Couple questions....
1) Having trouble deciphering the new command line arguments to run the 8 boards with bfgminer, anyone know what they are/mean?
2) How do I get the latest firmware and correct drivers to run the new bfgminer? (on Win 7 64 bit .... I did look at the drillbit forums but they are not complete or I misunderstand a step or two in the process)
Thanks!
Topic: DRILLBIT SYSTEM Miners - Avalon range available - Now $104.99! - page 12. (Read 273558 times)
January 16, 2014, 02:06:58 AM
January 15, 2014, 10:25:45 PM
Awesome!!!!
I'll give it a spin shortly....
Thanks Luke-Jr
January 15, 2014, 10:24:59 PM
Did you try the USB 2 ports on you MB (the GA-Z77X-D3H has both USB3 and 2 ports)? What about an external USB2 hub?
I have a single 8board which runs stable for days on Win7/64 on an old laptop, through a PCIExpress USB2 expansion (also had it running for some time on a USB2 external hub). It uses cgminer 3.10 now, but worked with earlier drillbit forked cgminer as well. The only occasional transient USB hiccups (and brief 'timing out unresponsive ASIC x') occur when switching pools, otherwise it is 'rock solid', averaging about 22.5 GH/sec (also reported by the pool) on drillbit options int:53:2:950.
Is your internet connection stable? You may also trying different pools (tried Slush and BTC here and they both work fine).
Good luck
I have a single 8board which runs stable for days on Win7/64 on an old laptop, through a PCIExpress USB2 expansion (also had it running for some time on a USB2 external hub). It uses cgminer 3.10 now, but worked with earlier drillbit forked cgminer as well. The only occasional transient USB hiccups (and brief 'timing out unresponsive ASIC x') occur when switching pools, otherwise it is 'rock solid', averaging about 22.5 GH/sec (also reported by the pool) on drillbit options int:53:2:950.
Is your internet connection stable? You may also trying different pools (tried Slush and BTC here and they both work fine).
Good luck
Thanks, I had tried the USB2 and USB3/2 ports controlled by the Z77 and USB3/2 ports controlled by the VL800 separately. They do behave differently (one seems to loose connection with the boards and then re-finds them before dying and the other dies with timing out unresponsive ASIC errors) but both seemed equally flakey. This morning I tried your suggestion of switching to Slush's pool from bitminter (not expecting it to work tbh), but so far things have been stable
. I've had good experiences with bitfury hardware and bitminter before, so was a bit surprised by this, but there you go. Fingers crossed they make it through the night! If not I have an OK powered usb hub to try and failing that I guess I'll need a different host.Cheers
Keep trying until you find the 'optimum solution'
Cheers
January 15, 2014, 07:17:25 PM
Been waiting for this, Thanx 
January 15, 2014, 07:09:03 PM
Did you try the USB 2 ports on you MB (the GA-Z77X-D3H has both USB3 and 2 ports)? What about an external USB2 hub?
I have a single 8board which runs stable for days on Win7/64 on an old laptop, through a PCIExpress USB2 expansion (also had it running for some time on a USB2 external hub). It uses cgminer 3.10 now, but worked with earlier drillbit forked cgminer as well. The only occasional transient USB hiccups (and brief 'timing out unresponsive ASIC x') occur when switching pools, otherwise it is 'rock solid', averaging about 22.5 GH/sec (also reported by the pool) on drillbit options int:53:2:950.
Is your internet connection stable? You may also trying different pools (tried Slush and BTC here and they both work fine).
Good luck
I have a single 8board which runs stable for days on Win7/64 on an old laptop, through a PCIExpress USB2 expansion (also had it running for some time on a USB2 external hub). It uses cgminer 3.10 now, but worked with earlier drillbit forked cgminer as well. The only occasional transient USB hiccups (and brief 'timing out unresponsive ASIC x') occur when switching pools, otherwise it is 'rock solid', averaging about 22.5 GH/sec (also reported by the pool) on drillbit options int:53:2:950.
Is your internet connection stable? You may also trying different pools (tried Slush and BTC here and they both work fine).
Good luck
Thanks, I had tried the USB2 and USB3/2 ports controlled by the Z77 and USB3/2 ports controlled by the VL800 separately. They do behave differently (one seems to loose connection with the boards and then re-finds them before dying and the other dies with timing out unresponsive ASIC errors) but both seemed equally flakey. This morning I tried your suggestion of switching to Slush's pool from bitminter (not expecting it to work tbh), but so far things have been stable
. I've had good experiences with bitfury hardware and bitminter before, so was a bit surprised by this, but there you go. Fingers crossed they make it through the night! If not I have an OK powered usb hub to try and failing that I guess I'll need a different host.Cheers
January 15, 2014, 06:49:23 PM
yay... but I literally just compiled 3.9.0 like 2 hours before you posted it and have my drillbits running on bfgminer that way, any new additions in the last 2hrs from git? lol
I compiled it from git, since it wasn't "officially" updated to 3.10 yet.
January 15, 2014, 06:45:45 PM
yay... but I literally just compiled 3.9.0 like 2 hours before you posted it and have my drillbits running on bfgminer that way, any new additions in the last 2hrs from git? lol
January 15, 2014, 06:42:11 PM
yay... but I literally just compiled 3.9.0 like 2 hours before you posted it and have my drillbits running on bfgminer that way, any new additions in the last 2hrs from git? lol
January 15, 2014, 05:48:26 PM
January 15, 2014, 05:45:55 PM
Hi i have the same issues with the boards and have them running with cgminer controlled by cgwatcher
The setting i use is 49:2:850 with cgminer 3.10
However weird is that the older 3.9 version does allow a bit faster clocking 50:2:850 but you need to pull and push back the usb plug several times to get the board running without the unresposive asic messages.
It happens often that the miner crashes and you need to monitor several times to check if the boards do not pump out the endless unresponsive asic messages. If it does you are screwed again and need to start messing with the usb plugs to get it running again
this far i have not been able to get them to run faster because either the screen is full of unresponsive asic messages or cgminer crashes instant.
settings at 48:2:950 seems to be impossible to run
52:2:950 seems to be the best all round setting, it gives just over 20GH/s per board. Make sure you have the latest firmware.The setting i use is 49:2:850 with cgminer 3.10
However weird is that the older 3.9 version does allow a bit faster clocking 50:2:850 but you need to pull and push back the usb plug several times to get the board running without the unresposive asic messages.
It happens often that the miner crashes and you need to monitor several times to check if the boards do not pump out the endless unresponsive asic messages. If it does you are screwed again and need to start messing with the usb plugs to get it running again
this far i have not been able to get them to run faster because either the screen is full of unresponsive asic messages or cgminer crashes instant.
settings at 48:2:950 seems to be impossible to run
Cheers
January 15, 2014, 04:28:27 PM
Hi i have the same issues with the boards and have them running with cgminer controlled by cgwatcher
The setting i use is 49:2:850 with cgminer 3.10
However weird is that the older 3.9 version does allow a bit faster clocking 50:2:850 but you need to pull and push back the usb plug several times to get the board running without the unresposive asic messages.
It happens often that the miner crashes and you need to monitor several times to check if the boards do not pump out the endless unresponsive asic messages. If it does you are screwed again and need to start messing with the usb plugs to get it running again
this far i have not been able to get them to run faster because either the screen is full of unresponsive asic messages or cgminer crashes instant.
settings at 48:2:950 seems to be impossible to run
52:2:950 seems to be the best all round setting, it gives just over 20GH/s per board. Make sure you have the latest firmware.The setting i use is 49:2:850 with cgminer 3.10
However weird is that the older 3.9 version does allow a bit faster clocking 50:2:850 but you need to pull and push back the usb plug several times to get the board running without the unresposive asic messages.
