Pages:
Author

Topic: Bitmain AntMiner U1 Tips & Tricks - page 33. (Read 106810 times)

full member
Activity: 138
Merit: 102
January 07, 2014, 12:44:20 AM
^^ could be your anti-virus blocking the install - some packages see the zadig driver as a trojan.

Are you using Norton?
member
Activity: 66
Merit: 10
January 07, 2014, 12:28:12 AM
Well, I can't get zadig to replace the drivers from CP210X to WINUSB .  It keeps saying "driver installation failed".  I tried to manually install the driver, but that was way over my head.  So I fired up bfgminer, and got all(6) miners working, one at a time.  Speeds suck though, best I'm seeing is just under 1 ghs. 
Any advice on the zadig problem?  I'm using Windows 7 and the correct cgminer version.  I just can't switch to the right driver.

~D
member
Activity: 99
Merit: 10
January 07, 2014, 12:20:47 AM

so is there anyway to make the display bigger or does it stop after 0- 14


easy bud.   If your using Windows Vista/7

1.  right click on your short cut that you start cgminer with  and go into properties

2. click on  "Layout" tab
3.  in the "window size" area make the height to what ever you want, I use 70.

4. click apply

Next time you start up from that shortcut it will be able to display all of your miners.   Additionally you can also change the layout colors.  Different background color, text color ect.  from the "Colors" tab.  I change the colors for each different pool, that way I know which pool or miners each window is  when more then one instance is open.

If you are using a .bat file

1. Right click on the top bar of of your command window and click properties

next is the same as steps 2 and 3 above.

If your using windows xp you will have to right click on the top bar of the command window.   If you have more then one shortcut for  maybe altcoins or different pools,  you will need to do this for each shortcut.   This works for cgminer and bfgminer.

Hope this helps.
member
Activity: 90
Merit: 10
Untitled
January 06, 2014, 11:10:55 PM
TLDR Sorry if this was mentioned
I was messing around with a hub trying to figure out available watts, and noticed how finnicky cgminer is with these damn things, I could never get more than 16 running. And adding them too fast would cause a crash. Had the same problem with modminer quads. I found it alot easier to open each device in its own. here is the bat file that i use to open the prompts. Each one also will have a named window. Find the usb settings by doing cgminer.exe -n where  --usb bus:device
start cmd /k C:\besr\cgminer.exe --usb 3:50 --bmsc-options 115200:20 -o stratum.btcguild.com:3333 -u reganreckman_a2 -p 123 --bmsc-freq 0781
start cmd /k C:\besr\cgminer.exe --usb 3:51 --bmsc-options 115200:20 -o stratum.btcguild.com:3333 -u reganreckman_a2 -p 123 --bmsc-freq 0781
start cmd /k C:\besr\cgminer.exe --usb 3:52 --bmsc-options 115200:20 -o stratum.btcguild.com:3333 -u reganreckman_a2 -p 123 --bmsc-freq 0781
What operating system are you using?

Edit: assuming Windows with that command line. Just for kicks...What version?
hero member
Activity: 711
Merit: 500
January 06, 2014, 11:08:19 PM
TLDR Sorry if this was mentioned

I was messing around with a hub trying to figure out available watts, and noticed how finnicky cgminer is with these damn things, I could never get more than 16 running. And adding them too fast would cause a crash. Had the same problem with modminer quads. I found it alot easier to open each device in its own. here is the bat file that i use to open the prompts. Each one also will have a named window. Find the usb settings by doing cgminer.exe -n where  --usb bus:device



start cmd /k C:\besr\cgminer.exe --usb 3:50 --bmsc-options 115200:20 -o stratum.btcguild.com:3333 -u reganreckman_a2 -p 123 --bmsc-freq 0781
start cmd /k C:\besr\cgminer.exe --usb 3:51 --bmsc-options 115200:20 -o stratum.btcguild.com:3333 -u reganreckman_a2 -p 123 --bmsc-freq 0781
start cmd /k C:\besr\cgminer.exe --usb 3:52 --bmsc-options 115200:20 -o stratum.btcguild.com:3333 -u reganreckman_a2 -p 123 --bmsc-freq 0781
full member
Activity: 192
Merit: 100
January 06, 2014, 10:05:47 PM
Has anybody managed to get their RPI setup working with the Adafruit 16x2 LCD + Piminer python script?...

