Pages:
Author

Topic: Hacking The KNC Firmware: Overclocking - page 41. (Read 144343 times)

hero member
Activity: 784
Merit: 504
Dream become broken often
January 23, 2014, 06:49:55 PM
the list again for reference


4C2 equals 967.5 MHz or ~ 185.76 GH per Asic (theoretical)

Your opinions ... On October Jupiter max Safe Mhz setting ?

tons of mixed info...you decide what you think...KNC engineer HNO in IRC said no more then 50a per vrm or 200a per board is safe...bitromia or something from knc sales said 64a per vrm is safe...so if you believe hno at 50a then 211 would be your safe max...and if ya believe da bit guy then hell you could probably get 261 outta it...depending also how cool you keep the vrm...stock cooling with cover on you probably won't get very far before temps get too high

I have copper heatsinks on vrms n extra fans in there to help with cooling Smiley
hero member
Activity: 560
Merit: 500
January 23, 2014, 07:08:22 AM
the list again for reference
Code:
SPEED     setting 
700.0 1B1      372
712.5      382
725.0 1C1      392
737.5      3A2
750.0 1D1      3B2
762.5      3C2
775.0 1E1      3D2
787.5      3E2
800.0 1F1      3F2
812.5              402
825.0 201      412
837.5      422
850.0 211      432
862.5      442
875.0 221      452
887.5      462
900.0 231      472
912.5      482
925.0 241      492
937.5      4A2
950.0 251      4B2



4C2 equals 967.5 MHz or ~ 185.76 GH per Asic (theoretical)

Your opinions ... On October Jupiter max Safe Mhz setting ?
hero member
Activity: 560
Merit: 500
January 23, 2014, 06:08:25 AM
and so for october jupiter.... max 0x86 line value ? I think that 211 is the safiest with cooled vrms. I could push more Jup ? Any suggestions ?

currently running with 4C2 for my coolest chip...pretty sure i'm gonna push it some more after I install heatsinks on the vrms (currently they are 'just' actively cooled by 4 dedicated 80mm fans)


I'm lost in this thread... 4C2 ?
hero member
Activity: 560
Merit: 500
January 23, 2014, 04:53:49 AM
and so for october jupiter.... max 0x86 line value ? I think that 211 is the safiest with cooled vrms. I could push more Jup ? Any suggestions ?
member
Activity: 100
Merit: 10
January 23, 2014, 03:30:16 AM

Jupiter installed 16 such heat sink 20mm x 15mm x 15mm, now 630GH / s.
member
Activity: 119
Merit: 10
January 23, 2014, 01:23:03 AM
here:

https://bitcointalksearch.org/topic/m.4621190

Hi

My procedure is as follows:

Get some aluminium heatsinks ( I used old graphics card-heatsinks from recycling center, sawed each to 4 pieces so I had 16 heatsinks).

Remove yellow stickkers from vrm, clean well. I used acetone to wipe them clean.

Glue these heatsinks to VRM.s using some stuff meant to conduct heat and bond together. I used arctic silver thermal adhesive.

Those VRMs that are under the aluminium bar I chopped this "rubber thingie" in half and insert that also in the front VRM under the metal bar.

Remove that "airflow" director metal piece thats screwed with 2 screws to asic-bar.

Screw those "case-fans" off, and reposition them in front of the asic-fan, leaning towards them in like 30 degree angle. This is not so easy but they have couple of positions suitable that they dont hit the other fan and when the case is on its easier to mount them "safely". I havent used any stuff that keeps them in place, they are just leaning there. Now they should be blowing straigth to asic-pcb.

People with jupiter i suggest that the get more fans, and dont screw case fans off as they "force" air in. One doesnt want any circulation to happen here.

FEEL with your hand after the pcb when airflow is feeling maximum after the pcb. Adjust the fan position correspondingly.

Now you just have to find proper value for 0x86 line (Please read the post I linked)

Current status:

Im hashing at 479GH/s second not with 2 ASIC-modules. The strange thing is that other board dropped it's amps to 24 but no change in hashing rate! I don't know what happened but my temps are now 34 and 44 degrees on temp sensors.

