Bitcoin Forum

Hi all,
I'm new to this world and would like to get an ASIC miner, just for fun: I don't need one to earn, but I'd like it to pay its current bills and, on long time, to pay back its price, maybe.

My choice is between two devices (of course suggestions are welcome):

- Dragon Miner T1 - 1TH/s and 1kW, low price, seems great but I'm still unsure about its overall quality.
- 2x Antminer S3 - 450GH/s and 350W -> 900GH/s and 700W in total. More GH/s per W, but price is slightly higher and the external power supply might be "unhandy"

What is your suggestion? Any experience with the Dragon T1? It seems like a Chinese miner (while I don't expect the Antminer to be USA-made :D).
Another question, what is the difference between the S3 and S3+? I might get the S3 or the S3+, the seller is not clear, is this a deal breaker?

Any help or suggestion is appreciated!
Thanks

There's no difference between a S3+ and a S3. Beside the first batch of S3. Some S3 and S3+ overclock better than other, either way its something you need to individually check with the seller before buying, its not specific to the + or non + version.

I would probably go with the S3's because they're more handy.

If this is your aim then the data you are looking at is missing the two of the most important parameters. The efficiency of the Miner in J/GH and your Electricity cost in Cents / KWh

You must use these to work out how much money you will be making at the current difficulty, and then factor in Difficulty increases and the all important Halving that will take place in the first half of 2016.

Someone will point you at an online calculator to help you in this.


Rich

 

the s-3's are better in the sense they are light and easy to move.  they run pretty quiet and there are low cost fans for it to be even quieter.

 if you are in the usa lots of good psu's are around. you will need cheap power for these units. if you want to earn a profit.

That is scaring me a bit. I rent a 1800GH/s miner for a day and earnt about 6€, so I guess that a 900GH/s machine would produce roughly 3€/day. Given the price for my current, I pay 1.1€/day to keep a 700W miner running, so I could get about 1.9€/day.
The halving means that those 3€ will became 1.5€? That might be pretty bad

Your electricity cost is way, way too high then. You'd have to go for something more efficient and even then, if you can't ROI before the halving, you probably wont at all. You could run the Math with a S5, see if you can get your return before then.

I might be able to lower the electricity cost, but I don't want to count on that. I'm working on the "worst case".
S5s are really harder to find.

Given that, even with the halving, I will still gain a minimum by the difference. How much will difficulty increases affect the revenue?
How are hardware prices related to halving? I know this is just the second one in history, but I guess that a mining machine, after the halving, will be much less valuable than now, isn't it?

The halving is kind of distant right now, but when its close, a ton of people are going to dump their hardware.

So whatever you get now, you need to plan to get your ROI before then.

And for difficulty. A 100% raise equal a 50% loss in BTC income. Effectively the same as a halving.

I would go with the Antminer setup, the S3's are built like tanks, and (if good ones) will overclock to 250mhz for roughly 1TH of performance.

I would go with an s3 also.  I have one and it really runs great, not to loud either.
I am not sure about 1TH performance but I would suggest you overclock it to a setting that gives you as less HW errors as possible and not to overclock it for the most TH that you can get, you will only compromise it's lifespan.  Every S3 has it's sweets pot.

He meant 250 freq gets 500gh

Op wants 2 thus 500 + 500 = 1000


@ op my worst s-3 was about 425 gh but I had many that could do 500gh

Oh my bad!
May I add that at the frequency 237.50 I am getting roughly 500ghs with HW errors in the range of 0.0120%/0.0110%
Again every s3 responds differently to overclocks.

I think you can also mod the s3 to reduce the voltage and frequency so that they are more efficient and last longer.

IF you can get your hands on some S3 ++ id buy them :)_ . notice i said S3 ++ not S3 +, i could have bought some over the summer but passed it up or saw it to late to buy at least two S3 ++, after, i noticed it said ++. now i can't get any in the US.

I think a few made it stateside.  But it was a very very few on S3 ++ your right on most not making it here to US was not offered on site to US.  With current electricity OP seems to have I would not count on S3 (or plus models) or dragon a1.  I think he will have to go newer if he want's it to pay for operation like he said.

Only thing on S3 dont get batch 1.  It was hit/miss we got refunds as some were a tad slower.  If you go buying and get a S3+ you know it's not a batch one with slower speed (or you can have owner show speed before buying on S3 regular to make sure it is up to speed)

700W = 16.8Kw a day

Paying 110 cents for that = 6.55 cents per Kwh. That is actually quite good for mining.

They were an interesting Miner but not sure that is worth trying to find one. Efficiency will be marginally better than an S3, but if you want string power then better to just go for the S5 with the better BM1384 ASIC, which I think the S3++ was possibly a trial for?

Rich

  and if he gets a newer s-5 he can under volt for better efficiency you have a long thread on that.

If I were to guess I would guess it was a trial for internal as much as anything.   It was not really widespread as I think just China's site had it.   So guessing they wanted all the efficiency they could get.

I would agree with there are a TON of S3/S3+ that causes lower price and much easier to find then S3++.  And going up to S5 is not a bad idea that phil mentioned.  All kinda depends on electricity  which his is decent.  With difficulty going up though might be worth looking into S5 or other miners around it.

with the current diff and the future one, i think not even 0.05 will save your ass in the future, the s7 is already doing only 0.03, when it was 0.04, lost 1/4

it's a race war between efficiency and diff, at this point only trading with your hash will save you

With the current difficulty, S7 is making profit, so is s5.

You can't make blanket statements like this there are far to many variables to be able to actually say this.  Look at electricity price it could make and break your comment.   

There is a point where only S7 would be profitable and S5 would not be if high electricity.   Some with high electricity can buy mine for a little bit then try to sell for a ROI.  It is kinda like gambling but there are some that still do it.

the s7 now, is losing half of its original production rate, 0.02 right now vs 0.04 at the beginning

the s5 is somehow trash right now if you don't have super very cheap electricity or better to have it for free at this point

I think S7 is the better of the two. However, when the 16 nm miner come out, it will be in toast as well.

That is the key on electricity costs.  Those with cheap/extremly cheap electricity are getting S5's at decent deals and can ROI possibly before having.   But takes the right cost, and price on miner.

S7 or Avalon 6 for most people currently.  But still some are set older gear that is good if price is right.

s7 was pointless after all, it did not last too much, because of the halving they were forced to go fast with another new asic(16nm)

those who were mining with s5, had not urgency to buy the s7, the best strategy would have to remain with s5 until the new 16 nm asic, and skip the s7 completely

Depends on your electricity price.  S7 is necessary for some operating conditions where electricity is higher.  In others S5's are great still.

So it's pretty individualized.  I would not say it served no purpose as I think it's efficiency did serve one but it is perhaps did the most for those with higher electricity prices.  As that is where it is harder and harder to mine.

With the current difficulty, if your electricity is $0.1, the mining revenue is double the electricity price. When the halving come, there will be no profit.

Having is still a LONG time away in mining terms though.  And we talk about having as some big thing... but if you look at data since January 13th ... we have almost haved just look at data:

Feb 19 2016   163,491,654,909   13.44%   1,170,318,852 GH/s
Feb 07 2016   144,116,447,847   20.06%   1,031,625,717 GH/s
Jan 26 2016   120,033,340,651   5.89%   859,232,121 GH/s
Jan 13 2016   113,354,299,801   9.12%   811,421,684 GH/s
total      48.51%   

So we make it sound like this big horrible thing but we did it in a month and a half in difficulty changes.