It happens often that the miner crashes and you need to monitor several times to check if the boards do not pump out the endless unresponsive asic messages. If it does you are screwed again and need to start messing with the usb plugs to get it running again
this far i have not been able to get them to run faster because either the screen is full of unresponsive asic messages or cgminer crashes instant.
settings at 48:2:950 seems to be impossible to run
January 15, 2014, 04:17:21 PM
Hi i have the same issues with the boards and have them running with cgminer controlled by cgwatcher
The setting i use is 49:2:850 with cgminer 3.10
However weird is that the older 3.9 version does allow a bit faster clocking 50:2:850 but you need to pull and push back the usb plug several times to get the board running without the unresposive asic messages.
It happens often that the miner crashes and you need to monitor several times to check if the boards do not pump out the endless unresponsive asic messages. If it does you are screwed again and need to start messing with the usb plugs to get it running again
this far i have not been able to get them to run faster because either the screen is full of unresponsive asic messages or cgminer crashes instant.
settings at 48:2:950 seems to be impossible to run
Luke-jr already a big thanks.
It would be awesome to have them running with multiminer controlling and monitoring it with bfgminer
The setting i use is 49:2:850 with cgminer 3.10
However weird is that the older 3.9 version does allow a bit faster clocking 50:2:850 but you need to pull and push back the usb plug several times to get the board running without the unresposive asic messages.
It happens often that the miner crashes and you need to monitor several times to check if the boards do not pump out the endless unresponsive asic messages. If it does you are screwed again and need to start messing with the usb plugs to get it running again
this far i have not been able to get them to run faster because either the screen is full of unresponsive asic messages or cgminer crashes instant.
settings at 48:2:950 seems to be impossible to run
Luke-jr already a big thanks.
It would be awesome to have them running with multiminer controlling and monitoring it with bfgminer
January 15, 2014, 03:00:07 AM
Rather strangely my up to now unstoppable Drillbit thumb now only gives Timing out unresponsive ASIC 0
I am at a loss of what to do to troubleshoot this?
Has it somehow bitten the dust?
Any suggestions welcomed.
Cheers
Marc
i have a similar problem with my 8board, what worked to fix it for me is using cgwatcher and using its schedule i change the timing to an invalid timing (49:1:850) run for a minute and then back to proper timing (53:2:950) also check power, heat usb etc. I am at a loss of what to do to troubleshoot this?
Has it somehow bitten the dust?
Any suggestions welcomed.
Cheers
Marc
January 15, 2014, 02:28:37 AM
Hi All, I'm having serious problems getting three "eight" boards to run stable for any length of time. I've had one good run of 12 hours or so, but mostly cgminer crashes out after less than an hour with one of the following types of error (a couple of examples below). They only arrived from the UK fixer last week and I've never been able to run them stable.
Looks like I have USB problems, but I've tried switching the boards to different ports on the mainboard (switching between the different USB controllers available and removing other, potentially competing devices from that controller), tried using usb cables with better shielding etc. with no luck. The problems occur whether or not the boards are plugged into the mini-plane for power or if they are separate and powered through their molex connectors. The PSU is supplying 12.24 V to the boards under load. Temperature readings on the board are always < 30 oC, typically 25-27 oC (cooled using a powerful, if noisy, 120mm server fan). I've mostly used the stock settings (int:40:1:850), but have tried a range of clock settings (faster and slower) and some, generally unsuccessful, runs at 950 mV with a range of clock settings (mostly resulting in "Timing out unresponsive ASIC" errors). I've re-flashed the firmware to the newest version several times and am using cgminer-3.10.0 built from the github (but have tried the drillbit 3.8.4 version too). In desperation I've even tried mining with these things under windoze, but they seem more prone to "Timing out unresponsive ASIC" errors than in linux
and ultimately crash out with USB errors there too. When hashing, the boards perform quite respectably (two at around ~19 GH/s, ~3% HW errors; one at ~17 GH/s, >10% HW errors - possibly a dead/unhealthy chip?), but they are down so often as to be useless.
