Bitmain AntMiner U1 Tips & Tricks

[chromosoma]

I have different experience. While I was  mining  with about 10-15% HW in pool, my pool  actually  did see only 90-85% (100%-HW)  of  the  hashrate  given in cgminer.

After I switched  to bfgminer, the pool showed me 1:1 Hashrate as it is in bfgminer software, which is equal to my expected hashrate.

[1715423697]

1715423697

[1715423697]

1715423697

[1715423697]

1715423697

[1715423697]

1715423697

[-ck]

I have different experience. While I was  mining  with about 10-15% HW in pool, my pool  actually  did see only 90-85% (100%-HW)  of  the  hashrate  given in cgminer.

I shall go back and audit the hashrate meter in the icarus driver then.

Correction, I'm getting kano to audit the hashrate meter for icarus.

He has now done so, and indeed the verdict is it was missing the code that correctly subtracted hw errors from the hashrate count. It has been fixed in git and will be in the next release (4.0.1). Thanks for the bug report.

[brush242]

Generating hardware errors will not harm it unless it overheats. Mine is actively cooled though.

Ah crap. I spent a WHOLE load of time fiddling with those things to make sure the HW errors were almost non-existent. But, that's good to know.

Now I can go back and fiddle again, but this time it will be right. And faster. I have to post my pics as well. VERY actively cooled. Heh.

[jbmac1967]

Im mining with 14 miners and around 10 of them go to the full 2ghs I have set the clock at but the others only manage around 0.3-0.5ghs

Anyone know why that happens

Im using latest BFG miner

[Sp00n1980]

Im mining with 14 miners and around 10 of them go to the full 2ghs I have set the clock at but the others only manage around 0.3-0.5ghs

Anyone know why that happens

Im using latest BFG miner

i have a similar thing with one off mine, it always starts out well then drops to almost nothing. I hate to say this but i am starting that my slow one be a duff one 

[brush242]

Im mining with 14 miners and around 10 of them go to the full 2ghs I have set the clock at but the others only manage around 0.3-0.5ghs

Anyone know why that happens

Im using latest BFG miner

I would bet they aren't getting enough power, but you haven't posted what your setup is like. All on one hub, I assume? Which hub?

[Jay_Pal]

He has now done so, and indeed the verdict is it was missing the code that correctly subtracted hw errors from the hashrate count. It has been fixed in git and will be in the next release (4.0.1). Thanks for the bug report.

ckolivas, perhaps one dumb question...
Does CGMiner reload the conf file each time I restart it by hitting "(S)ettings, (C)gminer restart"?

[jbmac1967]

Its a 20 Port Powered Hub from Cryptoware.co.uk

So it should have plenty of power

[Mudbankkeith]

Im mining with 14 miners and around 10 of them go to the full 2ghs I have set the clock at but the others only manage around 0.3-0.5ghs

Anyone know why that happens

Im using latest BFG miner

try them in groups of 7 miners, if each group work properly then your hub is low on power.

I have migrated from erupters to ants
While the hubs were happy with 10 erupters, they only like 8 ants and some ports fail completely.
I have now grouped my ants by the best speed each will hash.

I started with BFGminer and 0A01,(2100) the ants with large HW errors, were pulled and placed in another hub. (9 ants are now showing a combined average of 18.8Gh)
These were then re started in another session of BFG, this time at 0981(2000). (4 ants now showing a combined average of 8Gh)
Again I pulled the ants with HW errors, then placed them in the 3rd hub at 0901(1900). (these are now happy at 1.9Gh each)
The lower speed ants were initially showing only 0.5 to 1.0 Gh
By running them at a slightly lower clock speed, the actual hashing speed increased at each reduction level.

[jbmac1967]

Im mining with 14 miners and around 10 of them go to the full 2ghs I have set the clock at but the others only manage around 0.3-0.5ghs

Anyone know why that happens

Im using latest BFG miner

try them in groups of 7 miners, if each group work properly then your hub is low on power.

I have migrated from erupters to ants
While the hubs were happy with 10 erupters, they only like 8 ants and some ports fail completely.
I have now grouped my ants by the best speed each will hash.

I started with BFGminer and 0A01,(2100) the ants with large HW errors, were pulled and placed in another hub. (9 ants are now showing a combined average of 18.8Gh)
These were then re started in another session of BFG, this time at 0981(2000). (4 ants now showing a combined average of 8Gh)
Again I pulled the ants with HW errors, then placed them in the 3rd hub at 0901(1900). (these are now happy at 1.9Gh each)
The lower speed ants were initially showing only 0.5 to 1.0 Gh
By running them at a slightly lower clock speed, the actual hashing speed increased at each reduction level.

Will try running them at 0901 to see if that helps

[Proteu5]

Hey, I have a seemingly simple question...

On my desktop I can detect and run all of my Antminer (5) and Eruptors (4) on one single BFG terminal, or two if I overclock. I'm having the same issue as JBMac1967 with dropping preformance, but I suspect that it's caused by my Voltage and Amp of the USB Ports.

Any how, my real concern is my recent switch to an XP Laptop. I built a custom cooling housing and want to use it. My laptop can only detect 4 of my 5 AMU devices; if connected I can get two ICA Eruptors running as well. All devices show up in my manager.

Any ideas on how I can get atleast all 5 AMU's running on my Laptop?

[fractalbc]

Hey, I have a seemingly simple question...

