Topic: [CLOSED] batch #43 .34, .39 btc per ASICMiner Cube 30-38Gh/s - page 7. (Read 24915 times)

Thanks Canary.  Cube is running good.

I left positive feedback for you.

I have 2 Cubes in hand.  Ready to ship out instantly.  PM me. first come first serve.
1 btc each.

SOLD

Thanks, after 5 days of messing around they finally did  

Dude, if you are still having this problem and you own a droid phone - turn off the wifi on the droid.  That was my problem as documented here.  Sounds Fing crazy but worked instantly for me!

set your computer to the same network address range like 192.168.1.99 then connect and change it to an ip address that is usable on your network range.

I assume your on windows:

first, figure out what your current network range is. open a dos prompt (start / type "cmd" press enter).
in black dos prompt window

"ipconfig /all"

look for line showing 192.168.x.x or 10.x.x.x that is your current ip range.
line after it should show 255.255.255.0
then default gateway 192.168.x.1 or 10.x.x.1

write down the last two lines as we will need them later.

lets assume it showed 192.168.0.105 (most routers default to 192.168.0.x range or 10.x.x.x range)

that means your router is handing out addresses above .100 and your network is 192.168.0.x (the x is for workstation)

find an empty address below 100 to use.

try "ping 192.168.0.50"

if you see "reply from " that address is in use, try another until you find one that does not receive "reply form". it may say "ping timed out" or "host not reachable".

now go to control panel / network sharing

click "change adapter settings"

in the listing select the adapter that is in use. (it will not be greyed out).

right-click, click "properties".

scroll down in the list to "internet protocol version 4"

click "properties"

in the top box, select "Use the following ip address" fill in 192.168.1.99"

for the subnet enter 255.255.255.0

then click "Ok" then "close"

now try the "192.168.1.254:8000"

set the ip address to the number you found open for example 192.168.0.50
set the subnet and gateway to the setting you wrote down from above.

update it.

now to return your computer back to its settings.

in the listing select the adapter that is in use. (it will not be greyed out).

right-click, click "properties".

scroll down in the list to "internet protocol version 4"

click "properties"

in the top box, select "Obtain an address automatically "

click "OK" then "Close"

you  should now be able to use the new address you assigned to the cube.

what is the PC IP address and netmask?  if your are say 192.168.0.3 and netmask 255.255.255.0 you will not see anything on 192.168.1.x.

open a CMD window and type ipconfig to get this info copy and paste it here if you need help.

Well I got it, powered it up, but when I go to http://192.168.1.254:8000/ nothing happens.

Guess I'm not smart enough to figure this out... lol

Like Operatr I got a good one. Did open it up first, no loose heatsinks all boards seated.

Due to my general noobishment I first got it going with miner-proxy to slush's poolbut now have it running on a RasPi with minepeon ~31GH/s on Low (need to upgrade PS to keep it running on High)

Your Hashrate (MH/s):
30,935.51
Your Bitcoin Return
Daily                   Weekly      
0.01712738   0.12022102

Got my cube today  

Hashing nicely at stock clocks 32.5 Ghs/s   All chips show "0".  I guess I got one of the good ones, thanks Canary!

Running it with a Silverstone 850w btw.

I tried an Antec 650w at first, the miner would turn on initially but then shut down. Switching to my better PSU fixed it.

Like others reported the cards were out of their grooves in the case, easy fix, just pop the back cover off and slide the whole thing out, align the cards into the grooves. I didn't have any loose heatsink screws. The outer casing was scratched in a few places, but nothing I'm too worried about.

*edit   holding 37.942 gh/s at OC, Slush pool and Minectrl show it higher at 41-43 gh/s

I had similar issues... initially I was attempting to run the proxy off of a rpi as I have had in the past, but proxy was not happy no matter how many I had running - and it was resulting in lower hash rates.

After going through numerous troubleshooting steps, I moved the proxy (temporarily) to a macbook pro, just to test, and it had full hash rate... I have ultimately moved them to another machine entirely which has higher processing & memory power, my feeling was that it was probably memory related.

The processor load wasn't great but didn't seem too extreme on the rpi based on the 'uptime' commands I had ran, it was running at about .50-.60 on avg if I am remembering correctly, with just the proxy running. But it could have been processing power issues as well or a combination of both.

I didn't troubleshoot much further once I got them stable on another machine.  

I'm going to keep an eye on this thread.  I've had the same problem at work with blades and cubes.  They go sick after a couple minutes and reset.  Brought the cubes home and they are rock solid, even though one has two boards with 8 chips each 'x' out.  I'm going to have my IT guy take a serious look at our rounter/network settings to make sure AV or filters aren't blocking things.

The GPU rig is win 7 and has virtually no CPU load, that's why I though it might work.

The other is my ASCI control computer is an Intel Atom, Win 7 connected to multiple hubs and running the proxy (2 blades and 1 cube) and BFGminer (36 BE's). It runs 10 - 20% CPU load.

Im all 10/100 and cheap switches and router, good cable connection...

Every time I try to "optimize" my stuff it tends to complicate things and I usually have to start over, trying to keep it as simple as I can has been my solution.
A big, loud, hot, expensive solution. On the bright side my natural gas bill is very low for this time of year.

Good luck my friend, may the IT /ASCI gods look kindly upon you.

What was the difference in PCs? OS?  CPU Load? 

I ask because I am trying to figure out how to get my cubes up and running at full speed.

Thanks

That's 750 each x 3 + 2 600 watt Thermatakes @ .13 cents a Kwh
That long at 90% load is impressive to me. These thermaltake's have surprised me, they have ran over a year and a half at 450 watts continuous each.



Those receiving cubes should take apart your cube and reassemble it carefully making sure all the screws are tight and the boards all line up within the grooves inside the cube.

I initially ran slush's proxy on a GPU mining rig but it only let half or less of the cube hash, I had lots of chips sit at 000. Moved proxy to another PC on same switch it works fine.

Everything is Ethernet. 

I think the problem is that the chips are running very low (if at all) here is what the screen looks like:

ASIC_01-16: O O O O O O O O O O O O O O O O
ASIC_17-32: O O O O O O O O O O O O O O O O
ASIC_33-48: O O O O O O O O O O O O O O O O
ASIC_49-64: O O O O O O O O O O O O O O O O
ASIC_65-80: O O O O O O O O O O O O O O O O
ASIC_81-96: O O O O O O O O O O O O O O O O

M_01-16: 117 000 000 058 000 000 058 058 000 000 117 000 000 000 000 058
M_17-32: 058 058 000 000 000 058 058 000 058 000 000 058 058 058 000 000
M_33-48: 000 000 000 000 000 000 000 000 117 058 000 000 000 000 000 000
M_49-64: 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000
M_65-80: 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000
M_81-96: 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000 000

Jobs:0000000169 Accepted:0000000019 Rejected:0000000062 (0:0) F1 F2 F3
MHS:01112 Utility:015 Efficieny:011.24%
-------

This is with a direct connection to bitminter.  Local proxy (bfgminer or slush's proxy) has problems maintaining connection.  It seems to reset or be seen as sick after a few minutes.  Even when it stays connected it will run less than 1 GHs on local proxy - but spike up to 6 GHs on direct connection to bit minter.

Check the voltage output from your PSU, it should be 12v.  If it's less it might be affecting the speed. 

Also, check to see what each chip is running at, anything abnormally low?  they are spec'd to run at 333 per chip on low/std speed.

everthing should also be connected via ethernet, no wireless.

Thanks for the suggestion.  I have tried multiple setups over the last 3 days and can't seem to get it working right.

Even running 1 proxy with 1 cube on the same switch doesn't yield 15 GHs.  Which given the advertised 30GHs shouldn't be too much to expect.  I have tried BFGMiner 32 bit (on win7 and linux) as a proxy and I also tried the mining proxy on windows.  I also tried a direct connection to mint.bitminter.com (which does have get work enabled). 

On the direct connection, I could understand the traffic limits but no on the local proxy.  I am on a 1 GB Ethernet switch - even if block erupter only has 10 or 100 MPS the host PC is 1 GB and should be able to saturate both cubes.

I even tried running the mining proxy with an account on slush's pool but no luck.

I've run into bandwidth issues running multiple cubes on the same proxy or router.  Try running a second proxy on another machine for the second cube.

How are these coming in from china? air i presume.

boat would be too slow for the rapid pace of btc. Maybe shipping is causing the issues as I can see them being packed like that as they would have to be tested ("QC mining ;-)  )

Got mine yesterday.  Thanks Canary!!

As others have posted, it's best not to just try and power these on.  I took mine apart as soon as I unpacked it.  Mine had 2 heatsink screws rolling around inside.  All the other heatsink screws were loose.  Every board was misaligned and out of the guide slots in the chassis which made the mainboard bowed.  Once I re-assembled it properly it fired up fine and is getting ~37.6 GH/s with bfgminer 3.6.

Mine works great so far but there is obviously no quality control on the manufacture of these cubes so double check everything before applying power to them.