In addition to find a single power supply that will handle the load, you now have both S9's hostage to the failure of a single PSU until you replace it. I personally would rather have separate PSU's for the S9's for a variety of reasons.

You should visit and ask about ventilation and cooling, as well as how many miners they expect to host. Texas is known for getting hot in the summer, and while $500 is an excellent price with electricity, it won't do you any good if come June your miner gets "cooked" along with 200 others in a poorly ventilated building.

You'll also want to understand Internet access and controls. If there are 100 other "tenents", will they end up having access to your S9, both physical and on-line? What are their plans?

I would agree that this price sounds "almost too good to be true".

You'll want to take this thread to to one of the Alt-Coin sub-forums. GPU's haven't been viable for actual BTC mining in years.

Welcome. It never occurred to me that a crazy high BTC price would alter the dynamics of forum moderation, but it obviously has.

Best of luck in your new role.

Unless Amazon has some kind of "magic" hardware platform, this would be totally unworkable. They are offering fairly "generic" X86 processor cycles. They also don't expect that your usage will be 24x7 processor intensive, and will likely have a stipulation to that in the contract.

In any case, AWS processor cycles are completely outclassed for SHA-256 processing by any existing ASIC, by multiple orders of magnitude. General CPU processing hasn't been viable for BTC in well over 5 years.

Feel free to try it, but don't be surprised  if AWS cancels your contract with a week or two, and you haven't made $1 in BTC.

Cloud mining is an entirely different animal, and is best discussed in a different sub-forum.

One possible approach to debug this would be take the miner in question off the network and set up an isolated network that you can put it on. What I mean by that would be to have a separate router that isn't actually connected to the real Internet (and Wifi disabled). Power off the miner, disconnect it, and reconnect to the isolated router. power it on and let it reboot. Connect a PC or laptop to the isolated router using a cable as well. Once you are setup, you can then poke at and see how it shows up and if you can log into it. Yes it won't actually be hashing for you. You should be able to then get into and check the settings. This should simplify your efforts to figure out what works and what doesn't. It  should also rule out any possible IP address conflicts.

This approach assumes that your S9 is setup to use DHCP to acquire an IP address on your internal network.

You can also take a 2nd working S9 off and connect it to your isolated network.

Just a couple of ideas for your troubleshooting efforts.

I would encourage you to read the Sticky note at the top of this Sub-Forum. There you'll note that Bitcoins can no longer be mined efficiently using ANY GPU, You will be best served by looking to an Alt-Coin forum.

It's also quite possible that this thread will get moved by a moderator (e.g. -ck).

I would encourage you to read the Sticky note at the top of this Sub-Forum. There you'll note that Bitcoins can no longer be mined efficiently using ANY GPU, You will be best served by looking to an Alt-Coin forum.

It's also quite possible that this thread will get moved by a moderator (e.g. -ck).

This is true, as long as the price of your favorite coin doesn't drop because of all the confusion caused by Bitcoin forks and such. Bitcoin is only as valuable as people think it is. The price, or perceived value, of Bitcoin can drop way faster than mining difficulty.

I have never seen any legitimate person talk to me about a Bitcoin transaction using anything but wallet addresses (e.g. 1BURGERAXHH6Yi6LRybRJK7ybEm5m5HwTr). Nobody needs to know where your wallet is stored, or if it's a Web wallet like blockchain.info supports.

Also when you do the math, it just doesn't work out. If you provide .1 BTC, where is the supposed 1.7 BTC going to come from in a week? If this was offered to 1000 people, that would be almost exactly the entire BTC produced in one day of mining. Does that make sense to you? What about if it's offered to 10000 people. That would exceed the entire production for the week.

Could someone describe EXACTLY what an s10 is? I have seen nothing to suggest that an s10 even exists. To those folks that have been around for a while, Bitmain did have the s2 and s4, but never produced an s6 or s8. Way back then, the even numbered sx was a rack mountable miner as I recall. Bitmain seems to have completely abandoned the rack-miner form factor, and hence the s6 and s8 never came to be.

Clearly Bitmain can name their miners any way they wish, but it would really be attractive for folks not to "invent" a miner based on an N+1 naming strategy that hasn't been around for a while, N+1 wouldn't mean what you might think it means.

My simple view is that the usual tool I used (aka bitcoinwisdom.com) is pretty much useless these days. The graphs are usually horribly off, and I don't know of a good replacement. If you comments are base on bitcoinwisdom, then I'd attribute the weirdness to the tool. The continuing slosh back and forth between the mining of BTC and BCH (aka Bitcoin Cash) just makes things really unstable. The fact that BCH seems to be on a run in terms of USD price only makes things worse.

Just my simple $.02 (regualr old fiat, no crypto)

Up $500 in the morning, and then down $500 in the afternoon (US times). As usual, BTC never fails to deliver on volatility! 

I ran an Avalon 6 using an Iogear GWU624 (WiFi to Ethernet adapter) for well over a year. There is apparently a newer one on Amazon (aka the GWU627). There are almost certainly other vendors with similar devices.

If you are more inclined to tinker, you can also take something like Linksys WRT-54GL router and make into a WiFi to Ethernet bridge. I have also done that because I had an obsolete WRT-54GL router that was working just great when I upgraded to a newer device.

Lots of ways to skin this particular cat......

In addition to the entire discussion surrounding the process technology and ASIC fabrication, I'll add my $.02 on the drawing that suggests a PCIe card. The last time I saw anything like this was the final Butterfly Labs debacle. They ultimately did keep  their PCIe form factor, but it never went into a PC with PCI slots. The power and colling just rendered that idea absurd. I haven't seen anything to suggest that this would be any more practical, unless they decided to make a lower speed, lower power minor (say 100-150 watts). That still wouldn't be a particularly cost effective product, since a purpose built mining "box" doesn't have all the constraints of a PCIe card, nor it is chained to a to a larger PC with it's attendant costs and power consumption. All the current mining boxes pair up a low-power single board computer (e.g. Raspberry Pi) with purpose built hashing boards. All the power and colling is for the hashing boards, with no real concern for the "computer" part of it.

One other thing worth noting. While it's true that Intel, AMD, Nvidia and so forth produce 100+ Watt parts on the 14nm process, they have much bigger packages with thousands of pins and lot's bigger dies inside. Bitmain, and I expect all the others (e.g. Cannan et. al.) have gone with small packages as well. I expect that feeds into the power supply design, board layout, and cooling.

The last folks that I saw with a "large package" mining chip was Spondoolies. They were obviously willing to make the investment in supporting a larger chip. History might suggest that their choice was not a good one in terms of cost competitiveness, though they had reliable and configurable hardware.

Looks to me like Bitmain has decided the best way to deal with "variation" is to mix and match blades to get to a repeatable hashrate for the whole miner, without strictly consistent parts.

While it might be a nice theoretical exercise, ignoring electricity, cooling, and Internet costs for running a miner doesn't seem very realistic. An S9 running in a cool, low electricity cost environment remains economically productive far longer than an S9 in a hot, high cost environment. You also can't ignore the price of Bitcoin in you calculations. An S7 (for example) may not be able economical to run when BTC is priced at $1000, and it's just fine when BTC is priced at $5000.

If you want to ignore ALL costs and BTC price, then an S9 will produce something nearly forever. Reality however gets in the way for most folks long before forever.

Are you expecting to leave the doorway completely open? The air that is being exhausted at whatever rate, must come from somewhere, is that through an open doorway? Any noise generated inside the room will almost certainly "escape" into the area outside the doorway, or through the walls of the room. Is that OK for you?

Do you have it worked out to supply the roughly 6000W that you'll need to run 5 miners? I assume folks more knowledgeable on the Avalon will adjust that number up or down, so don't take the 6000W figure as gospel, just a starting point.

I only have experience with coinbase, so my experience is limited. Consider accordingly.

In my experience, the "price" that you see quoted is only an approximation of what coinbase will pay you for your 1 BTC. My thinking is that they operate like any customary currency exchange. There is a "spread" between what they will "pay" for a currency, and what they would require you to pay for the same currency. Hence if BTC is $4700, they might only give you $4630, or if you wanted to buy 1 BTC, they would charge $4775. These are made up figures, and I don't understand how the do the actual pricing. The only way I have found out is to actually start a transaction, and then cancel it at the end if I don't like the final result.

They also then charge a "fee" which looks to be related to the overall size of the USD involved. That is all disclosed before you agree to complete the transaction. It usually takes a few days for things to complete, although it might be possible to accelerate that if you pay more in fees. My first few exchanges were slower, than later ones. I attributed that to them wanting some experience with you at the outset.

You might well want to set yourself up with them to get a real feel about it, and concrete examples of costs.

With regards to your desire to "get rid of" the Trezor, I have a suggestion. Maybe you should have a difficulty based contest, but use the first two digits to the right of the decimal point for the next adjustment. I realize that it makes this more of "lottery" than a skill based adventure. There really wouldn't be any realtime activity beyond the periodic summary of "spot" taken stuff. It might still not get it "gone" the next go around. It would seem though that a Trezor will have a "shelf life" to some extent, won't it?

Just a random thought.

I agree that the whole thing with bitcoinwisdom is just plan "off". Can't tell if that's by design or defect. Maybe the mechanisms don't respond well to massive changes in actual hash rate (aka the "slosh" with BCC).