Pages:
Author

Topic: Hacking The KNC Firmware: Overclocking - page 28. (Read 144362 times)

legendary
Activity: 1036
Merit: 1000
Nighty Night Don't Let The Trolls Bite Nom Nom Nom
February 21, 2014, 06:58:06 PM


I have inspected what the November firmware has actually changed.

What I can see it has changed a value in one of the registers in the VRMs. It has set the "ON" bit in the OPERATIONS register,changing what conditions are needed to turn the VRMs on.
This change will be permanent even if rebooted or power-cycled. It can be changed using i2c commands:


Turn all VRMs on:

#!/bin/sh
# Stop SPI poller
i2cset -y 2 0x71 2 0

for b in 3 4 5 6 7 8 ; do
        for d in 0 1 2 3 4 5 6 7 ; do
                i2cset -y $b 0x1$d 1 0x80 >/dev/null 2>&1
        done
done


Restore to factory settings:

#!/bin/sh
# Stop SPI poller
i2cset -y 2 0x71 2 0

for b in 3 4 5 6 7 8 ; do
        for d in 0 1 2 3 4 5 6 7 ; do
                i2cset -y $b 0x1$d 1 0x0 >/dev/null 2>&1
        done
done


I suggest stopping cgminer before doing this to limit the power when the operation is done.

Reading the datasheet for the VRMs I can see that when using them in pairs (as we do when all are turned on) there are some requirements on how they should be turned on. This method does not do that!

I have heard that hizzing sound as well. Some boards do that. They did that on 0.90 firmware too. No idea why.

I do not know if this is safe or not, it might be unsafe and burn your boards.
hero member
Activity: 784
Merit: 504
Dream become broken often
February 21, 2014, 06:54:12 PM
A warning:

For you brave souls that have enabled all 8 VRMS (early Octobers) I saw today that the 1.0.0 firmware does not apply the voltage adjustment to all 8 VRMs. It only applies it to the 4 that are officially enabled. The old FW applied this adjustment to all VRMs.

The VRMs work in pairs and if the voltage adjustment (VOUT_TRIM register) of the 2 VRMs in a pair is different things will go bad.

If you still want to use all 8 VRMs I suggest you read and understand the datasheet of the VRM.
The VRMs are very resilient to abuse, but they have limits.


Other than that FW 1.0.0 works very good. HW errors have dropped a lot for me. Good chips are close to 0% errors (2% before), bad chips still close to 2%.

THanks to his help via IRC I was able to update fw and still use my 2 8vrm boards...and my hw error droped down to .11% from .25% yay hopefully it stays that low as i just restarted...Thxs bondus

do you care to share bondus help?

i would post his script if he gave me permission too...he's not on IRC so i can't ask him Sad sorry
legendary
Activity: 1036
Merit: 1000
Nighty Night Don't Let The Trolls Bite Nom Nom Nom
February 21, 2014, 06:27:00 PM
Got my second October 4VRM Jupiter in and on first bootup worked fine for a few hours.  I updated to 1.0 to test and I noticed when I turned off SPI Auto.

I've tried the hit it with SPI 3.3 and max voltage.  Let it run a few hours like that.  I have had them drop when OC to high but this I can't seem to rescue.

There is it at max voltage on that one.  I've swapped back to .99.1 , .99.2 and 1.0 and can't seem to get it to jump back up.

I've tried running 231 and 221 but sadly nothing helped kick that one up even.  Sad

Any other suggestions so I can put this Jupiter to a happy # Wink

Take a screen shot of all your current die settings and save to paint, you may need to zoom out on browser to do this. Then hit reset to default. This may work, then just set everything back.

The other way would be to create more heat. Remove heatsink fan or my preferred method is to place a large durable zip lock bag over the heatsink and fan.
j03
member
Activity: 62
Merit: 10
February 21, 2014, 06:07:16 PM
still running 291/790GH/810w oct. jupiter  Cool

291?! Hot damn! Smiley
sr. member
Activity: 486
Merit: 262
rm -rf stupidity
February 21, 2014, 05:53:37 PM
Got my second October 4VRM Jupiter in and on first bootup worked fine for a few hours.  I updated to 1.0 to test and I noticed when I turned off SPI Auto this:



I've tried the hit it with SPI 3.3 and max voltage.  Let it run a few hours like that.  I have had them drop when OC to high but this I can't seem to rescue.



There is it at max voltage on that one.  I've swapped back to .99.1 , .99.2 and 1.0 and can't seem to get it to jump back up.

I've tried running 231 and 221 but sadly nothing helped kick that one up even.  Sad

Any other suggestions so I can put this Jupiter to a happy # Wink
legendary
Activity: 2408
Merit: 1004
February 21, 2014, 05:53:47 AM
how i can delete the code
and put the new one code?

any new code

is safe to have amper at 50 a or will detsroy it?


Miner Status
Mining Status

CGMiner Status   Running (pid=19609)
Last Checked   Fri Feb 21 12:44:19 UTC 2014
Avg. Hash Rate   645 Gh/s
Difficulty Accepted   44,096
Difficulty Rejected   352 (0.8 %)
Hardware Errors   704 (1.5 %)
WU   9,025
Difficulty Stale   0
Network Blocks   3
Pool Rejected   0.8 %
Pool Stale   0
Best Share   7,434
Found Blocks   0
HW Status

ASIC slot #1   52.0 ℃
ASIC slot #2   -
ASIC slot #3   56.5 ℃
ASIC slot #4   55.0 ℃
ASIC slot #5   40.5 ℃
ASIC slot #6   -
ASIC Board Status
Update ASIC Status
Last updated on Fri Feb 21 12:17:41 2014

ASIC Board   Info
0   
Temperature sensor: 33.0 C

Die ID   Cores ON   Cores OFF   %
0   48   0   100
1   48   0   100
2   48   0   100
3   48   0   100
DC/DC ID   ON/OFF   Status   Input Voltage   Output Voltage   Output Current
0   ON   OK   11.8 V   0.782 V   40.1 A (31.4 W)
1   No DC/DC detected
2   ON   OK   11.8 V   0.798 V   39.8 A (31.8 W)
3   No DC/DC detected
4   ON   OK   11.9 V   0.785 V   41.2 A (32.3 W)
5   No DC/DC detected
6   No DC/DC detected
7   ON   OK   11.8 V   0.771 V   41.2 A (31.8 W)
2   
Temperature sensor: 35.5 C

Die ID   Cores ON   Cores OFF   %
0   48   0   100
1   48   0   100
2   48   0   100
3   48   0   100
DC/DC ID   ON/OFF   Status   Input Voltage   Output Voltage   Output Current
0   ON   OK   11.8 V   0.751 V   41.1 A (30.9 W)
1   No DC/DC detected
2   ON   OK   11.9 V   0.781 V   42 A (32.8 W)
3   No DC/DC detected
4   ON   OK   12 V   0.789 V   41.6 A (32.8 W)
5   No DC/DC detected
6   No DC/DC detected
7   ON   OK   11.8 V   0.804 V   42.4 A (34.1 W)
3   
Temperature sensor: 35.5 C

Die ID   Cores ON   Cores OFF   %
0   48   0   100
1   48   0   100
2   48   0   100
3   48   0   100
DC/DC ID   ON/OFF   Status   Input Voltage   Output Voltage   Output Current
0   ON   OK   11.9 V   0.783 V   40.7 A (31.9 W)
1   No DC/DC detected
2   ON   OK   11.8 V   0.748 V   41.9 A (31.3 W)
3   No DC/DC detected
4   ON   OK   11.9 V   0.778 V   42.6 A (33.1 W)
5   No DC/DC detected
6   No DC/DC detected
7   ON   OK   12 V   0.785 V   41.5 A (32.6 W)
4   
Temperature sensor: 31.0 C

Die ID   Cores ON   Cores OFF   %
0   48   0   100
1   48   0   100
2   48   0   100
3   48   0   100
DC/DC ID   ON/OFF   Status   Input Voltage   Output Voltage   Output Current
0   ON   OK   11.9 V   0.789 V   40.4 A (31.9 W)
1   No DC/DC detected
2   ON   OK   11.8 V   0.792 V   40.8 A (32.3 W)
3   No DC/DC detected
4   ON   OK   11.9 V   0.783 V   41.4 A (32.4 W)
5   No DC/DC detected
6   No DC/DC detected
7   ON   OK   11.9 V   0.808 V   42.2 A (34.1 W)
Total DC/DC power output: 517 W
legendary
Activity: 1260
Merit: 1008
February 21, 2014, 03:38:31 AM
A warning:

For you brave souls that have enabled all 8 VRMS (early Octobers) I saw today that the 1.0.0 firmware does not apply the voltage adjustment to all 8 VRMs. It only applies it to the 4 that are officially enabled. The old FW applied this adjustment to all VRMs.

The VRMs work in pairs and if the voltage adjustment (VOUT_TRIM register) of the 2 VRMs in a pair is different things will go bad.

If you still want to use all 8 VRMs I suggest you read and understand the datasheet of the VRM.
The VRMs are very resilient to abuse, but they have limits.


Other than that FW 1.0.0 works very good. HW errors have dropped a lot for me. Good chips are close to 0% errors (2% before), bad chips still close to 2%.

THanks to his help via IRC I was able to update fw and still use my 2 8vrm boards...and my hw error droped down to .11% from .25% yay hopefully it stays that low as i just restarted...Thxs bondus

do you care to share bondus help?
legendary
Activity: 1036
Merit: 1000
Nighty Night Don't Let The Trolls Bite Nom Nom Nom
February 21, 2014, 03:10:14 AM
still running 291/790GH/810w oct. jupiter  Cool
sr. member
Activity: 347
Merit: 250
February 21, 2014, 12:07:57 AM
Your issues sound somewhat similar to mine. I did the SD card reset. It works, but the machine seems to like me messing around with it less and less, because the "can't browse to machine IP" issue happened again relatively quickly.

How to do a SD-Card reset? Seems I have a bricked KnCMiner...

Hello,

I seemed to have that problem in early January. Following the guide here http://forum.kncminer.com/forum/main-category/hardware/23664-how-to-restore-a-broken-beaglebone-without-the-need-to-rma really helped me out.

If you have a mac you don't need to get an Ubuntu Live USB/CD, if you don't have a mac or linux PC then you need to get the Ubuntu Live USB/CD to format the micro-sd card.

Also make sure you micro-sd card is under 4 GB (Tip: Radioshack does not have them, Walgreens does though) 

Best of luck getting it back up and running!
hero member
Activity: 784
Merit: 504
Dream become broken often
February 20, 2014, 10:30:13 PM
A warning:

For you brave souls that have enabled all 8 VRMS (early Octobers) I saw today that the 1.0.0 firmware does not apply the voltage adjustment to all 8 VRMs. It only applies it to the 4 that are officially enabled. The old FW applied this adjustment to all VRMs.

The VRMs work in pairs and if the voltage adjustment (VOUT_TRIM register) of the 2 VRMs in a pair is different things will go bad.

If you still want to use all 8 VRMs I suggest you read and understand the datasheet of the VRM.
The VRMs are very resilient to abuse, but they have limits.


Other than that FW 1.0.0 works very good. HW errors have dropped a lot for me. Good chips are close to 0% errors (2% before), bad chips still close to 2%.

THanks to his help via IRC I was able to update fw and still use my 2 8vrm boards...and my hw error droped down to .11% from .25% yay hopefully it stays that low as i just restarted...Thxs bondus
j03
member
Activity: 62
Merit: 10
February 20, 2014, 09:06:06 PM
This 1.0 firmware - KnC states you can change clock speeds through the GUI. Can you actually overclock on a per die basis? They were always talking against overclocking and voiding the warranty. Can someone using 1.0 confirm this? Screenshot? Thanks.

Very interesting. I didn't notice this before, but it looks like the default clock of an October Jupiter is 750, and it can be increased to 775 in the 1.0 web GUI.

The default clock of a November Jupiter is 900, and it can be increased to 1000 in the 1.0 web GUI.

I guess it's time to play around now...  Smiley

Ok only to 775 via the GUI on a October Jupiter. That makes sense. They're not letting people do much. Custom hacks are still the way to go then. Thanks.
hero member
Activity: 742
Merit: 500
February 20, 2014, 08:52:25 PM
How we can do that?

Which option button..

Under the "Advanced" tab, each ASIC die now has a pull down menu that allows the user to adjust the chip frequency. It's right beside the voltage pull down button.
legendary
Activity: 2408
Merit: 1004
February 20, 2014, 08:50:23 PM
How we can do that?

Which option button..
hero member
Activity: 742
Merit: 500
February 20, 2014, 08:43:48 PM
This 1.0 firmware - KnC states you can change clock speeds through the GUI. Can you actually overclock on a per die basis? They were always talking against overclocking and voiding the warranty. Can someone using 1.0 confirm this? Screenshot? Thanks.

Very interesting. I didn't notice this before, but it looks like the default clock of an October Jupiter is 750, and it can be increased to 775 in the 1.0 web GUI.

The default clock of a November Jupiter is 900, and it can be increased to 1000 in the 1.0 web GUI.

I guess it's time to play around now...  Smiley
legendary
Activity: 2408
Merit: 1004
February 20, 2014, 07:57:29 PM
Can you post again the details and the code please

Also I can delete and copy the code?
j03
member
Activity: 62
Merit: 10
February 20, 2014, 07:51:24 PM
This 1.0 firmware - KnC states you can change clock speeds through the GUI. Can you actually overclock on a per die basis? They were always talking against overclocking and voiding the warranty. Can someone using 1.0 confirm this? Screenshot? Thanks.
full member
Activity: 147
Merit: 100
software developer
February 20, 2014, 07:44:29 PM
Any details how we can over clocked October Jupiter with new 1.00 firmware ?
I'm running the same script as I did with the previous firmware.
Works fine as usual.
legendary
Activity: 2408
Merit: 1004
February 20, 2014, 07:41:35 PM
Any details how we can over clocked October Jupiter with new 1.00 firmware ?
legendary
Activity: 1260
Merit: 1008
February 20, 2014, 06:03:35 PM
A warning:

For you brave souls that have enabled all 8 VRMS (early Octobers) I saw today that the 1.0.0 firmware does not apply the voltage adjustment to all 8 VRMs. It only applies it to the 4 that are officially enabled. The old FW applied this adjustment to all VRMs.

The VRMs work in pairs and if the voltage adjustment (VOUT_TRIM register) of the 2 VRMs in a pair is different things will go bad.

If you still want to use all 8 VRMs I suggest you read and understand the datasheet of the VRM.
The VRMs are very resilient to abuse, but they have limits.


Other than that FW 1.0.0 works very good. HW errors have dropped a lot for me. Good chips are close to 0% errors (2% before), bad chips still close to 2%.

many thanks bondus. very helpful as usual.

So how did you manage to apply the advanced settings to all the VRMs after upgrading to the new firmware? did you just use the old fetch_advanced_settings.cgi or did you also use the old waas executable?

newbie
Activity: 13
Merit: 0
February 20, 2014, 05:31:06 PM
A warning:

For you brave souls that have enabled all 8 VRMS (early Octobers) I saw today that the 1.0.0 firmware does not apply the voltage adjustment to all 8 VRMs. It only applies it to the 4 that are officially enabled. The old FW applied this adjustment to all VRMs.

The VRMs work in pairs and if the voltage adjustment (VOUT_TRIM register) of the 2 VRMs in a pair is different things will go bad.

If you still want to use all 8 VRMs I suggest you read and understand the datasheet of the VRM.
The VRMs are very resilient to abuse, but they have limits.


Other than that FW 1.0.0 works very good. HW errors have dropped a lot for me. Good chips are close to 0% errors (2% before), bad chips still close to 2%.
Pages:
Jump to: