Building computer for mining

[Anonymous]

It will threaten the network if one pool approaches 50% of the total. Has anyone thought of that?

[1715002456]

1715002456

[1715002456]

1715002456

[1715002456]

1715002456

[1715002456]

1715002456

[bitcool]

Now I think the income, in dollar term, for average miner may be constant.

There are two factors in play and they have opposite effects:
1. increasing competition and difficulty for the mining operations
2. because of this increased mining cost, the value of bitcoin in dollar term will increase

[sethsethseth]

I have a crappy video cards in my current rig, and it seems like at current bitcoin prices, I could just go slap a couple 5970's in there and, even at $600 or $700 each, have them paid off in a month or so since I don't pay for electricity.  Is this right? It seems crazy to me.

[Fractality]

It will threaten the network if one pool approaches 50% of the total. Has anyone thought of that?

Just wondering, how do the pools work, do they actually assign work to the participants (like ranges of nonces to try)? My thinking was that it probably wouldn't be necessary - my guess is nonces are tested randomly, and the search space is big enough so that conflicts in the pool are unlikely. Am I mistaken?

So if pools don't assign jobs, I don't think they could threaten the network?

[Garrett Burgwardt]

I have a crappy video cards in my current rig, and it seems like at current bitcoin prices, I could just go slap a couple 5970's in there and, even at $600 or $700 each, have them paid off in a month or so since I don't pay for electricity.  Is this right? It seems crazy to me.

Given that the difficulty is rising quite quickly, the odds of you paying off a couple of 5970s looks bleak. You'd likely recoup a portion but not necessarily all. And if you live with your parents, watch out. They don't want to be hit with an electric bill that is 100 USD higher than the previous month.

[sethsethseth]

with each 5970 doing 535 Mhash/s as it shows here: http://pastebin.com/AvymGnMJ
that is one block every 1day 5hr according to the calculator http://www.alloscomp.com/bitcoin/calculator.php
so i don't see how this can possibly be unprofitable at nearly $50/day in generated value.  Even building a dedicated mining rig seems hard to lose money on when bitcoins are $1 each.  Exactly how fast is the difficulty rising?

[Ricochet]

with each 5970 doing 535 Mhash/s as it shows here: http://pastebin.com/AvymGnMJ
that is one block every 1day 5hr according to the calculator http://www.alloscomp.com/bitcoin/calculator.php
so i don't see how this can possibly be unprofitable at nearly $50/day in generated value.  Even building a dedicated mining rig seems hard to lose money on when bitcoins are $1 each.  Exactly how fast is the difficulty rising?

See http://bitcointalk.org/index.php?topic=2345.0 for some data and charts on that.

[Raulo]

o i don't see how this can possibly be unprofitable at nearly $50/day in generated value.  

It can be unprofitable if these things happen:
1. Difficulty rises too quickly
Plug different values to this calculator (current hashspeed is about 250 GH/s) and see how many bitcoins you can generate with different assumptions.
http://www.taters.net/btcgc.php
2. BTC price drops
3. Hardware breaks

Even building a dedicated mining rig seems hard to lose money on when bitcoins are $1 each.  Exactly how fast is the difficulty rising?

Difficulty was rising 13-20% per 2016 blocks since the beginning of the year but the next difficulty increase on probably February 18th is going to be likely more than 30%. The rise in BTC/$ exchange rate lured a lot of new miners.

[racerx]

o i don't see how this can possibly be unprofitable at nearly $50/day in generated value.  

It can be unprofitable if these things happen:
1. Difficulty rises too quickly
Plug different values to this calculator (current hashspeed is about 250 GH/s) and see how many bitcoins you can generate with different assumptions.
http://www.taters.net/btcgc.php
2. BTC price drops
3. Hardware breaks

Even building a dedicated mining rig seems hard to lose money on when bitcoins are $1 each.  Exactly how fast is the difficulty rising?

Difficulty was rising 13-20% per 2016 blocks since the beginning of the year but the next difficulty increase on probably February 18th is going to be likely more than 30%. The rise in BTC/$ exchange rate lured a lot of new miners.

But he is correct that todays difficulty and exchange rate means that it is profitable.. today, who knows what tomorow will bring.

Im the the same boat, and am seriously considering cashing in ~700BTC and purchasing a HD5970. Its going to cost me $0 real money for a punt... Though perhaps spending that $700 on buying more BTC's would be a better financial decision?

Plug different values to this calculator (current hashspeed is about 250 GH/s) and see how many bitcoins you can generate with different assumptions.

Not 100% sure what this data means, and what Im supposed to plug into it... eg: here's the result for 600khashes (1x 5970) and the 250Ghashses you said..

Statistics
Current control of network   0%
Coins expected next block   0
Network growth per block   0.014177%
Blocks for network to double   4889.6

Block expectations
Block   Expected coins   USD Value   Network Size
1000   0.11   $0.12   288.075 ghash/s
5000   0.43   $0.44   507.888 ghash/s
10000   0.64   $0.66   1031.8 ghash/s
25000   0.82   $0.85   8651.254 ghash/s
50000   0.85   $0.87   299376.773 ghash/s

[Raulo]

Not 100% sure what this data means, and what Im supposed to plug into it... eg: here's the result for 600khashes (1x 5970) and the 250Ghashses you said..

5970 has 600000 khashes/s or 0.6 Gh/s.

[racerx]

oops, missed a few 0's. So how do I discover the current network hash rate?

Using .6G#'s for a 5970 and 250G#'s in total I get this:

Statistics
Current control of network   0.24%
Coins expected next block   0.12
Network growth per block   0.014177%
Blocks for network to double   4889.6

Block expectations
Block   Expected coins   USD Value   Network Size
1000   111.89   $115.25   288.075 ghash/s
5000   429.85   $442.75   507.888 ghash/s
10000   641.44   $660.69   1031.8 ghash/s
25000   822.1   $846.76   8651.254 ghash/s
50000   845.85   $871.23   299376.773 ghash/s


What do the expectations mean? Not sure I understand.

[Raulo]

oops, missed a few 0's. So how do I discover the current network hash rate?

Here is the graph:
http://bitcointalk.org/index.php?topic=3024.msg42173#msg42173
You can calculate past network hashspeed based on difficulty. The averaged hashspeed that set the current difficulty was
2^32*25998/600=186 Gh/s. Current hashspeed is 230-270 Gh/s (it's hard to calculate it accurately due to block generation variability) which you can roughly calculate based on expected difficulty in the next round:
http://nullvoid.org/bitcoin/

Block expectations
Block   Expected coins   USD Value   Network Size
1000   111.89   $115.25   288.075 ghash/s
5000   429.85   $442.75   507.888 ghash/s
10000   641.44   $660.69   1031.8 ghash/s
25000   822.1   $846.76   8651.254 ghash/s
50000   845.85   $871.23   299376.773 ghash/s

What do the expectations mean? Not sure I understand.

Assuming the entered hashspeed and global hashspeed growth, you should expect to mine 111.89 coins 1000 blocks from now (about 5 days), 429.95 BTC 5000 blocks from now, etc. Due to difficulty increase, it's going to be less and less each round. If the difficulty increases less, you will get more, if more, you will get less. It is absolutely certain that the difficulty is going to increase as long as mining is profitable. The more profitable, the more people will join mining or expand their mining. It's hard to estimate by how much.  It depends on bitcoin popularity, BTC price and other factors. As you can see, 5970 should pay back itself but if the BTC price drops or difficulty increases more than the entered value, it won't. Due to recent BTC price increases and Bitcoin being slashdotted, the next two difficulty increases will be brutal. What will be next is hard to tell.

[racerx]

Decisions decisions, not a clear decision on whether outlaying for a 5970 is going to be a prudent decision and pay itself off, Unless of course rates continue to grow. If btc's are x10 by the end of 2011 then its well well worth it. Very exciting times for sure.

Other issue is trying ti actually find a 2gb 5970 in stock anywhere in the UK!

[alikim]

Has anyone tried to use encryption chips, which have SHA256 hardwired in them?

[bitcool]

Now I think the income, in dollar term, for average miner may be constant.
There are two factors in play and they have opposite effects:
1. increasing competition and difficulty for the mining operations
2. because of this increased mining cost, the value of bitcoin in dollar term will increase

Big block production drop (30%) today due to difficulty increase:
http://mining.bitcoin.cz/stats/graphs/

It may take a few more days to work out excess due to recent surge in mining investment. For the longer term though higher difficulty begets higher $ value.

I expect btc $ will trade between $1.20~$1.40 soon.