I wonder if perhaps the analogy of "building your own cell phone" might help with all the folks asking about how to build their own miner. After all, why can't they just order all the parts to assemble an Iphone?

Why are folks happy to buy a phone from Apple or Samsung, and yet feel they don't want to directly buy a miner from Bitmain or Caanan but would rather "build their own"?

Just wondering since I too grow tired of the repeating threads.....

In case you haven't already noticed, the price for an S9 will likely be closer to $3000 rather than the $1600 you included in you original calculations. With the current Bitcoin frenzy (see FOMO), this may not be the last price hike for the S9.

This will almost certainly get moved by a Moderator to the an Alt-Coin forum. Might want to read the "stick" posting at the top of this forum about it being Bitcoin only.

https://bitcointalk.org/index.php?topic=401536.0

When you actually watch the video, the claim is for .003 BTC/day. He stated that was $40-45 per day. Towards the end of the video he says that it will be $100 in electricity for the month. I have no actual idea what his electric rate, but that $100 will likely be a little short.

In actual fact, his profit is likely to be closer to $40/day, which is paying off the hardware costs. He is clearly making money at the current BTC price. He also mentioned something about his Antminer D3 experience. He should "break even" in the first half of 2018 barring a collapse in BTC price.

If by "VPS" you meant some kind of generic X86 system you were renting, then your problem had absolutely nothing to with the pool. In case it isn't obvious to you, mining with a CPU is no longer profitable in any scenario. It hasn't been since 2012 .

I would encourage you to look at the stick post of this sub forum as well as it's parent.

If I had dozens of S9's I would do pool mining, and would split up my miners rate across different pools.

This all a very interesting topic.

It of course completely ignores the most basic real constraint today: Lack of actual hardware to purchase.

At the present time, there are essentially zero miners to purchase today. All large scale manufacturers are productions constrained and when they produce a "batch" (i.e. a production run of maybe a few hundred), they sell out in a matter of hours. Miss the purchase window, and wait another month or more to try again.

With the current speculative frenzy on BTC price, folks buy when possible.

Google the acronym FOMO for the current situation. 

I used to like that calculator as well. Just make sure to enter ALL the values, since it stopped filling in useful values for difficulty and BTC price back in August. I think the underlying calculations are correct as long as you don't accept any of the default values

I don't know exactly what correction you are referring to. One week fluctuations in BTC price will have zero impact on mining behavior and miner purchases. If BTC price were fall down to $5000, and stay there for 4-6 months, you could expect a reduction in miner pricing.  Unless and until that happens, I wouldn't be expecting a price reduction in miner pricing. Just my opinion for which you paid 0.0000 BTC .

The current speculative frenzy surrounding BTC price doesn't suggest that will be anytime soon.

If you are interested in a  "low cost" solar powered Bitcoin miner, you might want to consider using a USB based Bitcoin ASIC miner. They won't usually pay for themselves in terms of actual monetary return, but are an interesting entry to Bitcoin mining. I have a used a Raspberry Pi and different generations of USB miners from 2013 until about 2015. If your actual interest/goal is the FPGA part of this, then ignore my suggestion. 

I agree with fanatic26. While it would be an amazing chart to construct, I don't know what you would do with the result. I think it's predictive value would be nearly zero. I say this since (IMHO), Bitcoin price is a huge factor in any such chart. At this time it seems to me that Bitcoin is viewed by the financial world as a "speculative asset". That means that financial folks will trade it, regardless of what it's original intent might have been (e.g. a cryptocuyrency). They buy if they think's going to go up, sell if they expect it to decline. Rinse and repeat. In that scenario, information from prior to 2017 would seem to be pretty much useless.

While bitcoinwisdom.com was useful at one time, it seems like it essentially broke back in August of 2017 (when BCH forked from BTC). I am sure thereare  present day "calculators" for projections. I look at:

https://pool.btc.com/tools  (Mining Profit Calculator)

I can't vouch for the accuracy of the BTC calculator, but it shouldn't be too difficult to check if you want. What it does is pretty algorithmic, with BTC price as a huge wild-card variable (15 fold change in 2017).

As for Bitmain -vs- Caanan pricing, I think the most recent sale of 741's on Blokforge to be instructive. They literally sold out in minutes. They had 100 units, and well over 1000 folks interested. They set a limit of 2 per order, and still disappointed a lot of folks.

With the current FOMO (Fear Of Missing Out) attitude surrounding Bitcoin, there is plenty of money to go around, and neither Bitmain nor Caanan has any reason to engage in a price war. Both will sell all they can produce while BTC price holds up, and they are production constrained (probably by ASIC fab).

While it is possible to put the SHA-256 algorithm into an FPGA, in general it isn't cost effective to do so if you are seeking a profit on your effort. Back in the 2012-2013 time frame, there were multiple FPGA efforts for Bitcoin mining. At the time they were a step forward, but have been completely replaced by ASIC in terms of profit seekers since the end of 2013.

While I think it's a dandy home project, you should plan on getting your satisfaction from making it work at all.

Can anyone get into the blokforge.com website today?I've been trying sporadically over the last 2+ hours and get no where (Connect Reset).  I was looking for some definitive news today, but am stymied by the website issue.

bitcoinwisdom used to be my "go to" website for information. It seems that ever since the BCH hard fork, bitcoinwisdom really doesn't know what's going on.

I personally consider the numbers there to be highly suspect, if not  wrong.

Here's a acronym that sums it up (I think): FOMO

Fear Of Missing Out

One other little aside on difficulty changes. As stated above, it's every 2016. If it takes 14 days exactly, there is no change in difficulty. If it take less than 14 days difficulty increase, relative to how much less than 14 days. The increase is larger if 2016 blocks are completed in 12 days than 13 days. Conversely, if it takes 16 days to complete 2016 blocks, it drops more than if it takes 15 days.

The net effect of this is: Difficulty rises faster than it falls.

Difficulty adjustments are well understood and well defined. It's unpredictable because folks adding hash rate and removing hash rate. There is some element of luck for individual block solve times, but over many thousands of blocks, that all tends to even out.

What is completely unpredictable (IMHO) is what the BTC speculator in New York is willing to pay for 1000 BTC in the hopes that it will go up. That is what moves prices and influences peoples psychology on mining.

It's important to understand that there is NO algorithmic relationship between BTC price and difficulty. IIn my experience it's a feedback loop driven by the people that own and operate the mining hardware. When BTC price is high, or growing, people that own and operate mining hardware will try and acquire more Bitcoins. They do this by purchasing additional hardware (if it's available), or continuing to operate hardware that isn't all that efficient (but is likely all paid for).

If BTC price were to fall to $5000, withing a month you would likely see a decrease in difficulty. The folks with the lowest electricity price and most efficient hardware will just keep running and hope for an increase in BTC price and/or a drop in difficulty. Miners with high operating costs (i.e. high electricity prices) will be the first turn off thei hardware if it becomes to costly to operate. They may well wait a while to see if the price recovers, and then turn things back on. It's all driven by the psychology of 1000's of individuals (or companies), and what their current investment is in mining hardware. The total hardware investment in BTC mining is now huge, and nobody will quickly discard that hardware until they are convinced it won't pay off sometime (or they are forced to by economic circumstances).

Right now, the prevailing psychology for BTC price is "up, up, and away".

Assume it's going consume 80 KWh (Killowatt Hours) for the 48 hours you use the S9. It should actually come in less than that. You need to find out what he'll charge you for that (i.e. what the electricity rate is). If it's isn't profitable, then the actual consumption won't matter. I assumed the S9 will use 1666 Watts at the wall. That should be high, but it isn't high by more than 10-15%.

Will he let you point the S9 at the pool of your choice for 48 hours?

What's odd to me is that the 761 looks to have PCIe power connectors on it. That doesn't make a lot of sense if the power supply is "integrated". Also I don't see anything in the specifications about AC voltage requirements (e.g. Hz or voltage).

Maybe the "cut and paste" from the 741 description/picture needs some adjustment?

Don't sweat the 1950W -vs- the 2250W of three 750W power supplies. The 750W (or 1200W) indicates it's ability to supply that much power, not that it actually consumes that much power. The  actual power consumption is driven by the S9 itself. You just need to make sure the actual is less than the ability of the supplies to deliver power. The setup from parallelminer looks quite reasonable. The 1200W supply they include would power 2 of the 3 hash boards, and the 750W supply would power the 3rd hash board and controller/fan card. As I read it, the PCIe cables are sturdy enough. I don't have any actual experience with their kit, but looks pretty well thought out for a setup for a single S9 at home with 120V power. The sequence would be start the 1200W supply first (with 2 hash boards), and then the 750W supply (with 1 hash board and controller). The 1200W supply will likely run hotter since it's handling more load than the 750W supply.

Are you comfortable with the noise and the heat produced by an S9 in your space? Think a 1400W blow dryer running at full  blast, 24 hours per day. Other folks liken the noise to a vacuum cleaner.