I have 1.5 Antminers.
One that always performced fine, and one that always performed badly, with the 2nd board shipped back after dying, and now supposedly lost in mail. The bad performing board is april 2014 issue.

Got all 3 board modded to < 3kOhm, and Now I am running around 248 Watts for 210 GH/s (GHash.io average, 1 day). The single board is doing 65 GH/s, the fullminer is doing 145 GH/s. Both do have some HW errors, < 1%.

So I am at < 1.2W/GHS. So even the relatively high power costs here allows me to make a profit for a few more weeks/months.

Go look here: https://bitcoinwisdom.com/bitcoin/difficulty

Seems like you are losing value at with at least 15-20% per 2 weeks (Output drops by the difficult rate increase, in case exchange rate stable).

In reality output + value drops more, power price is a constant.

Antminer S1 running @ 200GH/s uses 400 Watts/hour = 9.6 kWh per day.
It produces $5.462 woth of bitcoins  at todays exchange rate, going down by the diff increase every 2 weeks.

Not sure what your power price is. But assuming it is in the high end = $0.40 /kWh. Then that is 9.6 * 0.40 = $3.84 /day.
So the profit is now $5.452 - $3.84 = $1.612 per day.

In 4.5 days, the difficulty goes up by say 12%, so the output is $5.462 / 1.12 = $4.88. Minus power $3.84 = $1.04.

So at the power price, your net mining profits drops 35% in 4.5 days.

You need to calculate yourself. As it looks now, profits is getting low, and there will be lesst investment in new equipment as it will never break even at the current retail prices. Only the manufacturers will turn a profit, or those with almost free electricity.

The Antminer S1 can be pencil modded to gain a significant efficiency boost. Slower hashrate, at much lower power cost. You can get it to be twice as efficient. That is the trick they used in the S2, undervolt the chips in the S1.

But when you do the calculations, remember that 1GH/s today is 15% more worth than 1GH/s in 2 weeks.

I have one question.

Been trying to solo-mine bfgminer -> bitcoind, but bfgminer keeps working on old stuff way after the block has changed.

How can I use blocknotify of bitcoind to tell bfgminer to actually pick up the new block and new work ?
With eloipool in between, all works fine, but it is overkill, and it eats 100% CPU on one core.

Any SIGxxx to send ? Something else ?

Use cron to rotate your own config in place and restart the miner.

One thing in the same category I am more intered in is switching pool based on luck.

Say I have worked at pool 1 for 1 hour until its luck has decreased to 100%. Now I feel that I will get less profit from the rest of my shares. If I work 1 more hour, I will get half the pay I would have received if I had dedicated my miner to another pool with 100% luck.

Actually, after 2 hours, switching will give me:
1 hour that gets dilluted to 50% value (2 hour block)
1 hour at 100% luck.

So by switching, I would get 1.50 times the value compared to mining the same block to finish.

I am seeing more and more very long blocks, and bad pool luck. So would like to try something like that.

I have set up local mining with 6x Antminer U1 on bitcoin, hoping for the lottery ticket to come out.

bitcoind + eloipoo.py - works fine, and I get diff 16 work to do.
One proboem, eloipool.py seems to eat 100% CPU on one core all the time. Can I do something to lower the CPU usage ? I wonder what it is doi9ng that eats so much CPU.

I have another question.

My bfgminer connected to bitcoind seems to keep working on old blocks.
I have disabled longpoll, scan-time 1 sec, queue = 0.

Would I need to set up pool software to get my solo miner to mine the current block without too much delay ?
Some startum proxy ?

I have a question about solo mining, or mining in general.

As I understand it, we are trying to find some sha-256 hash. We are all working on the same block, from hopefully different starting positions, and the one to find the right hash first wins the block. Is this more or less correct understood ?

If yes, then solo mining would always be considered as a valid lottery ticket ? No matter what difficulty we have ?

I was also hit. Learned more about how to find the transaction in other blockchains.
There was a warning when I ordered though, first time it would not create a vallet for me to pay to. So I knew it was probably a blockchain.info wallet I was going to pay to.

Now waiting to see if I get a "shipped" message within the next hour or so, or if they have stacked up so many orders that it will be impossible to ship all out today.

I also feel much more confident buying here than the "we hope to ship while it might still be profitable" companies.

When mining on eligius, I had it fail over to ghash quite a few times (I am on 60/60 Mbit fiber, so not my Intenet connection).
Switched it over to GHash as primary, and I have not seen a single failover.

I am in Europe, and the Eligius server in the US. Does not seem to work out too nice. For GHash I use the UK server.

Make sure you use servers with good connectivity.

Lots of existing hardware using more than 2W/GHS will start to get unprofitable soo, and some of it wil be turned off, or resold to poor souls who plugs in in, realises it is bad, and tries to shift it on. The 2W/GHS will be profitable a little longer.

We have the 1W/GHS units coming up (Antminer S2, KnC and others). They will be profitable longer. But I see most of the units running today being turned off within the next year. If many gets out from the mining business, maybe we will have a difficulty decrease ? Or a slow increase in difficulty.

Profitability in mining lies with the success of getting peple to not buy more equipment, and turn their old stuff off. Difficulty increase needs to slow down. Lst one was 12% instead of the 20% before that. High difficulty increase can only be caused by new hardware. And since there is so much hardware out there, I don't see it increasign that much any more. I see it slowing down.

Price is down because of the block chain.info issues.
Will soon correct itself. I see 10%+ upside over the next month.

It costs 0.988 BTC.
Playing with 15% difficulty - stable BTC price.

For now it will do like 5 days of 0.020 = 0.100
Then 2 weeks of 0.017 = 0.238
Then 2 weeks = 0.207
Then 2 weeks = 0.180
Then 2 weeks = 0.155
Then 2 weeks = 0.135
Total after 11 weeks is 1,015 BTC.

Power cost   = 61 kWh/week = 671 kWh. At $0.1/kWh that is $6.71 = 0.011 BTC.
Even at 4 times the power cost, it will be profitable within 11 weeks.

You can continue the calculations - say power costs = $0.40 / kWh. That is 0.088 for 2 weeks.
13 weeks = 0.117
15 weeks = 0.102
17 weeks = 0.088 <- Lifetime at $0.40/kWh
19 weeks = 0.077
21 weeks = 0.067
23 weeks = 0.058
25 weeks = 0.050
27 weeks = 0.044 <- lifetime at $0.20/kWh

The S2 has half the power costs, so will add like 10 weeks lifetime. But not shippinjg before in 3-4 weeks, so make it 6 weeks longer. And it will start earning at higher difficulty lower profits.

I have the same issue. And bitmaintech is saying lots of customers have the same issue.

blockchaininfo was having problems yesterday. Now they are down again. At some point number of unconfirmed transactions was 17000+.

I figured out my transaction was part of block 291285 by looking at one of the other block explorers. And it had at that point 46 confirms at ghash.io, where I mine some of my BTC.

This morning blockchain info was fine again for my transaction.

Guess the blockchain.info problems are causing the drop in BTC price. Hope it will recover soon.

I have bought and received one Atnminer S1. Looked fine. One blade died within 24 hours. It is on the way back. They promise a coupon for discount on next purchase.

Interesting enough, the bad blade was mounted with 5 torx screws, and one Philips screw. So it looks like the bad blade was not mounted at the main production line.

But I still have a good feeling about BitmainTech, and ordered another, and seriously considers the S2

I also had an order go in expired status while waiting for payment to get confirmed. My high-prio payment has been stuck hours with no confirmations to be found at blockchain.info.

Lets see what happens tomorrow

Bitmaintech is known for delivering on or before time.

Lets see around April 1st when batch 1 ships. Now is also the time to buy BTC; and then order when price for BTC is up 10%. You pay in $ exchange rate at the time of order.

Got 2 PSUs from a Compaq 9000 server, including the backplane + cables.
Have not tried to connect anything yet.

Anybody have an ide how to jump-start it ? Just use the ATX like cable ?
And then with all the cables, I can easily measure where I get 12V out.

It is rated 750W output @ 220V. 2 of those on the backplane = 1500W out.
Should be more than enough for 3 units, probably fine for 4 units.

4x360 @ wall = 1440 * 0.85 efficiency = 1225 power draw.

Did anybody measure actual power consumption by one Antminer S1 ?

The PSUs are rated 6A @220V input = 1320W, so 750W out is pretty poor. But maybe they are efficent, and the 1320W is startup peak.

You PSU is listed as having 2x 12V rails. You need to ensure you pull power from the different rails. Maybe one for PCI-e, and maybe you need to get the other 12V from the mainboard connector ? Or some other connector.

A single rail PSU is easier. Use any connectors you want.

PS about the 3 weeks wait for the Antminer S2, it will cost you some loist profit in the meantime, and you will have 2 difficulty adjustments. But just calculate that into your calculations.

I completely agree on the power costs.
Here in Denmark, power cost is $0.40/kWh.
180 GHS gets you around 0.020 BTC per day. On the S1 that is 0.36Wh * 24h = 8,64 kWh = $3,46/day = 0.005 BTC or thereabout.
That is already 25% of the mined BTC going to power here. If I can get that down to 12.5%, it would be nice. And the miner would be sustainable for a longer time.

Now I have solar power, on my root, I am taxed on that, so using my own power instead of selling gives me a power cost of 0.1$/kWh, but I only have surplus production of a few hundred kWh/year. Can't expand my solar installation without getting hit by new stricter tax rates (where I can't store surplus energy in the powergrid for free, but will have to sell surplus for low price, and buy back at the 0.40).

Power costs is important, especially in Europe. Enjoy cheap power where it is, but at some point, the power costs will be more than the mining output. When the S1 is worthless, I can still earn 5.6 * $3,46 = $19.4/day on an S2. Don't know what will happen to difficulty when older equipment starts hitting the not-profitable line.