It says Im using 380 watts on the bertmod with this 479GH  Grin

Im running 335 i I recal.

Jupiters could go as high as 1 TH/s with PROPER cooling, mine is like "DIY-version" with heatsink with footprint of 2x2 and 2x3 cm.

If you find that ORIGINAL-post useful, please donate=) Couple of heartstopping moments with this shit, before anything useful came out of it!   Grin Im sending some to Uski and this guy who found this 0x86 register stuff. I just found "correct" settings.

Do I have the FASTEST SATURN on the planet at the moment?!  Wink


 


ImI
legendary
Activity: 1946
Merit: 1019
January 22, 2014, 10:35:32 PM

hi all,

what vents do you use for additional vrm-cooling?

i am considering to get some of those:

http://www.newegg.com/Product/Product.aspx?Item=N82E16811999344

but i dont like that they come with an extra rpm-controller, thats alot of controllers flying around if i put one of those fans at each board...



Get the Arctic F12 Wink
http://www.newegg.com/Product/Product.aspx?Item=N82E16835186079

I use F12 + some Megaflows, which I found for a very cheap price: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=Megaflow&N=-1&isNodeId=1

ok, i was in doubt if those arctics would supply enough CFM.

i guess some adapters are needed to put them straight at the PSU?

They supply enough imo. Molex adapters are needed or get a cheap fan controller.

thx! lets go and clock those things!
hero member
Activity: 812
Merit: 502
January 22, 2014, 10:25:31 PM

hi all,

what vents do you use for additional vrm-cooling?

i am considering to get some of those:

http://www.newegg.com/Product/Product.aspx?Item=N82E16811999344

but i dont like that they come with an extra rpm-controller, thats alot of controllers flying around if i put one of those fans at each board...



Get the Arctic F12 Wink
http://www.newegg.com/Product/Product.aspx?Item=N82E16835186079

I use F12 + some Megaflows, which I found for a very cheap price: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=Megaflow&N=-1&isNodeId=1

ok, i was in doubt if those arctics would supply enough CFM.

i guess some adapters are needed to put them straight at the PSU?

They supply enough imo. Molex adapters are needed or get a cheap fan controller.
sr. member
Activity: 462
Merit: 250
January 22, 2014, 10:23:40 PM



ok, i was in doubt if those arctics would supply enough CFM.

i guess some adapters are needed to put them straight at the PSU?

before you go spending a bunch of money, did you drop your current fans to the crossbar?

ImI
legendary
Activity: 1946
Merit: 1019
January 22, 2014, 10:17:37 PM

hi all,

what vents do you use for additional vrm-cooling?

i am considering to get some of those:

http://www.newegg.com/Product/Product.aspx?Item=N82E16811999344

but i dont like that they come with an extra rpm-controller, thats alot of controllers flying around if i put one of those fans at each board...



Get the Arctic F12 Wink
http://www.newegg.com/Product/Product.aspx?Item=N82E16835186079

I use F12 + some Megaflows, which I found for a very cheap price: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=Megaflow&N=-1&isNodeId=1

ok, i was in doubt if those arctics would supply enough CFM.

i guess some adapters are needed to put them straight at the PSU?
hero member
Activity: 812
Merit: 502
January 22, 2014, 10:10:40 PM

hi all,

what vents do you use for additional vrm-cooling?

i am considering to get some of those:

http://www.newegg.com/Product/Product.aspx?Item=N82E16811999344

but i dont like that they come with an extra rpm-controller, thats alot of controllers flying around if i put one of those fans at each board...



Get the Arctic F12 Wink
http://www.newegg.com/Product/Product.aspx?Item=N82E16835186079

I use F12 + some Megaflows, which I found for a very cheap price: http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=Megaflow&N=-1&isNodeId=1
ImI
legendary
Activity: 1946
Merit: 1019
January 22, 2014, 09:54:17 PM

hi all,

what vents do you use for additional vrm-cooling?

i am considering to get some of those:

http://www.newegg.com/Product/Product.aspx?Item=N82E16811999344

but i dont like that they come with an extra rpm-controller, thats alot of controllers flying around if i put one of those fans at each board...

ImI
legendary
Activity: 1946
Merit: 1019
January 22, 2014, 08:58:11 PM


Doesn't raise FAULT 4, but still gives you ~800 (with case on).

case on. any additional vram-coolers?
full member
Activity: 147
Merit: 100
software developer
January 22, 2014, 08:44:48 PM
FAULT 4

FAULT 4 is a temperature warning, there is about 10F before it will force a complete thermal shutdown. While it is protection you need to be careful running it with a temperature warning since the part will fail and it may not take too long before it does, best to get the temp down some.

Thanks.

I've modified the line, so others do not experience it.
cmd=$(printf "0x86,0x%02X,0x02,0x85" $c)

Doesn't raise FAULT 4, but still gives you ~800 (with case on).

Code:
DC/DC ID 	ON/OFF 	Status 	Input Voltage 	Output Voltage 	Output Current
0 OFF OK 11.7 V 0.823 V 29.4 A (24.2 W)
1 OFF OK 11.8 V 0.822 V 29.2 A (24 W)
2 OFF OK 11.7 V 0.822 V 28.5 A (23.4 W)
3 OFF OK 11.7 V 0.82 V 29.3 A (24 W)
4 OFF OK 11.8 V 0.825 V 28.3 A (23.3 W)
5 OFF OK 11.8 V 0.825 V 28.2 A (23.3 W)
6 OFF OK 11.7 V 0.821 V 30.1 A (24.7 W)
7 OFF OK 11.7 V 0.822 V 29.9 A (24.6 W)
hero member
Activity: 812
Merit: 502
January 22, 2014, 08:06:04 PM
Thanks. I see that the hex values are different for the November units: 0x02,0xB5
Do you have the rest with their corresponding clocks?

Thank you.

https://bitcointalksearch.org/topic/m.4621190

Thanks. Do you know what Mhz is 305?
Also what is the stock value and stock Mhz?
legendary
Activity: 1428
Merit: 1000
https://www.bitworks.io
January 22, 2014, 08:03:01 PM
FAULT 4

FAULT 4 is a temperature warning, there is about 10F before it will force a complete thermal shutdown. While it is protection you need to be careful running it with a temperature warning since the part will fail and it may not take too long before it does, best to get the temp down some.
full member
Activity: 147
Merit: 100
software developer
January 22, 2014, 07:46:01 PM
Thanks. I see that the hex values are different for the November units: 0x02,0xB5
Do you have the rest with their corresponding clocks?

Thank you.

https://bitcointalksearch.org/topic/m.4621190
hero member
Activity: 812
Merit: 502
January 22, 2014, 07:10:16 PM
Thanks. I see that the hex values are different for the November units: 0x02,0xB5
Do you have the rest with their corresponding clocks?

Thank you.
full member
Activity: 147
Merit: 100
software developer
January 22, 2014, 05:48:26 PM
Hi everybody, waiting for donations  Grin

please use bertmod to see individual vrm stats.

Please keep in mind that i put the front blowers to different place, they aim straight to boards themselves. I also removed the bent-metal part that supplies airflow to beaglebone. Don't know if this has any effect on beaglebone in the long run. I put heatsinks on vrms, took the yellow tape off, cleaned with cleaner + acetone and put arctic silver thermal adhesive, as I didnt want to risk heatsinks "moving" and falling of from vrms. Temperature sensor wont give you correct reading, you can only watch if faults start to appear on vrm's.

Im having one issue at the moment, current readings from other asic have started to fall down. Hashing drops some, but not quite that much.

Yesterday I started to reverse engineer VRM voltage setting, but didnt have guts to try anything yet. On BMR464 manual it says that the output "master" voltage is set with resistor, but you can go like 15% higher by accidently issuing wrong command. Hope that this doesnt kill the asic=)

Im not to blame if it dies, please remember!

Thanks again. You'll receive a donation from me soon for sure  Wink
Hope others do so as well, because without your balls risking to blow up an asic we still wouldn't know about it.


Ok, did some further testing and measurements.

Notice please, case was off otherwise I'm not able to measure temps.
And I've used an infrared pistol, maybe I can borrow another thing which I can put directly on tiny parts to measure temps.

Standard clock-settings:
859W at the wall
67°C for those tiny blocks next to the voltage regulators
80°C for the voltage regulators (yellow area)
cores in the web-gui stats up to 53°C

no big diff between boards
Code:
DC/DC ID 	ON/OFF 	Status 	Input Voltage 	Output Voltage 	Output Current
0 OFF OK 11.8 V 0.829 V 25.2 A (20.9 W)
1 OFF OK 11.8 V 0.829 V 25.4 A (21.1 W)
2 OFF OK 11.8 V 0.827 V 25.3 A (20.9 W)
3 OFF OK 11.9 V 0.828 V 25.4 A (21 W)
4 OFF OK 11.8 V 0.83 V 25.4 A (21.1 W)
5 OFF OK 11.8 V 0.833 V 25.1 A (20.9 W)
6 OFF OK 11.8 V 0.83 V 24.8 A (20.6 W)
7 OFF OK 11.8 V 0.829 V 24 A (19.9 W)


Modified clock-settings:
1098W at the wall
70°C for those tiny blocks next to the voltage regulators
90°C for the voltage regulators (yellow area)
cores in the web-gui stats up to 66°C

For the overclocked one bertmod showed after some time several 'FAULT 4' for the two boards in the back, nevertheless it was hashing very nice.
No clue what this means.
Code:
0 	OFF 	FAULT 4 	11.7 V 	0.819 V 	30.8 A (25.2 W)
1 OFF OK 11.6 V 0.82 V 30.7 A (25.2 W)
2 OFF FAULT 4 11.7 V 0.822 V 31.5 A (25.9 W)
3 OFF OK 11.8 V 0.824 V 31.4 A (25.9 W)
4 OFF OK 11.7 V 0.821 V 30.1 A (24.7 W)
5 OFF OK 11.7 V 0.821 V 30.2 A (24.8 W)
6 OFF FAULT 4 11.7 V 0.823 V 30.7 A (25.3 W)
7 OFF FAULT 4 11.7 V 0.821 V 31.3 A (25.7 W)

BMR series offers a nice feature, btw: "Over temperature protection"
http://www.ericsson.com/ourportfolio/products/bmr464-series


Aaaaannd....here's what you all had been waiting for, I guess.
Gives me peaks in the pool over >850 Smiley

Keep in mind the usual stuff: donate a little to temen, warranty void, blow-up risk, ..

Code:
#!/bin/sh
PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin

use_bfgminer=
if [ -f /config/miner.conf ]; then
        . /config/miner.conf  
fi
if [ "$use_bfgminer" = true ] ; then
        DAEMON=/usr/bin/bfgminer
        NAME=bfgminer
        DESC="BFGMiner daemon"
        EXTRA_OPT="-S knc:auto"
else
        DAEMON=/usr/bin/cgminer
        NAME=cgminer
        DESC="Cgminer daemon"
        EXTRA_OPT=
fi


set -e

test -x "$DAEMON" || exit 0
echo "k";
do_start() {
        # Stop SPI poller
        spi_ena=0
        i2cset -y 2 0x71 2 $spi_ena

        good_ports=""
        bad_ports=""

        # CLear faults in megadlynx's
        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 3 >/dev/null 2>&1 || true
                done
        done

        for p in 0 1 2 3 4 5 ; do
                i2cset -y 2 0x71 1 $((p+1))
                good_flag=0
                ar="$(spi-test -s 50000 -OHC -D /dev/spidev1.0 0x80,3,0,0,0,0,0,0 | tail -c 13)"
                if [ "x$ar" = "x00 30 A0 01" ] ; then
                        good_flag=1
                fi
                ar="$(spi-test -s 50000 -OHC -D /dev/spidev1.0 0x80,2,0,0,0,0,0,0 | tail -c 13)"
                if [ "x$ar" = "x00 30 A0 01" ] ; then
                        good_flag=1
                fi
                ar="$(spi-test -s 50000 -OHC -D /dev/spidev1.0 0x80,1,0,0,0,0,0,0 | tail -c 13)"
                if [ "x$ar" = "x00 30 A0 01" ] ; then
                        good_flag=1
                fi
                ar="$(spi-test -s 50000 -OHC -D /dev/spidev1.0 0x80,0,0,0,0,0,0,0 | tail -c 13)"
                if [ "x$ar" = "x00 30 A0 01" ] ; then
                        good_flag=1
                fi

                if [ "$good_flag" = "1" ] ; then
                        good_ports=$good_ports" $p"
                else
                        bad_ports=$bad_ports" $p"
                fi
        done

        if [ -n "$good_ports" ] ; then
                for p in $good_ports ; do
                        # Re-enable PLL
                        i2cset -y 2 0x71 1 $((p+1))
                        for c in 0 1 2 3 ; do
                                cmd=$(printf "0x84,0x%02X,0,0" $c)
               spi-test -s 50000 -OHC -D /dev/spidev1.0 $cmd >/dev/null
               cmd=$(printf "0x86,0x%02X,0x02,0x85" $c)
               spi-test -s 50000 -OHC -D /dev/spidev1.0 $cmd >/dev/null
               cmd=$(printf "0x85,0x%02X,0,0" $c)
               spi-test -s 50000 -OHC -D /dev/spidev1.0 $cmd >/dev/null
                        done

                        # re-enable all cores
                        i=0
                        while [[ $i -lt 192 ]] ; do
                                i2cset -y 2 0x2$p $i 1
                                i=$((i+1))
                        done
                        spi_ena=$(( spi_ena | (1 << $p) ))
                done

        fi

        if [ -n "$bad_ports" ] ; then
                for p in $bad_ports ; do
                        # Disable PLL
                        i2cset -y 2 0x71 1 $((p+1))
                        for c in 0 1 2 3 ; do
                                cmd=$(printf "0x84,0x%02X,0,0" $c)
                                spi-test -s 50000 -OHC -D /dev/spidev1.0 $cmd >/dev/null
                        done

                        # disable all cores
                        i=0
                        while [[ $i -lt 192 ]] ; do
                                i2cset -y 2 0x2$p $i 0
                                i=$((i+1))
                        done
                        spi_ena=$(( spi_ena & ~(1 << $p) ))
                done
        fi

        # Disable direct SPI
        i2cset -y 2 0x71 1 0

        # Enable SPI poller
        i2cset -y 2 0x71 2 $spi_ena

        start-stop-daemon -b -S -x screen -- -S cgminer -t cgminer -m -d "$DAEMON" --api-listen -c /config/cgminer.conf $EXTRA_OPT
}

do_stop() {
        killall -9 bfgminer cgminer 2>/dev/null || true
}
case "$1" in
  start)
        echo -n "Starting $DESC: "
        do_start
        echo "$NAME."
        ;;
  stop)
        echo -n "Stopping $DESC: "
        do_stop
        echo "$NAME."
        ;;
  restart|force-reload)
        echo -n "Restarting $DESC: "
        do_stop
echo "cooling down for 60s.."
sleep 60 #let it cool down!
        do_start
        echo "$NAME."
        ;;
  *)
        N=/etc/init.d/$NAME
        echo "Usage: $N {start|stop|restart|force-reload}" >&2
        exit 1
        ;;
esac

exit 0
hero member
Activity: 812
Merit: 502
January 22, 2014, 03:01:11 PM
thx for sharing temen!

My experience on a November Jupiter with 305 is so far that it works for one board, avg is >700 and temperature is 80+
Enermax 1500W PSU seems not to be sufficient as it is shutting off when I overclock all 4 boards and cgminer starts putting work on them.

I'm running stable at 2B5 and 2C5 with max temperature up to 80 °C and reaching with this ~830GH/s
I'm using "spi-test -s 50000" not the 55000 Mhz value, btw.

I'll leave it for good now and keep watching Smiley

Can you share the vlock values and corresponding Mhz.
Also what file do you alter and what are the required changes?

Thank you.
Pages:
Jump to: