Officially no. But I dont mind people not believing me, it makes it more fun to say 'I told you so' later on 

They sell them together with their own 1500w power supply which is another $80 USD.

Kill-o-Watt is reporting about 1010 watts but as many of you know they are not the most accurate of devices.

Another thing to note about this miner. While they offer a 6 month warranty on the blades and a 1 year warranty on the controller as of right now the warranty is only valid in China. I have confirmed this direct with Ebang. As they are not officially selling outside of China the warranty does not cross the border.

While efficiency is the main factor in profitability, this is a false statement. While the Bitmain S9 is the leader in j/gh it is not the only viable solution. If you are a home miner trying to mine on overpriced power of course your options will be limited.

Other options include the Avalon 7 series as well as a new player on the market ebang who makes the ebit miner. Both are around .13-.14 which is similar to the new Bitmain T9 that is a more reliable replacement for the S9. Do your homework on each machine and you can figure out which one makes the most sense for you.

The point of the T9 is to trade a bit of efficiency for a more reliable machine.

As 16/14nm is pretty much the standard for the next few years it makes sense to buy a miner that is slightly less efficient but will run twice as long without problems. We all know the S9 has issues, and the release of this miner seems to support the fact that the s9 hardware is being run on the very edge of reliability to get the .10 spec, which in turn is causing a number of premature failures. It only takes one bad chip to take out an entire hashing board with the way the boards are designed.

In the ultra cynical world of bitcointalk I guess i better post some pictures before people start calling me a liar.

This is not vaporware as I just set 4 of them up last night. As they have only been officially selling in mainland China the software is in all chinese. All miners are slightly outperforming the stock 6.3t spec, sitting between 6.4 and 6.5t at 715 mhz. There is an option to turn them up to 730 and with that they will hit 6.6-6.7t but I didnt want to deviate from stock settings on the inital test phase.

Also, the production version of the hashing boards have 3x6 pin PCIE connectors not 2 per board as stated in the prerelease specs.

Edit: Using the unreliable Kill-o-Watt meter im seeing an efficiency of about .15 or so. Going to amp clamp it later and see if I can get slightly more accurate readings.

You can setup a Rasberry Pi to take care of that for you. You link the two pins that turn the system on to the RPis GPIO pins with breadboard connectors and you can set each bank of pins to turn individual machines off and on. You can manage 13 systems per RPi with the onboard header.  You can also buy GPIO expansion boards and have one RPi controlling a much larger number of systems.

Assuming this is a legit business with that many machines I would assume you are running a large scale antivirus solution with in house servers. The best way to proceed would be to add an exception to the anti virus software to ignore the specific exes that are tripping the antivirus software to begin with.

The solution to your problem has been posted by Hotmetal.

https://lbr.id.lv/6xgpu_mod/


^^^you need this in order to see more than 4 cards in Windows 7 and 10. (I have not tried 8 so i cannot speak to it.)

I have this exact same board and had to run this mod to make it work in windows. If you are mining ETH  I would personally recommend ethOS. It is a very simple linux setup that will pretty much just work out of the box.

Try reflashing the MM firmware with the device connected. The controllers are hard to brick and this will usually recover them if you cannot connect to the miner.

Need a bit more info and maybe a miner GUI screenshot. The light flashingred can mean a number of things that are causing issues. First I would check the fans and make sure they are both working. Then maybe try a factory reset on the controller.

You mention wireless network and a wifi router. You arent trying to run these on wifi are you?

The google machine or even reading other threads here will answer your question. Do a little research on your own before asking everyone to do it for you.

Uh oh...I dont see a section in the GUI for error codes anymore. Those were so helpful with the A6

Those are some really good temps considering the overclock. Have you checked power draw to see where the efficiency sits? The later batches are not as efficient (unmodded of course) as the original 54 chip boards were.

You all realize that as long as you arent running firmware from last year your board wont blow up if its powered and idle right? Running two power supplies is no big deal, i had to run a good 100 S7s with multiple power supplies and never had a problem. Its no different than the beagle crashing, the boards just sit there idle.

P.S. Dont listen to all the people that say upgrading firmware is dangerous. As long as you have half a brain and use the proper firmware for your system it will work just fine.

Try this command

lspci |grep VGA

If the system is physically detecting the second card it should show up there.

When you say large deployment what kind of numbers are you talking? I have found on these forums a 'large deployment' can mean only 20 miners.