Topic: I run a 35 GH/s mining operation, ask me anything! (Read 1964 times)

If you've already got a computer with an adequate power supply and a couple free pcie slots, go find a couple used 5870 cards and run them non-stop until ASICs flood onto the market. Otherwise, don't invest in a full GPU based solution at this point. You'll likely never fully recover your expenses on it from GPU mining alone if ASICs ship in October.

I will have free power from January to may, but I have never done any mining. Would you recommend jumping in the game at this point?

I felt the same but there's another way to look at it .. How would you feel if you invested thousands into a GPU/FPGA mining rig? Wouldn't you say "Damn, if only I built a small test rig to get a feel for it and waited for the ASICS to roll out, now I'm thousands in the hole with a few months to recoup those costs ". I spent 993$ AUD on two computers and Win 7.. I have sold parts I'm replacing for 300$.. I'm short one monitor and one keyboard. By November I should recover 320$.. If the ASICS start rolling out then, Thats not a bad loss and remember you need a computer to hook an ASIC to.. I'll have two at a cost of 353$ + monitor and keyboard.

The people who have preordered ASICS will make a profit.. But they are assuming some risk.. ButterflyLabs will probably bring out ASICS but they were smart enough to include a provision in their terms that there may be an extra 60 days before delivery.. so the time frame may be into January 2013, after the reduction to 25BTC per block. They will no doubtably make a fair bit of coin at the beginning, but those huge preorders will all be delivered pretty much at the same time.. So the wonderous profit to be had by ASICS won't be long as they all come online in a short space of time. And of course there is always the possibilty of technical glitchs (Though BFL has an excellent history)

You and I are in a position where we have little capital investment to recover, the ability to invest capital (Not that I've got any) in the next generation of hardware now , or to raise capital in preparation of investing in the next generation of hardware AFTER it has proven itself and the mining environment has stablised..

I'm taking the last option, it doesn't give me maximum profit, but it does give minimal risk.. and when all is said and done I've got two gaming boxes at 33% retail price (except for a monitor) and if it's time to quit trying to set up mining, I had a lot of stimulating fun messing about with bitcoin.

Don't be feeling too defeated. There's way more to Bitcoin than just the mining. And you never know, you may just have an idea for the next killer bitcoin app and make your fortune that way.

Hadn't even heard of bitcoins 3 days ago. Now I'm feeling rather defeated. Haven't even finished rebuilding the rig I was planning to mine with... Really wonder if I'll have enough BTC by end of the year to cash out and get BFL's or other FPGA board to tweak rig with.

When *IF* my ASIC gear shows up 

I mine in multiple pools. Solo mining would be slightly more profitable in the long run, but I like the consistency of pools and ability to remotely monitor what my machines are doing. If one goes offline, and occasionally they do, I can get alerts and all so I can get the issue resolved quickly. I like some of the features that pools provide, such as graphs, trends, etc. When my ASIC gear shows up, I'll probably just switch everything to solo mining.

Here you're assuming an order from BFL. Most companies don't have a 10 week lead time

Price will slowly make its way upward I think. Just remember that there are already over 9 million coins in existence. Even if coin production completely stopped in December, it still wouldn't budge the price too quickly. This is purely speculation of course. There is a possibility that bitcoin could become A) Easier for the common folk to use, and B) Easier for miners to transact into cash. If this happens, it's anybody's guess.

Personally, I plan to hold all the coins I don't spend on BFL ASIC SC's.

Miners are going to take a pretty heavy hit due to the difficulty increase. Some were talking about making money back in 6 months. I don't think we have 6 months before network power goes through the roof. If price doesn't keep up with difficulty (and it probably won't for quite a while), then it will actually cost you more money to mine with GPUs than it will make, assuming you pay for electricity.

For me, every piece of hardware I buy, I assume I will never resell, it will die, or become obsolete. Furthermore, I don't acquire hardware unless I think I can make my money back in 3-4 months time. I assume everything after December is going to make 0 for my GPU miners.

All that said, you can always mine at a loss if you think value will ultimately go up. Of course it would be more wise to just buy bitcoins if you think that is the case, but that's just not as fun.

Alright, you got me, I've only got two 5770s and...  

I am actually out of town right now so i am not able to get you full proof, but here's a good start. The first photo is a smaller lab that runs slightly less than 1/3 of the operation. The second photo is a new lab I recently built out which is roughly about the size of the first, but the photo doesn't show the current state. It's all I had access to at the moment. The third photo is actually my oldest and most powerful lab. It has the 208v bar visible on the floor. As with the power in the first photo, you can see digital readouts of how many amps are coming across the bar. What isn't visible in this room are a few cased machines running a few 6-series cards. Those were running off of 120V, so I had to place them closer to the outlets. Sorry for the low quality. I'll try and share more when I am back home. (I don't actually recommend mining with 6 series cards btw).

http://imgur.com/a/1QCeS/embed

Cheers!

Whooo, serious business !

Pictures or it didn't happen. 

I run 28 machines fully populated with a wide range of GPUs. Nearly all are 5-series GPUs. Each system runs on Windows. Systems are spread across 3 locations which all connect to the internet via wireless cellular modems. Most systems are running off of 208 volt power for higher power efficiency.

What is your mining configuration ?

Yes, I built a second rig recently. Only needed the GPUs, CPU, RAM, and Mobo - had all other parts on hand. Also have free electric. I'll turn a profit on that rig, but I'm certainly not planning on expanding any further with non-ASIC at this point.

Still we have several months until that (and with 50BTC reward). So I'm buying 3 video cards right now, because I'll get my money back in 5-6 months. I do not have to pay for computers or electricity though .

With that much mining power would it be better to solo mine or mine in a pool and which one do you do?

A bit of speculation, but I am curious what you think.
What will happen with the price when in December the bounty gets halfed?
Will you go with the market price or? What do you think what kind off affect this will have on the miners in general?

If you can get, or already have, an FPGA right now then you should have nothing to worry about until ASICs arrive. Even then, you can trade up to the ASIC thanks to BFL's generous trade-in program. The power consumption is pretty negligible compared to even the most efficient of GPU setups. Personally I wouldn't order an FPGA now due to the amount of time it will take to recover the investment, the amount of time it will take BFL to deliver, and I believe ASICs are going to render them insignificant in the end. If you already have a desktop computer you are better off buying a couple used 5-series GPUs at a good price and start mining immediately, so long as you have relatively cheap power and can deal with a little heat. Two used 5870s can produce the same hash rate as a BFL single at half the cost.

So if you have have an FPGA now, I'd get in on the trade-in list with BFL. If you don't, find yourself a couple used 5870s and stick them in an existing computer. At this point, building full computers doesn't make much sense when difficulty is outpacing the BTC price in the near future.

I think that FPGAs are still a good investment now, because they will remain profitable for most people indefinitely (assuming btc value stays up). There will be a significant profit margin change if BFL sends out ASICS, but I think its best to acquire some hardware now and go along for the ride. I know a lot of people are waiting for BFL to deliver, which is another option.