Anybody have any suggestions? These are definitely the most fussy hardware that I've come across yet, it would be great to get them fulfilling their potential, but at the moment I just feel like throwing them out of a window
.
Looks like I have USB problems, but I've tried switching the boards to different ports on the mainboard (switching between the different USB controllers available and removing other, potentially competing devices from that controller), tried using usb cables with better shielding etc. with no luck. The problems occur whether or not the boards are plugged into the mini-plane for power or if they are separate and powered through their molex connectors. The PSU is supplying 12.24 V to the boards under load. Temperature readings on the board are always < 30 oC, typically 25-27 oC (cooled using a powerful, if noisy, 120mm server fan). I've mostly used the stock settings (int:40:1:850), but have tried a range of clock settings (faster and slower) and some, generally unsuccessful, runs at 950 mV with a range of clock settings (mostly resulting in "Timing out unresponsive ASIC" errors). I've re-flashed the firmware to the newest version several times and am using cgminer-3.10.0 built from the github (but have tried the drillbit 3.8.4 version too). In desperation I've even tried mining with these things under windoze, but they seem more prone to "Timing out unresponsive ASIC" errors than in linux
and ultimately crash out with USB errors there too. When hashing, the boards perform quite respectably (two at around ~19 GH/s, ~3% HW errors; one at ~17 GH/s, >10% HW errors - possibly a dead/unhealthy chip?), but they are down so often as to be useless.Anybody have any suggestions? These are definitely the most fussy hardware that I've come across yet, it would be great to get them fulfilling their potential, but at the moment I just feel like throwing them out of a window
.Code:
Some system info:
GA-Z77X-D3H mainboard + i5-3570K
Ubuntu 12.04 LTS (up to date) (Have also tried in Win 8)
cgminer-3.10.0 (Have also tried Drillbit custom cgminer-3.8.4)
Corsair TX850 PSU
Did you try the USB 2 ports on you MB (the GA-Z77X-D3H has both USB3 and 2 ports)? What about an external USB2 hub?
I have a single 8board which runs stable for days on Win7/64 on an old laptop, through a PCIExpress USB2 expansion (also had it running for some time on a USB2 external hub). It uses cgminer 3.10 now, but worked with earlier drillbit forked cgminer as well. The only occasional transient USB hiccups (and brief 'timing out unresponsive ASIC x') occur when switching pools, otherwise it is 'rock solid', averaging about 22.5 GH/sec (also reported by the pool) on drillbit options int:53:2:950.
Is your internet connection stable? You may also trying different pools (tried Slush and BTC here and they both work fine).
Good luck
January 14, 2014, 07:41:15 PM
re: bfgminer
Still no Drillbit support...it's only been about 2 months now...
The code is in git, it'll be in 3.10. January 14, 2014, 07:39:21 PM
Hi All, I'm having serious problems getting three "eight" boards to run stable for any length of time. I've had one good run of 12 hours or so, but mostly cgminer crashes out after less than an hour with one of the following types of error (a couple of examples below). They only arrived from the UK fixer last week and I've never been able to run them stable.
Looks like I have USB problems, but I've tried switching the boards to different ports on the mainboard (switching between the different USB controllers available and removing other, potentially competing devices from that controller), tried using usb cables with better shielding etc. with no luck. The problems occur whether or not the boards are plugged into the mini-plane for power or if they are separate and powered through their molex connectors. The PSU is supplying 12.24 V to the boards under load. Temperature readings on the board are always < 30 oC, typically 25-27 oC (cooled using a powerful, if noisy, 120mm server fan). I've mostly used the stock settings (int:40:1:850), but have tried a range of clock settings (faster and slower) and some, generally unsuccessful, runs at 950 mV with a range of clock settings (mostly resulting in "Timing out unresponsive ASIC" errors). I've re-flashed the firmware to the newest version several times and am using cgminer-3.10.0 built from the github (but have tried the drillbit 3.8.4 version too). In desperation I've even tried mining with these things under windoze, but they seem more prone to "Timing out unresponsive ASIC" errors than in linux and ultimately crash out with USB errors there too. When hashing, the boards perform quite respectably (two at around ~19 GH/s, ~3% HW errors; one at ~17 GH/s, >10% HW errors - possibly a dead/unhealthy chip?), but they are down so often as to be useless.
