I spliced something up like that to run my old R-Box.  So that will do I suppose, although hopefully they have something better than a dc plug.  I guess it all depends on price.

Depending on price, I'd buy 2 to 4 of them.  That is, if I can run them off a PSU, I don't want to deal with dodgy AC adapter power bricks.  Do you know how these things are powered?

I hope that wink means you are joking.  You have better odds gambling on the "BTC Lotto" by solo mining as many have suggested, than gambling on the odds of some sha256 altcoin going up to a profitable price.

I'm interested in 2 upgrade kits for the R-Box and I have a friend that might want 2 as well.  It won't let me vote in the poll because I don't have enough posts, I guess I should spam up the forum more.  And depending on price, I'd also be interested in 2 sticks from this thread: https://bitcointalk.org/index.php?topic=995675.0

I switched the power cables around from one board to the other and the same board is still dead no matter how the cords are arranged.

I have a dead blade on my Antminer S5, which has been running for 2 months then this suddenly happened.  It shows all -'s in the miner status, one blade is cold and there is no red led lit on the blade.  I checked all the connections but it is still doing the same.  Any ideas?

My guess would be some type of firewall issue.  However, the problem seems to have worked itself out while I was sleeping.  So hopefully we are back to the usual flawless performance we are used to. 

Poolside hashrates are showing ~250 GH/s on one pool, ~550 GH/s on the other and 0 GH/s on the 3rd.    That is averaged out over the last 6 hours, actual rates varied from 94 GH/s to 937 GH/s.  Averaging around 45,000 stale shares per hour as well.   Lee's site shows me going at 1135 GH/s.

I don't think it's the pool, I'm running an AntminerS5 at my house on the same pool and it's been going at 1.19 TH/s with no stale shares.

I'm having similar issues, low poolside hash rate, miner disappearing from pool and reconnecting and large percentage of stale shares.  This didn't start happening until after the data center maintenance.   I suspect that the Chengdu center has connectivity issues with the pools.  Are there some pools that the Chengdu center is better able to connect to than others?

The poolside is low as well.  The internet was my main suspect but the networking diagnostics all passed and my other miners are running fine over the same connection.

EDIT:  I tried flashing an update to the firmware and it is now working again as it should.  I hadn't changed the firmware since I bought it.

Thanks for the help.

I've rebooted several times.  I turned off the PSU (Corsair 750) and reinserted the 4 cable connectors.  Still is the same.  All my chips are showing o's as well.

I've had an AntminerS3 running flawlessly since they came out.  I've had some networking issues and my internet service has been down all weekend, so I turned the miners off.    Since I turned it back on, it now hashes at 100 GH instead of 440 GH, fairly consistently.  All my other miners, computers and servers are all working.   I ran the antminer diagnostics and everything seems connected.  I've switched pools, with the same issue.   The fans don't come on either, but this may be because the temperatures are not coming up enough, my chips are running at 30 and 32 C.

Any ideas?

OK, just woke up this morning and my AntminerU3 was dead and so was the power brick.  Turning it over, looks like the power brick melted.  I was running this thing underclocked at 0755 and it still melted.  What now?

Although I had problems with it going Zombie, the USB power setting mentioned in post #210 seems to have fixed it.

I've had mine running for 49 GH/s stable for the last 48 hours, with no ZOMBIE and zero hardware errors.  I'm happy to run it a bit undervolted as long as it consistently runs, which it does.   Both the power brick and device are at reasonable temperatures, so I don't have to worry about melting the power brick.

Here's what I'm using:


Changing the baud rate from 1 to .57 gave me 10 more GH/s at these same voltage and frequency settings, possibly fiddling with it more might get it even better.

OK, it settled down but not as I was hoping.  It did run in the high 50's for a few minutes then started to hash lower and lower and lower rates.  Still getting lots of hardware errors and the power brick is hotter than the device.  Since it is now hashing in the 30 GH/s range, I switched back the voltage & frequency to 0755 and 0782 before my power brick melts too.  Back to getting 40 GH/s with no HW errors.

OK this seems to work, I'm getting high 50's GH/s, close enough to 60.  But it does kick out a bunch of "AMU: Invalid noonce - HW error" messages, at about the same rate as it solves hashes.  At lower voltage and frequency settings it gives no hardware errors but lower hashrate.  I've only been running it for 5 mins at this new frequency, so maybe it will settle down.  Should I worry about these hardware error messages or is this normal?


I tried that last night before I went to bed and it was still running this morning so it appears the zombie problem is solved.  I think I may have this thing working more or less properly now as long as the HW error notices are not an issue.

Thanks I changed the setting and it is running (so far), I'll see in the morning if it is still running, if not I'm doing a fresh install on another PC.

I just have the one miner hooked up to the computer's USB port.  It has never lasted through the night, every morning it is no longer hashing.   Other than today, it's only went Zombie one other time, usually when "it's not hashing" it just sits there showing 120 MH/s or 250 MH/s on cgminer and zero hashrate at the pool.   

Between all the downtime and the low hashrate, I'm averaging 18.33 GH/s per day on this device.

I agree that it sounds like a software issue.  I think the driver and/or the cgminer fork need more work.

EDIT:  Just went ZOMBIE again, right after I made this post.  Restarted cgminer and it never found the device, restarted computer same deal.  Changed the Baud rate from .65 back to 1, ran cgminer, still nothing.  Pulled USB and Power Cord then plugged power then USB back in, ran cgminer again, now is hashing back away at 40 GH/s.   That lasted about 10 minutes.  For some reason, ever since I changed the settings to 0800 and 1306 and .65, it won't run for more than 10 minutes without going zombie, even though I changed the settings back to what it was sort of working on before.  I've just unplugged the miner for now.  I'm going to try and install it on one of my other computers tomorrow to see if that makes a difference before I give up on this device and Bitmain forever.

Using these settings raised my GH/s from 40 to 50 but now I am getting large number of Hardware Errors saying Invalid Nonce.  The vast majority of all frequency settings spit out this same error except for 0755 voltage 0782 freq and 0755 voltage and 0882 freq, but I only get 30 and 40 GH/s on these settings.

Also, I haven't been able to get it to run for more than 12 hours without going ZOMBIE or just sitting there hashing at 250 MH/s.

Any ideas before I just get rid of this thing?

EDIT:  Lasted for a whole 7 minutes at 50 GH/s, now is ZOMBIE.  Going back to the 0755 setting for a whopping 40 GH/s. 

Try your voltage setting at 0755 and find a frequency that it likes at that voltage.   This should improve your hashrate, or at least it did for me.