I've run a sample python script for the LCD and the display and all buttons are working perfectly (which is surprising considering I hadn't soldered anything for a decade before today!), but when I launch "python PiMiner.py &", the display comes on and just sticks on displaying the RPI's IP address... Occasionally after stopping and starting cgminer the display will change to "connecting to cgminer...", but it then just sticks on that indefinitely... None of the buttons respond when it's stuck on the IP address or "connecting" screens.

The default port is definitely listening;

root@RASPBIAN:~# netstat -l
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State
tcp        0      0 *:ssh                   *:*                     LISTEN
tcp        0      0 *:4028                  *:*                     LISTEN



The following lines are in my .conf file;
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4028",
"api-allow" : "0/0"
Did that ever work for you?  I "could" read that as saying "listen on port 4028 but don't allow anyone to talk to me".  My .conf has the following:

Code:
"api-listen" : true,
"api-mcast-port" : "4028",
"api-network" : true,
"api-port" : "4028",

I needed the "api-network" before mine would work.
full member
Activity: 192
Merit: 100
January 06, 2014, 09:57:56 PM
I figured out how to tell the difference between an erupter usb and an antminer u1 with a little help from the kindly folks over at #cgminer.

So, the good news is the branch at https://github.com/fractalbc/cgminer.git can run ant miner u1's, block erupter usb's and red/blue fury's all at the same time assuming you enable them at configure time.  I used

Code:
./configure --disable-opencl --enable-bmsc --enable-bitfury --enable-icarus

The bad news is the hash rate drops for all devices when more than one type of device is present.  I'll play around a bit more but figured I would share what I had so far.

legendary
Activity: 4326
Merit: 8950
'The right to privacy matters'
January 06, 2014, 05:12:30 PM
Check the screws on the heat sinks are tight.

One of my 10 was slow as shit (7-900MH), on careful examination the heat sink was about 0.1mm off the board.  The screws weren't tight.  Tightened up and it's running around 1.8GH no problems now.

All 10 are hashing fine in an Anker 10-port hub plugged in to a Celeron NUC.  Running the terrible cgminer on Windows 7.  What a faff getting them working, I have to start cgminer then plug each stick in individually, waiting until the previous stick starts hashing.  cgminer also seems to refill the queue slowly - the sticks stop hashing for a few seconds, then are given new work.  I don't know if cgminer just sucks, or the antminer fork is broken.  Sad



  a 3mm hex or a 3/32 hex will work.  you can start up with  up 3 sticks plugged in with windows 7.  of course I have 36 running on a 49 port hub so a head start of 3 is not much.

I think I may be able to do up to 7.

 I will f around with it some more.  all on windows 7 ultimate. with an asus maximus gene 5  mobo 8 gb ram   and I forgot what the cpu is an i5 sandy 3470s comes to mind.  also a ssd for the os drive . I  have to get back to the idea that 36 sticks clocked to 2gh are only using 115 watts.  for the older am sticks 36 would pull 135 watts
hero member
Activity: 1246
Merit: 501
January 06, 2014, 05:06:15 PM
Check the screws on the heat sinks are tight.

One of my 10 was slow as shit (7-900MH), on careful examination the heat sink was about 0.1mm off the board.  The screws weren't tight.  Tightened up and it's running around 1.8GH no problems now.