On my desktop I can detect and run all of my Antminer (5) and Eruptors (4) on one single BFG terminal, or two if I overclock. I'm having the same issue as JBMac1967 with dropping preformance, but I suspect that it's caused by my Voltage and Amp of the USB Ports.

Any how, my real concern is my recent switch to an XP Laptop. I built a custom cooling housing and want to use it. My laptop can only detect 4 of my 5 AMU devices; if connected I can get two ICA Eruptors running as well. All devices show up in my manager.

Any ideas on how I can get atleast all 5 AMU's running on my Laptop?

You want a powered hub with a power supply capable of at least 3 amps at 5 volts to run 5 AMU's at stock frequencies.  You want a powered hub and a power supply capable of 5 amps at 5 volts to run 5 AMU's at a modest overclock.  Five AMU's and four ICA's will want a minimum of 5 amps at 5 volts to run stable at stock frequencies.  You may find it easier to run two x 7 port hubs and spread the devices evenly across the hubs.  I.E.  2 or 3 AMU's plus 2 ICA's on a dlink 7 port hub with the 3 amp power supply should work well at stock frequencies.

[-ck]

He has now done so, and indeed the verdict is it was missing the code that correctly subtracted hw errors from the hashrate count. It has been fixed in git and will be in the next release (4.0.1). Thanks for the bug report.

ckolivas, perhaps one dumb question...
Does CGMiner reload the conf file each time I restart it by hitting "(S)ettings, (C)gminer restart"?

Yes it does, but restart is not as reliable as quit and start again due to some devices sometimes not cleanly shutting down I'm afraid.

[Jay_Pal]

He has now done so, and indeed the verdict is it was missing the code that correctly subtracted hw errors from the hashrate count. It has been fixed in git and will be in the next release (4.0.1). Thanks for the bug report.

ckolivas, perhaps one dumb question...
Does CGMiner reload the conf file each time I restart it by hitting "(S)ettings, (C)gminer restart"?

Yes it does, but restart is not as reliable as quit and start again due to some devices sometimes not cleanly shutting down I'm afraid.

Yeah, unfortunately it isn't.
Well, back to old school loop: mine for some hours. Quit mining. Edit the conf (to give time to lower temperatures on the equipment and flush them completely) with the new freq. Restart mining.

Thank you ckolivas!
You're doing a great job!

[bens.btc]

Hi, Im running 14 Antminers, BFGminer, Linux. All run perfectly fine with < .10 HW errors. However I have two U1's that when plugged in just rack up the HW errors. Like a HW error every five seconds. I went one-by-one and found the problem ones, and they don't run now with the good ones. Im clocking the 14@ 2 gh/s. The two hubs that I have the U1's split between are both high quality and Im sure its not a hub power problem.

Is there anything I can do with these two error-prone U1's? Like, can I specify BFGminer to run AMU 14 & 15 at a clock speed lower than all of the others? Can I mount them to a larger heatsink and have the HW errors go away? Or are they just plain toasted and should go in the E-cycle bin?

Thanks

[Mudbankkeith]

Hi, Im running 14 Antminers, BFGminer, Linux. All run perfectly fine with < .10 HW errors. However I have two U1's that when plugged in just rack up the HW errors. Like a HW error every five seconds. I went one-by-one and found the problem ones, and they don't run now with the good ones. Im clocking the 14@ 2 gh/s. The two hubs that I have the U1's split between are both high quality and Im sure its not a hub power problem.

Is there anything I can do with these two error-prone U1's? Like, can I specify BFGminer to run AMU 14 & 15 at a clock speed lower than all of the others? Can I mount them to a larger heatsink and have the HW errors go away? Or are they just plain toasted and should go in the E-cycle bin?

Thanks

Try a second session of BFG with the clock speed one step lower, (not all ants are equal) eventually you will find the best speed for each device, (some of your better sticks may even perform at a higher speed)

[jaakkton]

I have problem my RPI wont find my Antminer U1 but when I but it at my computer it works about 2.0 Gh/s. I my RPI gone haywire and need to program again. Is the anybody else had a similar problem. RPI did work at yesterday.

[Mudbankkeith]

I have problem my RPI wont find my Antminer U1 but when I but it at my computer it works about 2.0 Gh/s. I my RPI gone haywire and need to program again. Is the anybody else had a similar problem. RPI did work at yesterday.

May need to re burn your SD card.
While you do, also make up a spare one for next time.
SD cards can/will eventually fail.

[-ck]

Posted cgminer 4.0.1 with fixes for these.

[kabopar]

Hi, Im running 14 Antminers, BFGminer, Linux. All run perfectly fine with < .10 HW errors. However I have two U1's that when plugged in just rack up the HW errors. Like a HW error every five seconds. I went one-by-one and found the problem ones, and they don't run now with the good ones. Im clocking the 14@ 2 gh/s. The two hubs that I have the U1's split between are both high quality and Im sure its not a hub power problem.

Is there anything I can do with these two error-prone U1's? Like, can I specify BFGminer to run AMU 14 & 15 at a clock speed lower than all of the others? Can I mount them to a larger heatsink and have the HW errors go away? Or are they just plain toasted and should go in the E-cycle bin?

Thanks

You may also want to check that the hex screws holding the heatsink plate are not loose on those.  Did you try to run the 2 'problem' ones alone on a hub?  Also running 14 here (but spread among 3 hubs).  1 of these has a slightly higher HW error rate (just under 1%), while all others have nil error, all running at about 2 GH/sec average.

Cheers