Litecoin Theoretical Mining Rig

First of all you do not use crossfire for mining builds. Each card needs to work individually. You want a cheap CPU like an AMD sempron or cheap athlon. then you want as many hd7950's you can afford and a HUGE gold or platinum certified 1000w+ PSU. You then need a motherboard with 4 or 5 PCI express lanes.

Yes, you need risers for sure. POWERED ones for sure so you don not fry the motherboard.

http://www.gigabyte.com/products/product-page.aspx?pid=3880#ov this seems to be a popular motherboard.

Make sure you DISABLE crossfire. And DO NOT use a bridge between the cards.

Also you need to think about PSUs. Generally two 750w gold or platinum certified units are cheaper than one 1500w platinum unit.

You need to dump as much as possible in. With 2 GPUs mining litecoin, you will take a LONG time to even break even. You need to get 4 or 5 cards minimum going NOW and even then you may want to look into other coins because litecoin is becoming to difficult, much like bitcoin did.

IT should. Some people have issues getting certain gigabyte boards to see more than 4 GPUs, so read into that. ASrock seems to be the standard for favorite boards for mining.

I think the hd7xxx cards are discontinued for the rx lines.

What you should do is read this: https://litecoin.info/Mining_hardware_comparison

For a baseline, whatever combination you go with, you want 2000 kh/s or more to actually generate enough profit to pay your electricity.

Most 7970s are WORSE than 7950s for mining.

If you buy the correct r9 280x, that looks like the way to go now if 7950s don't come back. Buying used 7950s is always a good deal too.

You want to worry about PSUs as well. the CPU should be an amd sempron or the cheapest athlon you can buy.

Looks goo assuming you mean to have 2 of the PSUs. You will need two of those for running 4 cards.

Don't forget those calculators are NOT accurate. They account for being in the BEST private mining pools and ZERO pool taxes.

Your actual return will be lower for sure.

Well, that is the risk taken. Also, you don't trade it for USD. You trade it for BTC which you then trade for USD. This can be done safely online with BTC-e.

You cannot just atke anything to the bank, because crypto currencies are not real. You can't print litecoins. They just exist as a list of numbers. Thats it. You trade it for BTC which is another list of numbers, then you trade that for USd on a market online like BTC-e.

Transfer to your bank account.

Only GPU matters. In the link you can see R9 series performance charts for LTC mining

http://www.youtube.com/watch?v=m_0j5ntR4w0

I think it's misleading to say you need a huge rig to recover the cost of electricity. Huge rigs use huge amounts of power: small rigs use proportionally less. Well, almost proportionally.

True, with more GPU's the power and capital "wasted" on the motherboard, CPU, boot device, and case is factored over a greater hash rate. However, modern cpu/MB's are pretty thrifty on power. Even my 8 core Xeon "power" workstation draws only 66W at the wall through a Bronze PS when it's not busy folding. I expect the rigs I'll be building next week to draw less than 30W w/o cards. Factored over the 400W I expect a pair of 270x cards to draw, that's not much of a factor. Further, I doubt California is alone in having a steeply teired pricing structure for electricity. For the first MW/hr or so each month, I pay .10-.15 per KW/hr. Over that amount. pricing skyrockets to .32 KW/hr. Though still (barely) profitable, it's a brick wall that _seriously_ impacts payback period. As long as I keep total draw around a KW, payback time is very short. Over that, payback time is greatly extended. Further, I've found that many of the capital costs don't scale very well either. Unless you're building open-air rigs, I think you'll find that shooting for 2 cards per rig is as economical as buying exotic MB's, cases, and PSU's to support 3 or more cards. I started mining on Jan. 1 2014 with parts on hand: a pair of 7790's. The combo cranks about 510 Kh/s. I've mined about $100 worth, and power was about $30. As it happens, my house uses electric heat anyway: $100 for Jan. Replacing the space heaters around the place on a watt-for-watt basis is a no-brainer; scaling much higher than that becomes a real head-scratcher.

Although I'm sure that the brick wall I mentioned is widely discussed, t I think it's very odd I have never seen it mentioned in the dozen or so intros to litecoin mining I've read. I've certainly never seen a profit calculator that takes it into account. It wouldn't be too hard to code such a calculator, though you'd need a copy of your rate tariff.

I'm pretty tickled with my small scale experiment and its benefits. For me, the next thing to analyze is how the power problem impacts the recovery time for solar power installation. It bugs me at some level that the "work" our miners do is, ultimately, to pollute our atmosphere with more CO2 and compound our world's energy problem. But if we can give some serious thought and a loud voice to the concept of solar altcoin mining, we could potentially turn this Very Bad Thing into a Very Good Thing. It's pretty cool my litecoin mining will pay for my heat this Winter (and then some). It would be doubly cool if it makes the very difficult decision as to whether or not to "go solar" a no-cost, no-brainer as well.