Anybody have any suggestions? These are definitely the most fussy hardware that I've come across yet, it would be great to get them fulfilling their potential, but at the moment I just feel like throwing them out of a window.
Code:
DRB 2: Got unexpected response E to command W
DRB 2: Timing out unresponsive ASIC 0
DRB 2: Got unexpected response ^H to command W
DRB 2: Timing out unresponsive ASIC 2
DRB 2: Got unexpected response a to command W
DRB 2: Timing out unresponsive ASIC 3
DRB 2: Got unexpected response <90> to command W
DRB 2: Timing out unresponsive ASIC 4
DRB 2 BFRequestWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
DRB 2 attempted reset got err:(0) LIBUSB_SUCCESS
DRB 2: Got unexpected response ^A to command W
DRB 2: Timing out unresponsive ASIC 5
DRB 2: Got unexpected response
DRB 2: Timing out unresponsive ASIC 6
DRB 2: Got unexpected response
DRB 2: Timing out unresponsive ASIC 7
DRB 2 BFRequestWork usb write err:(-7) LIBUSB_ERROR_TIMEOUT
Accepted 0758f3ea Diff 35/32 DRB 1
DRB 2 attempted reset got err:(-5) LIBUSB_ERROR_NOT_FOUND
Code:
DRB 3: Config: int:40:1:850 Serial: 4e0c4700
Hotplug: Drillbit added DRB 3
DRB 4: Config: int:40:1:850 Serial: 401c6500
DRB 5: Config: int:40:1:850 Serial: 54384400
Hotplug: Drillbit added DRB 4
Hotplug: Drillbit added DRB 5
DRB 6: Config: int:40:1:850 Serial: 4e0c4700
Hotplug: Drillbit added DRB 6
DRB 7: Config: int:40:1:850 Serial: 54384400
DRB 8: Config: int:40:1:850 Serial: 401c6500
Hotplug: Drillbit added DRB 7
Hotplug: Drillbit added DRB 8
Pool 0 difficulty changed to 4
New block detected on network before longpoll
Work available from pools, resuming.
DRB 1 BFRequestWork usb write err:(-4) LIBUSB_ERROR_NO_DEVICE
Looks like I have USB problems, but I've tried switching the boards to different ports on the mainboard (switching between the different USB controllers available and removing other, potentially competing devices from that controller), tried using usb cables with better shielding etc. with no luck. The problems occur whether or not the boards are plugged into the mini-plane for power or if they are separate and powered through their molex connectors. The PSU is supplying 12.24 V to the boards under load. Temperature readings on the board are always < 30 oC, typically 25-27 oC (cooled using a powerful, if noisy, 120mm server fan). I've mostly used the stock settings (int:40:1:850), but have tried a range of clock settings (faster and slower) and some, generally unsuccessful, runs at 950 mV with a range of clock settings (mostly resulting in "Timing out unresponsive ASIC" errors). I've re-flashed the firmware to the newest version several times and am using cgminer-3.10.0 built from the github (but have tried the drillbit 3.8.4 version too). In desperation I've even tried mining with these things under windoze, but they seem more prone to "Timing out unresponsive ASIC" errors than in linux
and ultimately crash out with USB errors there too. When hashing, the boards perform quite respectably (two at around ~19 GH/s, ~3% HW errors; one at ~17 GH/s, >10% HW errors - possibly a dead/unhealthy chip?), but they are down so often as to be useless.Anybody have any suggestions? These are definitely the most fussy hardware that I've come across yet, it would be great to get them fulfilling their potential, but at the moment I just feel like throwing them out of a window
.Code:
Some system info:
GA-Z77X-D3H mainboard + i5-3570K
Ubuntu 12.04 LTS (up to date) (Have also tried in Win 8)
cgminer-3.10.0 (Have also tried Drillbit custom cgminer-3.8.4)
Corsair TX850 PSU
January 14, 2014, 06:09:05 PM
So far flashed the firmware with no problem and removed all other load from the hub (block erupted and fan)
But sadly no different yet
CGMiner next
But sadly no different yet
CGMiner next
January 14, 2014, 04:41:02 PM
Rather strangely my up to now unstoppable Drillbit thumb now only gives Timing out unresponsive ASIC 0
I am at a loss of what to do to troubleshoot this?
Has it somehow bitten the dust?
Any suggestions welcomed.
Cheers
Marc
My Thumb is going fine, still getting an average of 3.01 GH/sec.I am at a loss of what to do to troubleshoot this?
Has it somehow bitten the dust?
Any suggestions welcomed.
Cheers
Marc
You can try: Different hub/USB port, check the hub's power supply, try different voltage/clock settings, flash the latest firmware, try cgminer 3.10, check that your mining pc is ok.
Good luck
January 14, 2014, 02:43:25 PM
Rather strangely my up to now unstoppable Drillbit thumb now only gives Timing out unresponsive ASIC 0
I am at a loss of what to do to troubleshoot this?
Has it somehow bitten the dust?
Any suggestions welcomed.
Cheers
Marc
I am at a loss of what to do to troubleshoot this?
Has it somehow bitten the dust?
Any suggestions welcomed.
Cheers
Marc
January 14, 2014, 06:38:12 AM
Here's some quick documentation I wrote up from the cgminer driver code, in preparation of writing the BFGMiner driver:
Code:
L Identify*
I Get info
0x10 bytes response:
uint8: protocol version
protocol version 2 has an implicit external clock if chip count is 1
char[8]: product name
if "DRILLBIT", fill in "Thumb" or "Eight" appropriately
uint32le: serial number
uint8: chip count
uint16le: capability bitmask
0x0001 Temperature sensor
0x0002 External clock
R Reset*
C Configure
Request followed by 6 bytes:
1 byte: core voltage config
0x01 850mV
uint8: clock level
uint8: "clock_div2"
uint8: use external clock? 0 or 1
uint16le: external clock frequency
1 byte response: 'C'
T Get temperature
2 byte response:
uint16le: Temperature in decidegrees celcius
E Get completed work
Response:
uint32le: number of works finished (max 1024)
For each work, 68 bytes:
uint16le: chip id
uint8: number of nonces found (max 16)
uint8: non-zero if chip idle, otherwise 0 if busy
uint32le[16]: nonces
W Send work to chip
Request followed by 46 bytes:
uint16le: chip id
uint256le: midstate
uint8[12]: data tail
1 byte response: 'W'
I Get info
0x10 bytes response:
uint8: protocol version
protocol version 2 has an implicit external clock if chip count is 1
char[8]: product name
if "DRILLBIT", fill in "Thumb" or "Eight" appropriately
uint32le: serial number
uint8: chip count
uint16le: capability bitmask
0x0001 Temperature sensor
0x0002 External clock
R Reset*
C Configure
Request followed by 6 bytes:
1 byte: core voltage config
0x01 850mV
uint8: clock level
uint8: "clock_div2"
uint8: use external clock? 0 or 1
uint16le: external clock frequency
1 byte response: 'C'
T Get temperature
2 byte response:
uint16le: Temperature in decidegrees celcius
E Get completed work
Response:
uint32le: number of works finished (max 1024)
For each work, 68 bytes:
uint16le: chip id
uint8: number of nonces found (max 16)
uint8: non-zero if chip idle, otherwise 0 if busy
uint32le[16]: nonces
W Send work to chip
Request followed by 46 bytes:
uint16le: chip id
uint256le: midstate
uint8[12]: data tail
1 byte response: 'W'
Is bfgminer live with Drillbit support yet?
Cheers
Jump to: