Questions about a mining rig.

[Sheep]

I'm thinking about building a mining rig and upgrading it every few months until all my PCIe slots are full, but I do have some concerns before I order parts.

Is it worth building a mining rig  when the butterfly labs singles are going to start shipping soon? If so, I have one more question

 Is a 16x gpu running in a 1x slot (via a 1x To 16x cable) still capable of the same hash per second that it normally would? (in a x16 slot)
   a. Even in crossfire with other cards?

[kliffen]

A GPU placed in a 1x -> 16x PCI-e riser gets the same mhashes as one in a 16x motherboard slot, mining doesn't need much data transfer.

[Sheep]

That's good news! I hope the butterfly labs singles ship soon so I can read up on whether it'd be better to get one of those.

[kliffen]

The butterflylabs singles have shipped, atleast to those who ordered in 2011. 2012 orders don't seem to have been shipped yet, i might be wrong though.

They do give out about 850 mhash/s at around 80W. If electricity is cheap in your country/where you live, GPU mining is still the way to go. If not, then FPGAs will be your thing i guess.

[Sheep]

I'm not entirely sure on the price per kwh where I live, but I think if GPU's would be easier to resell once profit dies down then that would be the way to go regardless of electricity price.

But since the butterfly labs use such low low power, they would turn profit much longer than a gpu rig. Bitcoins value would have to crash very very low for the butterfly to not turn profit still.

My only concern with the butterfly boxes is that they won't have any uses when bitcoin mining DOES become unprofitable. When that does happen what would it be good for? At least with GPU mining you can sell the GPUs and make some of that money back.

[kliffen]

The only things i can think off that a FPGA can be used for BESIDES mining is either password cracking, or a rendering farm. Not sure if buttcoin FPGAs would be very optimal for rendering though

[Sheep]

Damn, so no one would want to buy them once bitmining stops.

What would be really helpful is if somebody knew if the money you saved on power over the course of using the butterfly box is MORE than what you would get if you sold your GPUs. You know what i'm sayin'?

If the money I can save with the low power solution is more than what I would get for selling my used GPUs after everything is said and done. Than obviously the low power solution would be the way to go regardless if you were stuck with it after mining stops.

Calculating this though would be a major pain in the ass. It also wouldn't be completely accurate considering you would have to predict when bitcoin mining becomes unprofitable, and how much bitcoins would be worth.

I suppose this is where the real "risky investment" comes into play. Do I choose option A, or option B?

[legolouman]

Damn, so no one would want to buy them once bitmining stops.

What would be really helpful is if somebody knew if the money you saved on power over the course of using the butterfly box is MORE than what you would get if you sold your GPUs. You know what i'm sayin'?

If the money I can save with the low power solution is more than what I would get for selling my used GPUs after everything is said and done. Than obviously the low power solution would be the way to go regardless if you were stuck with it after mining stops.

Calculating this though would be a major pain in the ass. It also wouldn't be completely accurate considering you would have to predict when bitcoin mining becomes unprofitable, and how much bitcoins would be worth.

I suppose this is where the real "risky investment" comes into play. Do I choose option A, or option B?

Calculate that. A gpu can sell for around 50% of its value after vigorous mining.
All you need to do is calculate how much you spend on electricity, and see which is a larger amount

[Sheep]

You're right, I completely over thought that. I blame it on lack of sleep 

[Sheep]

I'm sold on the Butterfly single! If bitcoin calculator is correct, you can get 1k USD a year from running a single.

Even if I made a quarter of that i'd still be content. So this thing works just by plugging it in to an existing computer and using the software? Doesn't use any extra resources other than that right?

edit:The calculator I used was set to $10 per BTC, for the sake of accuracy in todays market I halved the amount in order to bring it to $5 per BTC.

[DeathAndTaxes]

I'm sold on the Butterfly single! If bitcoin calculator is correct, you can get 1k USD a year from running a single.

Even if I made a quarter of that i'd still be content. So this thing works just by plugging it in to an existing computer and using the software? Doesn't use any extra resources other than that right?

edit:The calculator I used was set to $10 per BTC, for the sake of accuracy in todays market I halved the amount in order to bring it to $5 per BTC.

Maybe you should do some more calculations and thinking.

If you got "a quarter" = $250 a year you would be happy?  The single cost $600 so you are talking 2.5 years just to break even.  Of course if you got $250 in first year you are unlikely to get $250 every year due to rising difficulty.  You also need to consider the electrical cost.  80W continually for a year @ $0.10 per kWh = $70.  so $250 gross isn't $250 net.  It is $180 net.  The block reward halves in Dec 2012 so that becomes $125 gross and $55 net.  

What if the device works for 3 years you collect $480 and then it breaks?  What if it breaks at 91 days (on 90 day warrant) and you have collected $80 in gross revenue?

Now I am not saying don't buy a single or that you will only make $250 gross revenue.  I am saying jumping in based on your horribly flawed "caclulations" and assumptions is just gambling not investing.

[Sheep]

I'm sold on the Butterfly single! If bitcoin calculator is correct, you can get 1k USD a year from running a single.

Even if I made a quarter of that i'd still be content. So this thing works just by plugging it in to an existing computer and using the software? Doesn't use any extra resources other than that right?

edit:The calculator I used was set to $10 per BTC, for the sake of accuracy in todays market I halved the amount in order to bring it to $5 per BTC.

Maybe you should do some more calculations and thinking.

If you got "a quarter" = $250 a year you would be happy?  The single cost $600 so you are talking 2.5 years just to break even.  Of course if you got $250 in first year you are unlikely to get $250 every year due to rising difficulty.  You also need to consider the electrical cost.  80W continually for a year @ $0.10 per kWh = $70.  so $250 gross isn't $250 net.  It is $180 net.  The block reward halves in Dec 2012 so that becomes $125 gross and $55 net.  

What if the device works for 3 years you collect $480 and then it breaks?  What if it breaks at 91 days (on 90 day warrant) and you have collected $80 in gross revenue?

Now I am not saying don't buy a single or that you will only make $250 gross revenue.  I am saying jumping in based on your horribly flawed "caclulations" and assumptions is just gambling not investing.

I'm sorry, when I edited my post I didn't think about this. My original post said 2k per year. A quarter of that for one year would be $500 after power bill and blah blah blah. The calculator I was using was set to $10 per bitcoin as default and I didn't catch that until after I posted.

I know block values are being halved, and that actually raises a question of mine. Since bitcoins would be harder to get after the halving, would they go up in price? If so this would obviously mean that I should buy the single now and start mining, and then sell when the blocks are halved.