Making progress in building cgminer 4.7.0. Some errors left but getting there 
good luck!
Topic: Ultra Under-overclock image for A2 Innosilicon by Emdje - V5.0 - page 21. (Read 79786 times)
November 01, 2014, 03:12:05 AM
good luck!
October 31, 2014, 06:11:05 PM
Pff another error. Anyone any ideas??:
Code:
.......
make[1]: Entering directory '/home/pi/Downloads/cgminer-4.7.0-A2-A'
CC cgminer-cgminer.o
cgminer.c: In function ‘watchdog_thread’:
cgminer.c:8635:12: error: ‘struct cgpu_info’ has no member named ‘restart’
Makefile:876: recipe for target 'cgminer-cgminer.o' failed
make[1]: *** [cgminer-cgminer.o] Error 1
make[1]: Leaving directory '/home/pi/Downloads/cgminer-4.7.0-A2-A'
Makefile:1486: recipe for target 'install-recursive' failed
make: *** [install-recursive] Error 1
October 31, 2014, 05:43:42 AM
Making progress in building cgminer 4.7.0. Some errors left but getting there
October 30, 2014, 08:58:00 PM
I have a very crappy photo to go on from my phone, but after some enhancement with photoshop I think the component on the left is the following:
ARM STM32F103 vet6
Datasheet: http://www.espruino.com/datasheets/STM32F103xC.pdf
I think the component on the left is the following:ARM STM32F103 vet6
Datasheet: http://www.espruino.com/datasheets/STM32F103xC.pdf
You got it on the microcontroller.
Unfortunately I have no idea what that part that I think is a voltage regulator that is marked with '33Z' is.
October 24, 2014, 04:35:57 PM
Better images:
MCU:
Possible voltage regulators:
MCU:
Possible voltage regulators:
October 24, 2014, 03:49:23 AM
I have a very crappy photo to go on from my phone, but after some enhancement with photoshop I think the component on the left is the following:
ARM STM32F103 vet6
Datasheet: http://www.espruino.com/datasheets/STM32F103xC.pdf
I think the component on the left is the following:ARM STM32F103 vet6
Datasheet: http://www.espruino.com/datasheets/STM32F103xC.pdf
October 23, 2014, 08:38:53 PM
yep, setting initial difficulty appears to work for west/nicehash 
good I'm probably going to try editing the main page for more main pools, I'm going away for 2 1/2 months soon...
What I did is registering my miner on miningrigrentals and set up 5 pools there. You can put you miner on unavailable when you don't want to rent it out. Mine is available most of the time for a price I am willing to rent it for (higher than what I can mine), so when it gets rented it is profit. Then I set 1 or 2 pools on the 'main page' for when miningrigrentals goes down.
Would you liase with Michelle who does minera? I've got 38 undervolted gridseeds blades doing 20% more mh/J than A2s running on minera, less Mh, but higher profit .
I'll be happy to try out anything you two cook up. I'm currently running a 48 chip A2box, but can also test a 16 chip a2mini. I had 2 dead blades in the A2box so the mini donated its blades.
A2 probably getting undervolted soon.
I don't know Michelle (you got info?). And I am not sure if I am able to undervolt the A2's. There are some suggestions that it can be done throught the software, I haven't found out how yet. When it must be done by hardware I will not be of any help, because I have not much experience in that field. .I'll be happy to try out anything you two cook up. I'm currently running a 48 chip A2box, but can also test a 16 chip a2mini. I had 2 dead blades in the A2box so the mini donated its blades.
A2 probably getting undervolted soon.
Did you ever undervolt the A2? What Does it look like emdje's picture that I marked up?
If we can figure out how to adjust the voltage in software I want to try dropping the voltage while it's running. I was able to do that with an A1 board that never made it into production because of other issues (with the power supply), but it seems worth trying.
Right now I'm more interested in using my A2 mini as a space heater so hash/j is not very relevant (yet).
On the other hand, if anyone has a stack of innosilicon A2 minis (or even full A2 terminators) I might be interested in buying two or three to play with.
October 13, 2014, 04:03:06 PM
Seems there is no acknowledgment send back when the reset command is given. Because a reset of the chips is mandatory for the miner to start hashing (and not detected) it thinks there are no chips, thus does not start. Don't know yet how to fix it though 
October 13, 2014, 01:28:15 PM
October 13, 2014, 02:33:21 AM
October 11, 2014, 06:00:27 AM
I can't get the v2.1-MiniTerminator image working with my new A2 Mini Terminator. Here a debug from run.sh:
Code:
[2014-10-11 18:33:24] Started cgminer 3.9.0
[2014-10-11 18:33:24] Run Reset=1
[2014-10-11 18:33:24] ST MCU hardware reset start
[2014-10-11 18:33:28] SPI Speed 4000 kHz
[2014-10-11 18:33:28] ST MCU - Enable (Pre-header)
[2014-10-11 18:33:28] A1 = 1200,10
[2014-10-11 18:33:28] A1 PLL Clock = 1200MHz
[2014-10-11 18:33:28] A1 = 1200,10
[2014-10-11 18:33:28] A1 PLL Clock = 1200MHz
[2014-10-11 18:33:28] A1 = 1200,10
[2014-10-11 18:33:28] A1 PLL Clock = 1200MHz
[2014-10-11 18:33:28] A1 = 1200,10
[2014-10-11 18:33:28] A1 PLL Clock = 1200MHz
[2014-10-11 18:33:28] A1 = 1200,10
[2014-10-11 18:33:28] A1 PLL Clock = 1200MHz
[2014-10-11 18:33:28] A1 = 1200,10
[2014-10-11 18:33:28] A1 PLL Clock = 1200MHz
[2014-10-11 18:33:28] AUTO GPIO CS
[2014-10-11 18:33:28] SPI '/dev/spidev0.0': mode=1, bits=8, speed=4000000
[2014-10-11 18:33:29] Failure(cs0)(20): missing ACK for cmd 0x04
[2014-10-11 18:33:29] ACK(cs0) timeout:cmd_RESET_BCAST-1.5556s
[2014-10-11 18:33:29] SPI '/dev/spidev0.0': mode=1, bits=8, speed=4000000
[2014-10-11 18:33:29] SPI(cs1) no device
[2014-10-11 18:33:29] ACK(cs1) timeout:cmd_RESET_BCAST-0.2142s
[2014-10-11 18:33:29] SPI '/dev/spidev0.0': mode=1, bits=8, speed=4000000
[2014-10-11 18:33:29] SPI(cs2) no device
[2014-10-11 18:33:29] ACK(cs2) timeout:cmd_RESET_BCAST-0.2165s
[2014-10-11 18:33:29] SPI '/dev/spidev0.0': mode=1, bits=8, speed=4000000
[2014-10-11 18:33:29] SPI(cs3) no device
[2014-10-11 18:33:29] ACK(cs3) timeout:cmd_RESET_BCAST-0.2161s
[2014-10-11 18:33:29] SPI '/dev/spidev0.0': mode=1, bits=8, speed=4000000
[2014-10-11 18:33:30] Failure(cs4)(20): missing ACK for cmd 0x04
[2014-10-11 18:33:30] ACK(cs4) timeout:cmd_RESET_BCAST-1.3649s
[2014-10-11 18:33:30] SPI '/dev/spidev0.0': mode=1, bits=8, speed=4000000
[2014-10-11 18:33:30] SPI(cs5) no device
[2014-10-11 18:33:30] ACK(cs5) timeout:cmd_RESET_BCAST-0.2177s
[2014-10-11 18:33:30] No any A1 board
No any A1 board
: Inappropriate ioctl for device
October 09, 2014, 08:51:24 PM
Anyone tried the image on the new A2's?
nope - sold mine
October 09, 2014, 02:51:44 PM
Anyone tried the image on the new A2's?
September 29, 2014, 05:19:38 AM
yep, setting initial difficulty appears to work for west/nicehash
good I'm probably going to try editing the main page for more main pools, I'm going away for 2 1/2 months soon...
What I did is registering my miner on miningrigrentals and set up 5 pools there. You can put you miner on unavailable when you don't want to rent it out. Mine is available most of the time for a price I am willing to rent it for (higher than what I can mine), so when it gets rented it is profit. Then I set 1 or 2 pools on the 'main page' for when miningrigrentals goes down.
Would you liase with Michelle who does minera? I've got 38 undervolted gridseeds blades doing 20% more mh/J than A2s running on minera, less Mh, but higher profit .
I'll be happy to try out anything you two cook up. I'm currently running a 48 chip A2box, but can also test a 16 chip a2mini. I had 2 dead blades in the A2box so the mini donated its blades.
A2 probably getting undervolted soon.
I don't know Michelle (you got info?). And I am not sure if I am able to undervolt the A2's. There are some suggestions that it can be done throught the software, I haven't found out how yet. When it must be done by hardware I will not be of any help, because I have not much experience in that field. .I'll be happy to try out anything you two cook up. I'm currently running a 48 chip A2box, but can also test a 16 chip a2mini. I had 2 dead blades in the A2box so the mini donated its blades.
A2 probably getting undervolted soon.
September 28, 2014, 06:30:12 PM
yep, setting initial difficulty appears to work for west/nicehash
I'm probably going to try editing the main page for more main pools, I'm going away for 2 1/2 months soon...
I'm also going to try having some sort of reset going, probably a 4 hourly cron reboot. I'd prefer something that detects when that dodgy blade goes down and reboots/restarts then, but my coding skills are not great...
The really annoying thing is that the dodgy blade is the only one that runs significantly quicker than stock, the rest are 1160-1220, the dodgy one usually runs 1300, until it stops... It does that at stock speed too.
Would you liase with Michelle who does minera? I've got 38 undervolted gridseeds blades doing 20% more mh/J than A2s running on minera, less Mh, but higher profit.
I'll be happy to try out anything you two cook up. I'm currently running a 48 chip A2box, but can also test a 16 chip a2mini. I had 2 dead blades in the A2box so the mini donated its blades.
A2 probably getting undervolted soon.
I'm probably going to try editing the main page for more main pools, I'm going away for 2 1/2 months soon...
I'm also going to try having some sort of reset going, probably a 4 hourly cron reboot. I'd prefer something that detects when that dodgy blade goes down and reboots/restarts then, but my coding skills are not great...
The really annoying thing is that the dodgy blade is the only one that runs significantly quicker than stock, the rest are 1160-1220, the dodgy one usually runs 1300, until it stops... It does that at stock speed too.
Would you liase with Michelle who does minera? I've got 38 undervolted gridseeds blades doing 20% more mh/J than A2s running on minera, less Mh, but higher profit
.I'll be happy to try out anything you two cook up. I'm currently running a 48 chip A2box, but can also test a 16 chip a2mini. I had 2 dead blades in the A2box so the mini donated its blades.
A2 probably getting undervolted soon.
September 28, 2014, 08:27:43 AM
I found out why 2.0 was throwing errors and not mining:
I'm using nicehash as my main pool, for the password, I have d=4096;p=0.41 which sets the initial difficulty (essential) and the profit above which it will use that pool. The semicolon is breaking something.
The flaky blade just stops working without crashing the rest of the system. I'll try adding a reboot into cron
2.1 didn't like my old a2s
I'm using nicehash as my main pool, for the password, I have d=4096;p=0.41 which sets the initial difficulty (essential) and the profit above which it will use that pool. The semicolon is breaking something.
The flaky blade just stops working without crashing the rest of the system. I'll try adding a reboot into cron
2.1 didn't like my old a2s
No it didn't 'like' my old A2 as well. It worked and it hashed with the same speed according to miningrigrentals, but weird logs.
You can set your initial difficulty using the image as well (and you can go up to 4096), so there should be no need to put that in the password field.
September 28, 2014, 06:32:38 AM
Image is posted, see first post for the link.
And let me know if it works for the guys/girls with the A2MINI Terminator
And let me know if it works for the guys/girls with the A2MINI Terminator
v2.0 worked fine on my 16 chip mini.
it worked for a day on a 48 chip a2box, then barfed with:
[2014-08-29 03:40:46] Started cgminer 3.9.0 [2014-08-29 03:40:46] Run Reset=0 [2014-08-29 03:40:46] SPI Speed 4000 kHz [2014-08-29 03:40:46] ST MCU - Disable [2014-08-29 03:40:46] GPIO CS is OFF [2014-08-29 03:40:47] Failure(cs0)(20): missing ACK for cmd 0x04 [2014-08-29 03:40:47] ACK(cs0) timeout:cmd_RESET_BCAST-1.1678s [2014-08-29 03:40:47] No any A1 board
I'd really like a working version for the 48 chip a2box as it has several blades with very high hw errors that get reasonable if the frequency is dropped a little, 1180 and it's a helluva lot better, though down to 14mh per blade at the pool.
that box has a flaky blade which could be messing with your image, but it doesn't crash the scripta one.
Minereu scripta seems able to cope with it, the dodgy blade sometimes goes down, but the rest keep on hashing.
reflashing didn't help.
I'll try your new version.
I found out why 2.0 was throwing errors and not mining:
I'm using nicehash as my main pool, for the password, I have d=4096;p=0.41 which sets the initial difficulty (essential) and the profit above which it will use that pool. The semicolon is breaking something.
The flaky blade just stops working without crashing the rest of the system. I'll try adding a reboot into cron
2.1 didn't like my old a2s
September 27, 2014, 06:36:55 PM
Tat is weird because in the image the reset is turned on as well as the mcu and the gpio. But your log states different. Maybe tge new image works. You might try the original one as well. The only other explanation could be the amount of hardware errors or non working cores. This might trigger a turnoff. I have to look it up when this is triggered.
Let me know if a different version works.
Let me know if a different version works.
September 27, 2014, 12:44:35 PM
Image is posted, see first post for the link.
And let me know if it works for the guys/girls with the A2MINI Terminator
And let me know if it works for the guys/girls with the A2MINI Terminator
v2.0 worked fine on my 16 chip mini.
it worked for a day on a 48 chip a2box, then barfed with:
[2014-08-29 03:40:46] Started cgminer 3.9.0 [2014-08-29 03:40:46] Run Reset=0 [2014-08-29 03:40:46] SPI Speed 4000 kHz [2014-08-29 03:40:46] ST MCU - Disable [2014-08-29 03:40:46] GPIO CS is OFF [2014-08-29 03:40:47] Failure(cs0)(20): missing ACK for cmd 0x04 [2014-08-29 03:40:47] ACK(cs0) timeout:cmd_RESET_BCAST-1.1678s [2014-08-29 03:40:47] No any A1 board
I'd really like a working version for the 48 chip a2box as it has several blades with very high hw errors that get reasonable if the frequency is dropped a little, 1180 and it's a helluva lot better, though down to 14mh per blade at the pool.
that box has a flaky blade which could be messing with your image, but it doesn't crash the scripta one.
Minereu scripta seems able to cope with it, the dodgy blade sometimes goes down, but the rest keep on hashing.
reflashing didn't help.
I'll try your new version.
September 27, 2014, 12:35:49 PM
where does it say you can play with pin settings?
Page 25 - Table 6 Pin description. But who knows what does it mean... 0.93V is the target.
That's the chip specs, the voltage regulators are external to the chips.
In fact, why don't I get the iron hot and see what I can do with a dud blade right now!
Jump to: