Author

Topic: Industrial Sized Bitcoin Mine Farm (Read 2618 times)

legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
February 24, 2015, 11:54:32 AM
#9
And responded. It was definitely sleeping time here in the US.
newbie
Activity: 25
Merit: 0
February 24, 2015, 04:01:09 AM
#8
Ive sent you a PM sidehack!

Thanks again and look forward to hearing from you
legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
February 24, 2015, 02:26:57 AM
#7
http://money.cnn.com/gallery/technology/2013/12/17/bitcoin-mine/4.html

That picture at least is ASICMiner Blades in immersion cooling, but I'd be surprised if they're still running them. I think that article is discussing an ASICMiner facility and it's certainly no longer current. AM Blades run about 8W/GH, something like 16 times the power of gear being made now.

The second article, there's more hardware in there than I can identify, but I see some older Avalons and such. I work mostly with power supplies and haven't really bought much mining hardware in the last year. I do recognize their crate of PSUs as the Z750P/N750P which my business works with quite a bit.

I'd say they have quite a bit more than 200 rigs, but they're not likely the same hashrate as 200 SP35. At 5.5TH per SP35 you'd be kicking 1.1PH, about .3% of the total network right now. That'd average about 11BTC per day, currently valued around US$2.6K in revenue daily if I did the math right - so at the current difficulty, you'd pay off one machine per day if your operating costs were zero. That's a 200-day breakeven if power is free and the difficulty is level and the coin price remains flat. Since none of those things are accurate you'll have to figure for yourself if that's a good plan.

Additionally, Spondoolies is out of stock on SP35 I believe. The SP31 is still shipping from stock, but I don't know how it compares. They're focusing right now on developing their new generation of hardware, which will be larger, more powerful, more efficient and undoubtedly much more expensive than anything in existence right now.

I'd take a look at what Bitmain announces after the holidays; if they move on an S6 it will be the most efficient miner on the market for a while and, if pricing is anything like the S5, not terribly expensive especially if you buy several hundred. Unless they pull something unexpected, however, it won't be as adjustable as a miner could be as far as undervolting for future efficiency gains, without additional hardware.

Also, when factoring difficulty changes into your revenue analysis, consider that within the next few months both Spondoolies and ASICMiner will be moving a lot of new gear that will be very high hashrate and very power-efficient.
newbie
Activity: 25
Merit: 0
February 24, 2015, 01:57:37 AM
#6
sidehack:

Yup, i sorta gathered the same with cloud mining.

The current advantage we have over "other" miners is that we consume a bucket load of electricity for our plastics manufacturing plant.

In terms of AMPS, we have incoming supply of around 6000 - 7000 which is enough to power many many neighbourhoods. As such, we get electricity with the following benefits

1. High Tension and Maximum Demand - 97% of our maximum load always available with SLA from the utility company.
2. Off peak rates - Electricity consumed during the hours of 8pm - 8am is 50% off the tariff
3. Industrial High Tension Tariff - On top of the 50% discount, we further get extremely good tariff rates, same rates offered to car manufacturers, aluminium extruders etc. Its as cheap as we can get and probably the cheapest in South East Asia, ontop of the 50% off peak discount.

Hence, after research, ive found that electricity cost is the "make it or break it" when it comes to mining for bitcoins. Everything else can be researched and mitigated such as choice of hardware, set up, efficiency etc.

Im trying to take advantage of our cheap electricity rates and the fact that we have acres of warehouse space to retrofit to house this project. Since the building belongs to us already, there is also no real estate cost.

Hence the only cost is the cost of equipment and setting it up (cables etc) and the cost of electricity monthly.


Does this sound like the operations can be viable?

With these guys here in these 2 links :

http://money.cnn.com/gallery/technology/2013/12/17/bitcoin-mine/1.html
http://www.thecoinsman.com/2014/08/bitcoin/inside-one-worlds-largest-bitcoin-mines/

Is it possible to tell what equipment these guys are running?

Based on my "eye-level" calculation, it would seem that they have approx 150 - 200 mining rigs. Am i wrong? If we are prepared to pay that kind of money, at approx USD2,235 per ST35 plus a further USD765 for set up per unit rounding it up at USD3,000 per miner and total investment of approximately USD450,000 - USD600,000, would we be a "sizeable" miner in comparison to those 2 in the articles?

I also never explained why our sudden interest to do this, my apologies.

At this moment, our packaging manufacturing business serves customers such as KFC, Starbucks, Airlines etc and it is a stable revenue. We generate approximately USD60 million - 75 million per annum although it has declined due to decline in our raw material prices. (Our profits remain the same, just revenue down due to raw material down).

We walk away with 32% - 33% gross and approximately 8% - 9% net profits. On top of that, we are exposed to currency risk substantially as we export a bulk of our products in USD.

Our profits are shrinking. Along with the general malaise with the economy globally, we see our profit margins shrinking with bad sentiment, bad government policies and a myriad of reasons. I wouldnt know where to begin.

Hence whilst we are still profitable, we would like to divert some of that profit into starting and innovating a new business. Of course, Bitcoin is by no means new. However if we consider the potential applications, like Bitcoin Atms that have sprout up, paying for coffee in Bitcoins, Digital wallets, this is just the start of a new frontier.

Smiley
legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
February 24, 2015, 01:28:19 AM
#5
Cloud mining is a big question mark these days. A lot of them appear to be scams, wherein you pay up-front for a certain amount of hashrate and they bank your money. Periodically they pay you back a portion, which diminishes over time as difficulty increases, but the rest is still there gathering interest. You'll probably never actually get back as much as you put in, and they start profiting immediately.

A legitimate hashing operation is leasing its hardware to cloud mining in order to offset both the odds of losing money and the time required to pay off the equipment. They buy the hardware and set it up, then you pay them to operate it, so a lot of their expenses, which normally would be paid off in installments over the next year or so, are reimbursed up-front by cloud customers. This puts the machine costs back in their pocket and the only recurring costs now are operating costs - the cost of power, internet and maintenance. It's not quite as much of a concern to them if the equipment doesn't break even, because it's already been "paid for" by cloud customers lacking discernment.

Starting a cloud hashing operation is pretty risky these days, because everyone is going to assume that you're a scam until you very definitively prove otherwise.
newbie
Activity: 25
Merit: 0
February 24, 2015, 01:17:45 AM
#4
Hello sidehack,

Thank you so much for the swift response! Much much appreciated!

I've also noticed that there's a lot of "cloud mining services" available now. However, if it were profitable, would the cloud miner provider not be better off mining himself with his equipment? Or is it more profitable to offer cloud mining services?

Thanks again!
legendary
Activity: 3374
Merit: 1859
Curmudgeonly hardware guy
February 24, 2015, 12:47:52 AM
#3
For SP20 calculations, don't use 1.7TH - it's not very stable at that point. A more likely stable (and more efficient) figure is probably 1.4TH I think. They're also currently out of stock, and Spondoolies is focusing on next-generation hardware and probably won't be building any more of them.

Bitmain's next hardware is probably not too far away, possibly around 3TH for under USD1000 and 0.4W/GH but right now everything is speculation.

Cost of electricity definitely has to be a factor in longevity calculations. The operating cost of the miner will slow down how fast the miner pays itself off and then begins to profit. A more power-efficient miner/PSU combination (measured in Watts per Gigahash, typically) will have a higher operating margin and be profitable longer.

Another thing to consider is undervoltability and underclockability. Some people are still running AntMiner S1 hardware which was built something like 15 months ago, because with some hardware hacks it can be taken from its stock configuration of 2W/GH down to around 0.8W/GH. The hashrate is substantially lower, but as you've mentioned, as long as the miner can turn out more BTC than it costs to run, it's still profitable. So miners with inherent undervolting and underclocking, which can be made to run at higher efficiency points, will be feasible longer.

For gear that doesn't have built-in power supplies, external PSU can be had for under $40/KW if you know where to look (I can source them ready-to-go for that in certain configurations).
newbie
Activity: 25
Merit: 0
February 24, 2015, 12:18:49 AM
#2
Sorry.. A quick addition.

Ive also come to realize that the information is somewhat "all over the place" so i became rather confused..   Huh

The king of miner's now it seems would be the

SP35 Yukon which operates at 6 T/H

So on that basis, i did the following calculations just on efficiency of capital expenditure.

1. SP35 Yukon

USD2,235/6 T/H = USD372.50 per T/H

2. SP20 Jackson

USD479/1.71 T/H = USD281.76 per T/H

3. Antminer S5 Batch 4

USD370/1.155 T/H = USD320.35 per T/H


So it would seem that based on this, the SP20 Jackson delivers the greatest bang for your buck? I understand that it does not come with a power supply but surely the power supply cant be substantial in price?

But yet the review's all state that the SP35 Yukon is a leader.

How many SP35 Yukon's would be a good sized operations? Would an approximate 100 - 150 SP35's generate a good profit?

Some advise on how to structure the set up, lay out and operations would also be much appreciated.


Also here in certain parts of South East Asia, our internet quality is rather spotty.

How much "internet power" do we need to have for how many rigs? What kind of upload and download speed do we need to have for how many rigs?

If we are unable to obtain high quality industrial/commercial internet, can we opt for "multiple home type" fibre internet and is there a way to "merge" them into 1 large and strong consistent internet supply?

So sorry for the myriad of questions. Its all coming slowly like a train through my brain right now.

Thanks again for reading and all the help in advance.
newbie
Activity: 25
Merit: 0
February 24, 2015, 12:10:19 AM
#1
Hello all,

Would like to ask for some advise and potentially offer a collaboration as well. The forum has been heaps informative and ive learnt alot here frm trudging and trawling the forum so a great big thank you in advance!

Abit of background about how much I/We know about Bitcoin. (I continuously refer to me and Us as we as a company intend to invest in building a Bitcoin Farm)

Sometime in 2012, i stumbled across "the Silk Road" forum and the whole TOR phenomenon. Got online, got on Silk Road and even attempted to order a thing or two. Paid with Bitcoin, which i bought off someone else at a rate i cant remember. I do know it was nowhere near what it is now.

I delved abit deeper on my free time about understanding more about Bitcoin and Bitcoin Mining. I attempted to "mine" bitcoins using a PC at home and was able to mine 1 coin in a month or so if im not mistaken. I may be wrong as its been a couple of years now so apologies in advance for any mistakes in chronology.

I was contributing to a pool called "slush's pool" at that time and with abit of research, it seems this pool is still around so i could "log in" through the Slush's pool website to see how much Bitcoin i had earned and i could cash out into my Bitcoin wallet.

Anyhow, due to the shallow nature of my mind, i assumed the currency wouldnt take off as at that time, its primary function was for the purchase of illicit goods. Or at least thats how my shallow mind made it out to be.

Now ive seen otherwise and it seems the future of this currency may be exponential.


After further research, it seems that now people use dedicated ASIC rigs to mine for Bitcoins that mining for Bitcoins has become an industrial sized operation occupying acres and acres of space.

This is where i have some questions and i was hoping someone could fill in the gaps and maybe we can take this further together as well..

1. Electricity Cost : Ive come to understand that the biggest hurdle to overcome is electricity cost in mining. Besides the real estate location being the physical location where you house the mine farm, the electricity cost is a large hurdle.

Is there any "other" substantial operating costs besides electricity?


2. There are some "LARGE LARGE INDUSTRIAL SIZED" Bitcoin mining operations ive read about online.

http://money.cnn.com/gallery/technology/2013/12/17/bitcoin-mine/1.html
http://www.thecoinsman.com/2014/08/bitcoin/inside-one-worlds-largest-bitcoin-mines/

These articles write about large industrial sized Bitcoin Mines.

In these mines, are the equipment used standard "buyable" equipment that i can find on the internet or are they custom made and custom designed?

Meaning, if i wanted to replicate what they've done, would i go out and buy 5,000 x Spondoolies SP35 Yukon or Antminer of some kind?

Or is the equipment they use custom designed and built for them alone and not bought off the shelf like the 2 above quoted items?


3. Obsolete rate of equipment

Looking at this site :

http://www.rigwarz.com/

It would seem that the ROI (return on investment) for the equipment is approximately 1 year (365 days). Im assuming that this takes into consideration an "average" power cost into the calculation to generate the balance of the profits? If so, what is the rate of electricity?

Also reading the above 2 articles, it would seem that the rate of improvement and upgrade for the efficiency and speed of these ASIC rings is substantially fast!

Would your rig be still relevant in 1 year, being your ROI? If it stops generating enough Bitcoin after 6 months due to the increase in difficulty of the algorithms, i would not even recover my investment, much less make a profit?

OR

Perhaps once a miner becomes 6 - 7 months old, due to the increase in difficulty of the algorithms, my rig will generate less return BUT i can continue to add more and more rigs incessantly and the old rigs doesnt necessary become obsolete. It can continue running, just less efficiently as new rigs. Am i correct?

It only becomes unfeasible when it consumes more money in electricity than it can produce in Bitcoins. Am i correct?

4. Mining Pools.

I generally understood that you had to join a "mining pool" as no 1 individual can mine successfully due to the difficulty. Would an industrial sized operation also need to join a mining pool? And if so, how do you identify the best pool? If your farm is substantially large, can you contact the pool owner/administrator and work out how to start a partnership?


The reason why we are asking is as follows.

We currently operate a "plastics manufacturing business" here in South East Asia. We operate approximately 250 plastics processing machines here over a factory/warehouse space of around 13 acres. Our products are items like microwaveable takeaway containers, meat trays, deli trays, foam insulated cups, foam trays, take away boxes, straws, drinking cups (starbucks frappucino, 7-11 Slurpee), party plates, party cups and many others. Our products are exported globally to many many countries.

Due to the heavy consumption of electricity of our plastics processing machineries, we have a LARGE supply of electricity and a very very attractive rate from the utility company as well alongside with "off-peak" rates for consumption after 8pm and weekends.

We have been interested in starting a large industrial scale Bitcoin mining operations in our existing factory areas. We have twin 11 KVA incoming high-tension supply (approximately 6,000 amps) and capacity to upgrade to higher amounts in our own substation on-site.

Hence, the same thing that has made our plastics manufacturing successful thus far, being low electricity cost, we hope will allow us to mine Bitcoin effectively as well.

Although, further research tells us that equipment that may take 1 year to recover the cost on, may become obsolete in 4 - 6 months. Further, the volatility of the rate of exchange to common world currencies (USD, EUR, JPY) makes it unfeasible.

We are still extremely interested and would like to invest a large sum to create a commercially viable operation in the long run.

As the technology evolves very quickly and we dont constantly monitor the technology, as for us it is a business rather than a passion, we would also like to know if there are any forum members that would like to join us as a consultant, imparting knowledge and just filling in a few blanks here and there. I think our general understanding is somewhat sound but alot of help is needed. Once we've cleared the preliminary "blanks" and gotten a full understanding, we are happy to offer the person a share/stake in the operations which means an included stake in the equipment/assets as well. Basically a full business partner and part owner of the entire operations.

I understand that we may not have a thorough understanding of the concept as a whole and our understanding may be rudimentary so please accept my apologies in advance if ive made any mistakes above either in concept or understanding. Looking forward to having more information and someone to work together with too!

Have a great day!
Jump to: