least some cases, large utility bills slap them with reality pretty quickly.

Here's what I see coming in one possible future:

FPGA and sASIC devices will come to market and make GPU mining unprofitable. BFL (http://butterflylabs.com) has had a shaky start, and some of their performance claims may have been exaggerated, but the chips required for efficient $ / MH/s, even at $2 Bitcoin, are coming, and I believe that it is going to completely eliminate energy hungry mining (GPU).

There are two features that are important in such a change.

• Miners will be paying mostly for the sunk cost of the equipment, and have 1/40th of the electric bill as compared to current hashing methods
  
• Miners will have higher profit margins, with significantly reduced ongoing costs. Miners will amortize the cost of equipment over a longer period of time, and we will see a new wave of hoarding begin. Selling demand will drop, and the 2nd wave of early adopters, who think in terms of years rather than weeks or months, will become the norm. Rather than pay an ongoing $700/mo electric bill, you're now paying $4000 up front for 5GH/s + of hashing power, for the price of a home computer monthly power draw.
  

What do YOU think the future of Bitcoin network is? A mix of GPU, FPGA and sASIC block miners, or a wholesale shift to FPGA and sASIC designs as we saw in the crossover from CPU to GPU mining?

I appreciate your thoughts.

-Jonathan

[1715216814]

1715216814

[teflone]

http://2.bp.blogspot.com/-c_Ugm0CNAaE/Tcd4K4R-1oI/AAAAAAAAAfs/FxhbZZJwI1g/s1600/in-the-year-2000.jpg

[Jonathan Ryan Owens]

FTFY



Ron Paul becomes president, abolishes the Federal Reserve and adopts the Gold Standard. China mobilizes as does Russia. Bitcoin is an internationally recognized form of scrip, with many derivatives in circulation. In the 3rd world war, EMP's are all the rage, and since everyone has become reliant on computers for work, entertainment and finally, currency, we are all reduced to the stone age.

[notme]

Don't go and spoil it for them now.

[FlipPro]

Your numbers are all over the place, as with most bears..

I calculated how much I pay for 3G/Hs in electricity and it comes out to $3...

There are millions of people with the same situation as me...

EDIT: And I actually live in a pretty big state. I calculated how much my electricity would be if it were "EXPENSIVE" (0.15KWH) and it came out to about 7 dollars a day. Still a far cry from the $700 you were crying about.

[Jonathan Ryan Owens]

Your numbers are all over the place, as with most bears..

I calculated how much I pay for 3G/Hs in electricity and it comes out to $3...

There are millions of people with the same situation as me...

EDIT: And I actually live in a pretty big state. I calculated how much my electricity would be if it were "EXPENSIVE" (0.15KWH) and it came out to about 7 dollars a day. Still a far cry from the $700 you were crying about.

You pay $0.0015 per KW/h? That's amazing.

There aren't millions of people mining.

$0.15 @ 2500 Watts = $9 per day

$0.15 is the listed average base rate that represents most urbanized areas. Tiered rates apply, and aren't usually factored in the two dimensional forecasting that most miners seem to use. Rates can go as high as $0.50 per Kw/H. It does depend on where you live, of course. In California, there are quite a few areas where mining is simply out of the question. This also applies to other states.

Thanks for your input, and congratulations on the (almost) free electricity!

-Jonathan

[trogdorjw73]

One problem with the above ideal is that it assumes just because an FPGA might come out that can hash faster than a GPU at a lower power cost, people using those FPGAs will suddenly be profitable at $2/BTC. What will likely end up happening is that all the people heavily invested in GPU mining will start to switch to FPGAs, and then we'll have a lot more hashing power but the total amount of BTC generated per day will still stay at 7200 BTC. That would work out to $14400 per day available to split among all the miners, and there will be a race to get a bigger piece of the pie by mining faster, but no matter how many GH/s you can do you can't make more than $14400 per day.

I'd guess there are at least several hundred people trying to make money off of BTC mining--let's just call it 1000 for kicks. So each of those has the potential to make $14.40 per day if the price is $2 per BTC. Right now as one example, Deepbit.net has nearly 3000 GH/s of compute power, and that brings in about 2500 BTC per day. Like FlipPro I pay about $1 per day per 1GH/s, so the cost for 3000 GH/s at Deepbit is probably around $3000 per day. If we get FPGAs that can do $0.10 per day per GH/s, then we'll simply end up increasing compute power ten fold and we're still back to right where we are now (minus the people that drop out because they don't want to go with FPGAs -- like most CPU miners have quit outside of those that think playing the lottery is cost effective).

Really, to make FPGAs a practical solution, it needs to be cost effective in terms of buying, programming, and deploying the hardware. It looks like $400 to $500 for an FPGA board might be typical for 200MH/s, but power use will be 1/10 of a GPU doing equivalent work. The rest of the system running those FPGA miners will probalby use 40W minimum, and that might be able to push four FPGAs. So 800MH/s for 100W of power use seems reasonable, which is now only about five times as efficient as GPUs. What's more, you're looking at sinking $2000 into a system that would generate maybe 0.7 BTC per day at the current rate. Cost of power would be maybe $0.20 per day, for net income of $1.20 per day (per miner PC), but unless the FPGA boards are substantially cheaper you'd need 4.5 years just to cover the cost of the hardware investment. So a better target would be FPGAs that cost less than $50 that can do 200MH/s, and that's not looking likely.

What will make BTC mining with FPGAs profitable? That's right: higher prices of BTC. Just like with GPUs. When you need to expect your income from mining to decrease over time (i.e. the difficulty will increase if mining is profitable, so your BTC per day goes down), the only way investing in hardware makes sense to me is if you can pay off the hardware within six months. Otherwise you're sailing a sinking ship.

[freshzive]

California is ridiculous. PG&E increases the rate you pay per KWH with each 50KWH you consume per month. After 600KWH, I was paying over 0.30/KWH. Who knows how ridiculous it would get with a real mining set up...totally not worth it.

[FlipPro]

Your numbers are all over the place, as with most bears..

I calculated how much I pay for 3G/Hs in electricity and it comes out to $3...

There are millions of people with the same situation as me...

EDIT: And I actually live in a pretty big state. I calculated how much my electricity would be if it were "EXPENSIVE" (0.15KWH) and it came out to about 7 dollars a day. Still a far cry from the $700 you were crying about.

You pay $0.0015 per KW/h? That's amazing.

There aren't millions of people mining.

$0.15 @ 2500 Watts = $9 per day

$0.15 is the listed average base rate that represents most urbanized areas. Tiered rates apply, and aren't usually factored in the two dimensional forecasting that most miners seem to use. Rates can go as high as $0.50 per Kw/H. It does depend on where you live, of course. In California, there are quite a few areas where mining is simply out of the question. This also applies to other states.

Thanks for your input, and congratulations on the (almost) free electricity!

-Jonathan

I'm at 0.04 which is about 2.40 USD a day...

[Jonathan Ryan Owens]

Your numbers are all over the place, as with most bears..

I calculated how much I pay for 3G/Hs in electricity and it comes out to $3...

There are millions of people with the same situation as me...

EDIT: And I actually live in a pretty big state. I calculated how much my electricity would be if it were "EXPENSIVE" (0.15KWH) and it came out to about 7 dollars a day. Still a far cry from the $700 you were crying about.

You pay $0.0015 per KW/h? That's amazing.

There aren't millions of people mining.

$0.15 @ 2500 Watts = $9 per day

$0.15 is the listed average base rate that represents most urbanized areas. Tiered rates apply, and aren't usually factored in the two dimensional forecasting that most miners seem to use. Rates can go as high as $0.50 per Kw/H. It does depend on where you live, of course. In California, there are quite a few areas where mining is simply out of the question. This also applies to other states.

Thanks for your input, and congratulations on the (almost) free electricity!

-Jonathan

I'm at 0.04 which is about 2.40 USD a day...

Ah, so you make about $170 monthly at the current exchange rate. How much was your hardware, and have you broken even yet?

What do you think about FPGA and sASIC as a replacement of GPU? Is it just another upgrade hassle, as CPU to GPU mining was?

Sorry to be snarky.

-Jonathan

[FlipPro]

Your numbers are all over the place, as with most bears..

I calculated how much I pay for 3G/Hs in electricity and it comes out to $3...

There are millions of people with the same situation as me...

EDIT: And I actually live in a pretty big state. I calculated how much my electricity would be if it were "EXPENSIVE" (0.15KWH) and it came out to about 7 dollars a day. Still a far cry from the $700 you were crying about.

You pay $0.0015 per KW/h? That's amazing.

There aren't millions of people mining.

$0.15 @ 2500 Watts = $9 per day

$0.15 is the listed average base rate that represents most urbanized areas. Tiered rates apply, and aren't usually factored in the two dimensional forecasting that most miners seem to use. Rates can go as high as $0.50 per Kw/H. It does depend on where you live, of course. In California, there are quite a few areas where mining is simply out of the question. This also applies to other states.

Thanks for your input, and congratulations on the (almost) free electricity!

-Jonathan

I'm at 0.04 which is about 2.40 USD a day...

Ah, so you make about $170 monthly at the current exchange rate. How much was your hardware, and have you broken even yet?

What do you think about FPGA and sASIC as a replacement of GPU? Is it just another upgrade hassle, as CPU to GPU mining was?

Sorry to be snarky.

-Jonathan

I REALLY don't think of Bitcoin in Fiat terms anymore. If I can make a successful trade here and there great, but in the end Bitcoin is BETTER than any fiat currency. And my goal in the end is MORE BITCOINS, not fiat... If I were to go by your standards I really make "nothing" since I haven't "cashed" out a dime since I got involved with this project 5 months ago.

EDIT: And to answer your question, I believe that when FPGA boards become more affordable/accessible, we will see a mass switch.

I will continue mining with my GPU'S till they wear out, then I plan to replace my hardware to more modern equipment using Bitcoins .

[jwzguy]

I believe most miners are currently underwater, where perhaps 10% of the network is highly profitable (20% above costs monthly) while the rest of the network is either in a zero sum or negative position to some degree.

Based on what?