Bitmain AntMiner U1 Tips & Tricks

[philipma1957]

Its and Anker 10 port 3.0 hub so plently power, cant get one running on its own. I normally use a GUI based miner so am really struggling with things like cgminer

I have written my bat file ok etc, but when i plug them into the port only 1 of them is actually working, anyone got any ideas

 you got one to work?  you loaded the zadig driver ?  and you are hashing at ? 1.4 - 1.6

  if so  let  us know.

[brush242]

do you think the famous pencil mod used on redfury sticks  would work?

Anyone got a good link for the pencil trick (never heard of it before) on the BE USBs? I have a few of them left that I would like to experiment on. Thanks!

[twib2]

do you think the famous pencil mod used on redfury sticks  would work?

Anyone got a good link for the pencil trick (never heard of it before) on the BE USBs? I have a few of them left that I would like to experiment on. Thanks!

Pencil trick doesn't work on the Block Erupter USBs; they have a forced oscillator speed so you need to swap the oscillator AND change out a feedback resistor.  There's a thread where someone did do that on here somewhere and a guy was selling kits to get them up to almost double speed IIRC.

The pencil trick mentioned here was for the Blue/Red Fury USB miners based on the bitfury chip.  Just search "blue fury pencil mod" and you should find it.  The frequency of those is determined by the resistance at a key location which can be lessened by tracing a pencil around the resistor.  Fun stuff!

[brush242]

Its and Anker 10 port 3.0 hub so plently power, cant get one running on its own. I normally use a GUI based miner so am really struggling with things like cgminer

I have written my bat file ok etc, but when i plug them into the port only 1 of them is actually working, anyone got any ideas

Yes, this post: https://bitcointalk.org/index.php?topic=390929.msg4263091#msg4263091

You really do have to have the miner running, plug them in, take them out, and plug them in again. I had to do the whole thing twice.

For the larger crowd, what will the *new* version from Luke(?) change that will (I get the impression) alleviate some of these issues AND speed them up beyond 0981?

[brush242]

Pencil trick doesn't work on the Block Erupter USBs; they have a forced oscillator speed so you need to swap the oscillator AND change out a feedback resistor.  There's a thread where someone did do that on here somewhere and a guy was selling kits to get them up to almost double speed IIRC.

The pencil trick mentioned here was for the Blue/Red Fury USB miners based on the bitfury chip.  Just search "blue fury pencil mod" and you should find it.  The frequency of those is determined by the resistance at a key location which can be lessened by tracing a pencil around the resistor.  Fun stuff!

Lisa: "While you were gone, I got some really good advice from twib2!"

Homer: [awed] "Twib2...Is there anything he doesn't know?"

[twib2]

Pencil trick doesn't work on the Block Erupter USBs; they have a forced oscillator speed so you need to swap the oscillator AND change out a feedback resistor.  There's a thread where someone did do that on here somewhere and a guy was selling kits to get them up to almost double speed IIRC.

The pencil trick mentioned here was for the Blue/Red Fury USB miners based on the bitfury chip.  Just search "blue fury pencil mod" and you should find it.  The frequency of those is determined by the resistance at a key location which can be lessened by tracing a pencil around the resistor.  Fun stuff!

Lisa: "While you were gone, I got some really good advice from twib2!"

Homer: [awed] "Twib2...Is there anything he doesn't know?"

LOL, and I was just about to respond to your other post above about having to have cgminer running and plug in the next U1s.  Turns out you don't actually have to do that if you're running on linux, but everyone knew that and it would have just been me chiming in pro-linux drivel.  Good thing I didn't do that.

[colinfx]

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.

You are a GENIUS - thank you! I spent HOURS trying to get this to work! I made the change you suggested and BAM it worked straight away. Cannot thank you enough

[brush242]

LOL, and I was just about to respond to your other post above about having to have cgminer running and plug in the next U1s.  Turns out you don't actually have to do that if you're running on linux, but everyone knew that and it would have just been me chiming in pro-linux drivel.  Good thing I didn't do that.

Okay Mr. Pro-Linux Drivel:

For the larger crowd, what will the *new* version of cgminer from Luke(?) change that will (I get the impression) alleviate some of these issues AND speed them up beyond 0981?

I used to be Mr. Tech Genius a while back and installed Ubuntu on an old Windows desktop. After about a year with it, I realized that while it may be safer and faster, that I was effin around with it ALL OF THE TIME! Jeebus is it fiddly. I understand the arguments between Windows and *nix, but I got tired of fiddling. Terminal Emulator on my Android phone now is as close as I get to "su".

[chromosoma]

hi,
what is the  safe temperature range for antminer u1?
I overclocked it to 1.8 GHS  and it runs  at 55° C and HW=20% with passive cooling
At 2.0 GHS it runs at 65 °C, HW= 25%.

Is it ok, or too much?

I will organise some kind of active cooling system, but  just want to know, which temperature is safe.

[philipma1957]

hi,
what is the  safe temperature range for antminer u1?
I overclocked it to 1.8 GHS  and it runs  at 55° C and HW=20% with passive cooling
At 2.0 GHS it runs at 65 °C, HW= 25%.

Is it ok, or too much?

I will organise some kind of active cooling system, but  just want to know, which temperature is safe.

what do you measure to get the temp?  the chip the metal heat sink?  if noise does not matter use this fan


http://www.amazon.com/ARCTIC-USB-Powered-Portable-Cooling-Solution/dp/B003XN24GY/ref=sr_1_1?ie=UTF8&qid=1388812120&sr=8-1&keywords=arctic+breeze+usb+fan


if noise matters use this fan

http://www.amazon.com/Thermaltake-Mobile-External-Cooling-AF0007/dp/B002OJN250/ref=sr_1_2?ie=UTF8&qid=1388812155&sr=8-2&keywords=thermaltake+usb+fan

[chromosoma]

I measure the Temp on heat sink.
just want to know if it is safe to run  them on passive cooling till Monday.

[Powell]

Been following the thread (how I got my cgminer working with my U1's from iluvpcs), but has anyone had any luck getting them to work with BFGMiner on the Pi yet?

[Reebzy]

Hi all; I've been following this thread for a few days, initially pointing my efforts to getting my U1's working under OSX.

Firstly, I'll state that I'm running a VM of Win7 on OSX because I can't get the AntGen1/AnMiner repo to make on OSX. Therefore, I'm having some trouble with 2x U1 sticks under Win7.

What I'm experiencing on Win7, now that I have all drivers installed, is a rollercoaster ride of hashing.

Inserting the first U1, it hashes up to the desired speed (1.6GH/s+), and then stabilizes. Inserting the second U1, it too hashes up to the desired speed, and then stabilizes. It's stable for approximately 20 seconds. Then, the hashrate plummets slowly (down to >100MH/s) for each device. Wait another 20 seconds, and both will bounce back to the desired speed. Additionally, my HW error rate is around 7-10%, with the vast majority of the errors appears in the "plummet" phase.

I can't figure out what the heck is going on; if anyone has any advice I'd love to hear it.

[Powell]

I think there is an option in Fusion to dedicate all USB power to it not just when needed.  Unless you are running off the factory USB ports then there is the problem.  Little rant LOL...  MacBook Pro's non-retina share 2 HUBs even though you figured they would have more.  I could never get mine to use the USB to CAN to flash my brothers 997TT it was never happy about bandwith so I always had to use my OG OBD2 to USB cable on it to let it do K-line based flashes (slowwww).  My 996TT (when had stock ECU) took about 6 minutes and the 997TT took 30 minutes with that cable.  When I got the $450 cable I was so excited then the shocking truth hit after some research on the USB hubs built in.  Probably should have mentioned Fusion said FU from the get go, and my bootcamp drive with Win7 x64 was the biggest FU that got me looking into it.  Also the performance even on 16gig i7 Macbook Pro 17" running logging software for ProEFI, Syvecs, etc. was actually very poor compared to Bootcamp.

[threaded]

Hi all; I've been following this thread for a few days, initially pointing my efforts to getting my U1's working under OSX.

Firstly, I'll state that I'm running a VM of Win7 on OSX because I can't get the AntGen1/AnMiner repo to make on OSX. Therefore, I'm having some trouble with 2x U1 sticks under Win7.

What I'm experiencing on Win7, now that I have all drivers installed, is a rollercoaster ride of hashing.

Inserting the first U1, it hashes up to the desired speed (1.6GH/s+), and then stabilizes. Inserting the second U1, it too hashes up to the desired speed, and then stabilizes. It's stable for approximately 20 seconds. Then, the hashrate plummets slowly (down to >100MH/s) for each device. Wait another 20 seconds, and both will bounce back to the desired speed. Additionally, my HW error rate is around 7-10%, with the vast majority of the errors appears in the "plummet" phase.

I can't figure out what the heck is going on; if anyone has any advice I'd love to hear it.

Hi I'm running on OSX but my setup will be quite different as I've a couple of aftermarket USB 3.0 cards in it: an old one I can't remember the name of and a CalDigit.  I've tried running various miners off it.  Got a similar sounding USB polling weirdness messing with me.  To stamp on that I played with "USB Prober" in Developer/Applications/Utilities. (You need the developer tools installed)  I went to USB logger and hit start.  My USB polling problems went away.  I've not bothered digging further than that as the last few bitcoin-qt's were discouraging so've given up on the OSX version for now and went the VM route for that part.  USB logger also gives a nice log of what's plugging in and out.

[philipma1957]

Been following the thread (how I got my cgminer working with my U1's from iluvpcs), but has anyone had any luck getting them to work with BFGMiner on the Pi yet?

bfgminer is not ready to do the job.  luke has yet to update his software.  maybe  he will do a weekend update of his software.

[philipma1957]

I measure the Temp on heat sink.
just want to know if it is safe to run  them on passive cooling till Monday.

I would not overclock unless I had a fan.  but that is just me.

[chadgroover]

I don't suppose anybody who managed to get these working on a RPI fancies helping a simpleton out a little bit here?...

Running Raspbian (2013-09-25) with one of them connected to a powered hub at the moment. I've tried the following;

Code:

apt-get update
apt get install libusb-1.0 libtool pkg-config libcurl4-openssl-dev libncurses5-dev libudev-dev autoconf automake git -y
git clone https://github.com/bitmaintech/cgminer.git
mv cgminer cgminer_bitmaintech
cd cgminer_bitmaintech
./autogen.sh
./configure --enable-bmsc 
make

(I've also tried "make -j 6" as mentioned in an earlier post)

When I try running cgminer with the same command line parameters which work for me in Windows (see below), I get the error "Segmentation fault"

Code:

./cgminer --bmsc-options 115200:20 -o stratum+tcp://un.coinmine.pl:6200 -u ####.### -p ##### --bmsc-freq 0881

lsusb shows the following;
Bus 001 Device 002: ID 0424:9512 Standard Microsystems Corp.
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 009: ID 0835:8502 Action Star Enterprise Co., Ltd
Bus 001 Device 003: ID 0424:ec00 Standard Microsystems Corp.
Bus 001 Device 004: ID 7392:7711 Edimax Technology Co., Ltd EW-7711UTn nLite Wireless Adapter [Ralink RT2870]
Bus 001 Device 005: ID 0835:8500 Action Star Enterprise Co., Ltd
Bus 001 Device 006: ID 0835:8501 Action Star Enterprise Co., Ltd
Bus 001 Device 007: ID 10c4:ea60 Cygnal Integrated Products, Inc. CP210x UART Bridge / myAVR mySmartUSB light
Bus 001 Device 008: ID 0835:8500 Action Star Enterprise Co., Ltd



Any ideas where I'm going wrong?...

[fasmax]

@ chadgroover
Did you edit the driver file ?

[chadgroover]

@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?...