All 10 are hashing fine in an Anker 10-port hub plugged in to a Celeron NUC.  Running the terrible cgminer on Windows 7.  What a faff getting them working, I have to start cgminer then plug each stick in individually, waiting until the previous stick starts hashing.  cgminer also seems to refill the queue slowly - the sticks stop hashing for a few seconds, then are given new work.  I don't know if cgminer just sucks, or the antminer fork is broken.  Sad
legendary
Activity: 4326
Merit: 8950
'The right to privacy matters'
January 06, 2014, 03:58:57 PM
how is this for power used 36 sticks on a 49 port hub and 2 thermal take 120mm usb fans  clocked to 2gh total of 71.3gh at 115 watts

note I am using a sea sonic 760 watt plat psu.  kill a watt meter. at the psu plug.  at best it is 90 percent  efficient  so real watts are under 100.  pretty good for usb sticks.  


these sticks will run at a difficulty of 30000 mill and still make some money.  at 900 a btc and 18 cents a kwatt.
newbie
Activity: 42
Merit: 0
January 06, 2014, 02:00:13 PM
I don't get stuck there, but the changes Bitmain have made to CGminer break something that adafruits scripts reply on. Guess we will just have to wait until the drivers are merged back into the main Git of CGminer
That could be the api change they made.  They pissed kanoi off when they changed the hashrate from MHS to GHS.  I changed it back on my branch.  Try it and see if it works with your adafruits scripts.  It did get miner.php to work properly when there were both old and new miners in the config.

It built and ran perfectly for me just now (no more fiddling around with the driver! Yay!!!), but the Adafruit display is still stuck on the IP address unfortunately...
member
Activity: 90
Merit: 10
Untitled
January 06, 2014, 01:10:52 PM
I resurrected my >10 year old Pentium 4 and installed Ubuntu 12.04 Server on it.
Installed any dependencies needed, most of which are listed in the regular cgminer README document in the 'Basic *nix build instructions:' section.
Combined that with parts of iluvpcs instructions and made the modification to driver-bmsc.c as fractalbc outlined. (I did run cgminer without the update and cgminer did crash.)
My gratitude to the both of you.  Cool
I have a bunch of U1 running @ 2GH/s stable for a couple days now. Installing Ubuntu Server 12.04 and configuring with the updates fractalbc lists was actually less time consuming than messing around in Windows 7.

nwoolls is working on a bfgminer fork for U1 (https://bitcointalksearch.org/topic/m.4341064) and release support from bfgminer shouldn't be too far off but I haven't seen recent postings regarding that.

Hoping to see Windows support improve and also see the U1 play well with other miners (which I haven't seen so far)((but I could have missed it in the big wall of text)).




full member
Activity: 192
Merit: 100
January 06, 2014, 12:49:54 PM
I don't get stuck there, but the changes Bitmain have made to CGminer break something that adafruits scripts reply on. Guess we will just have to wait until the drivers are merged back into the main Git of CGminer
That could be the api change they made.  They pissed kanoi off when they changed the hashrate from MHS to GHS.  I changed it back on my branch.  Try it and see if it works with your adafruits scripts.  It did get miner.php to work properly when there were both old and new miners in the config.
newbie
Activity: 28
Merit: 0
January 06, 2014, 11:14:14 AM
Has anybody managed to get their RPI setup working with the Adafruit 16x2 LCD + Piminer python script?...

I've run a sample python script for the LCD and the display and all buttons are working perfectly (which is surprising considering I hadn't soldered anything for a decade before today!), but when I launch "python PiMiner.py &", the display comes on and just sticks on displaying the RPI's IP address... Occasionally after stopping and starting cgminer the display will change to "connecting to cgminer...", but it then just sticks on that indefinitely... None of the buttons respond when it's stuck on the IP address or "connecting" screens.

The default port is definitely listening;

root@RASPBIAN:~# netstat -l
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State
tcp        0      0 *:ssh                   *:*                     LISTEN
tcp        0      0 *:4028                  *:*                     LISTEN



The following lines are in my .conf file;
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4028",
"api-allow" : "0/0"


I don't get stuck there, but the changes Bitmain have made to CGminer break something that adafruits scripts reply on. Guess we will just have to wait until the drivers are merged back into the main Git of CGminer
legendary
Activity: 4326
Merit: 8950
'The right to privacy matters'
January 06, 2014, 10:52:38 AM
20 minutes and I have it running on my BAMT GPU linux box.  So I decided to try this on one of my BAMT devices with 4 GPU's running mining Scrypt (NO I'm not mining Scrypt with these, just using the same Linux device).

I logged in on my BAMT device and did this exactly:

git clone https://github.com/bitmaintech/cgminer
cd cgminer
sudo apt-get install libusb-1.0
sudo apt-get install libudev-dev
./autogen.sh
./configure --enable-bmsc
make -j 6

and then did (as in the example from the .PDF)
./cgminer --bmsc-options 115200:20 -o 50.31.149.57:3333 -u ktzhan_3 -p 123 --bmsc-freq 0781
Two things.

First of all, the "--bmsc-options 115200:20 " is MANDATORY.  I was not using it and having it on the command line made all the difference in the world.

Next, what processor / OS are you running?  i.e. does "uname -p" say  "i386" or "x86_64"?

The same exact sequence on an intel i3 running ubuntu 12.04 x86_64 gives a compiler warning that it will crash because of a memory overflow and really does.  Fixing the memcpy from trying to copy 5 bytes into a 4 byte uint32 stops the crash and lets the program run.  Your compiler / OS / processor may be more forgiving of the driver trying to stuff 5 bytes into a 4 bytes entity.

FWIW, what I did to make it work on an i3 running ubuntu 12.04 was

--- a/driver-bmsc.c
+++ b/driver-bmsc.c
@@ -1276,7 +1276,7 @@ static int64_t bmsc_scanwork(struct thr_info *thr)
                goto out;
        }

-       memcpy((char *)&nonce, nonce_bin, sizeof(nonce_bin));
+       memcpy((char *)&nonce, nonce_bin, sizeof(nonce));
        nonce = htobe32(nonce);
        curr_hw_errors = bmsc->hw_errors;
        submit_nonce(thr, work, nonce);

which is essentially to only copy 4 of the 5 bytes returned by the driver before the byte swap.  Only 4 bytes are used by the code so this is a safe change.

After all your fixes, only 1 out 5 miners is detected? :?





pull out 1 stick of the five while you are running hot. hold it in your hand count to 10 plug it in wait 30 seconds see what happens. 

if you get a amu 1 hot plug after doing this.

 repeat on the second stick. do all five sticks.

  the one that is running will turn zombie but should come back to life.  let us know if this works.  I can tell from your printout that the sticks are ready to work and want to be hot plugged into hashing.
newbie
Activity: 42
Merit: 0
January 06, 2014, 09:56:54 AM
Has anybody managed to get their RPI setup working with the Adafruit 16x2 LCD + Piminer python script?...

I've run a sample python script for the LCD and the display and all buttons are working perfectly (which is surprising considering I hadn't soldered anything for a decade before today!), but when I launch "python PiMiner.py &", the display comes on and just sticks on displaying the RPI's IP address... Occasionally after stopping and starting cgminer the display will change to "connecting to cgminer...", but it then just sticks on that indefinitely... None of the buttons respond when it's stuck on the IP address or "connecting" screens.

The default port is definitely listening;

root@RASPBIAN:~# netstat -l
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State
tcp        0      0 *:ssh                   *:*                     LISTEN
tcp        0      0 *:4028                  *:*                     LISTEN



The following lines are in my .conf file;
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4028",
"api-allow" : "0/0"
member
Activity: 107
Merit: 10
January 06, 2014, 09:21:18 AM
20 minutes and I have it running on my BAMT GPU linux box.  So I decided to try this on one of my BAMT devices with 4 GPU's running mining Scrypt (NO I'm not mining Scrypt with these, just using the same Linux device).

I logged in on my BAMT device and did this exactly:

git clone https://github.com/bitmaintech/cgminer
cd cgminer
sudo apt-get install libusb-1.0
sudo apt-get install libudev-dev
./autogen.sh
./configure --enable-bmsc
make -j 6

and then did (as in the example from the .PDF)
./cgminer --bmsc-options 115200:20 -o 50.31.149.57:3333 -u ktzhan_3 -p 123 --bmsc-freq 0781
Two things.

First of all, the "--bmsc-options 115200:20 " is MANDATORY.  I was not using it and having it on the command line made all the difference in the world.

Next, what processor / OS are you running?  i.e. does "uname -p" say  "i386" or "x86_64"?

The same exact sequence on an intel i3 running ubuntu 12.04 x86_64 gives a compiler warning that it will crash because of a memory overflow and really does.  Fixing the memcpy from trying to copy 5 bytes into a 4 byte uint32 stops the crash and lets the program run.  Your compiler / OS / processor may be more forgiving of the driver trying to stuff 5 bytes into a 4 bytes entity.

FWIW, what I did to make it work on an i3 running ubuntu 12.04 was

--- a/driver-bmsc.c
+++ b/driver-bmsc.c
@@ -1276,7 +1276,7 @@ static int64_t bmsc_scanwork(struct thr_info *thr)
                goto out;
        }

-       memcpy((char *)&nonce, nonce_bin, sizeof(nonce_bin));
+       memcpy((char *)&nonce, nonce_bin, sizeof(nonce));
        nonce = htobe32(nonce);
        curr_hw_errors = bmsc->hw_errors;
        submit_nonce(thr, work, nonce);

which is essentially to only copy 4 of the 5 bytes returned by the driver before the byte swap.  Only 4 bytes are used by the code so this is a safe change.

After all your fixes, only 1 out 5 miners is detected? :?




sr. member
Activity: 452
Merit: 250
January 06, 2014, 05:12:19 AM
Have had one for a little bit now, works good. is stable overclocked to 2gh, at 2.2 hardware errors become an issue. At 2.4 they escalate to rejects hw errors and intolerable mining. Hashing on preferred sha256 alts at 2gh happily, ordered 2 more of these. I think with a little hardware hacking they could go way faster. (cooled with a fan about 1/4 inch above chips).

Happy Hashing Smiley


 Grin
sr. member
Activity: 447
Merit: 250
January 06, 2014, 01:33:09 AM
quote author=klondike_bar link=topic=390929.msg4245839#msg4245839 date=1388533065]
has anyone taken their U-1 apart yet to see if a better thermal paste can be used to keep it cooler? I want to see some pics if so.

I haven't even got mine home yet but from looking at it the U1 uses a thermally conductive pad between the PCB and heat sink.

To increase cooling you would increase the surface area of the heat sink or increase air flow.

stick a little adhesive-backed heatsink to the U1. you can get them on ebay for pretty cheap, i just got some like this that you could fit 3 or 4 on a U1:
[/quote]

Hi, I have seen the bigger performing fury with a large sink behind the smooth metal side, first pic...this other guy on ebay sells little sinks but puts them on the other side on top of the chip itself, pic 2 and 3...do you know which is the proper way to put sinks on these? The fury it appears come from the factury with the sink on the smooth metal..is it safe to put the sink on top of the chip itself? What you recommend and could you tell me the name of the seller on ebay that sold you these. Thanks.




full member
Activity: 192
Merit: 100
January 06, 2014, 12:06:15 AM
@fasmax

Sorry, I'm not entirely clear on what I need to do with the driver file... I take it you mean driver-bmsc.c?
Do I need to just paste the code from post #37 into that, or does it need to go somewhere in particular... Do I need to recompile after doing that?...

Yes You need to change the driver-bmsc.c file.

Change this line
memcpy((char *)&nonce, nonce_bin, sizeof(nonce_bin));
to
memcpy((char *)&nonce, nonce_bin, sizeof(nonce));

Then build.

Finally got RaspPi and AntMinerU1 working. This saved me a headache. I had enough trouble with other things. The biggest was every time I ran cgminer -n the device would come up ERR-3. I am not running in root, so my user name had to be part of sudo and plugdev groups and then sudo cgminer -n. Works great on PiBang.
I got 1 issue left, it gets HW errors in cgminer; even in WIN7. Its a VERY low percentage. What is considered too many and I should be concerned?
You do not need to run cgminer as root (sudo ...).
You can run it as yourself as long as you a member of the plugdev group and have installed the "udev" rules as described in the documentation.
Pages:
Jump to: