• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

llano overclockers: your highest ?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Soulkeeper said:
any idea how many cpu's newegg orders at a time ?
I wonder how many of these things they are selling.
would be nice if they said X in stock like some other websites.
Click to expand...

I get the impression that llano isn't the most popular enthusiast system, but it is wildly popular in prebuilt/oem systems.

My system is a light gaming rig, I play eve online at all max settings, skyrim 1080p... Awesome for HTPC uses as well.
 
I have a A6-3670K and a Gigabyte A55M-DS2 board. I was trying to overclock the GPU/IGX from 444 Mhz to 600 Mhz and undervolt it to -0.2V CPU core. I changed the settings in the BIOS, then save and exit. The system then failed to reboot - no hard disk activity, no display, just the fan and power led turned on. No beep code cuz my speaker is probably broken. I turned it off and on a few times but that didn't work.

Now, if I turn it off completely and wait for about 30 seconds, then it will usually turn on normally and can get into Windows. I went into BIOS again and removed the CPU core voltage mod but keep the 600Mhz GPU speed, but once again the system failed to boot properly. Eventually I gave up and just leave everything at default, which it works fine for a few days, without any "burn test" done yet.

I know it is a cheap board that is not meant to OC, but all I want to do is to increase the GPU to 600Mhz (the same as the A8). Has anyone seen a similar problem like this?
 
I decided to spend my money on my truck instead
I'm going to rebuild the engine and put a new exhaust system on.
I've already wasted enough money on this pc for now.

~3.4GHz is fast enough for everything I do 🙂

the only games I ever play are:

oilrush (med/high detail) 30+fps
ut99 (high detail) few hundred fps
wesnoth 2d game
 
Last edited:
I figured out the problem. The problem was caused by the voltage mod (originally I applied -0.2V on CPU core, but on the second attempt I think I didn't remove it completely). I tried -0.1V and it didn't work. Kind of strange.
 
I have an A6-3650, Gigabyte A75M-UD2H, Zalman CNPS9900MAX (5v), CX430 Builder PS, Kingston Blu 1600MHz DDR3 originally configured as an Windows HTPC but recently switched to Gentoo Linux for media (XBMC) and light gaming (Linux & Wine).

Operating at 3.510Ghz with a 135Mhz FSB and RAM at 1,440 Mhz (5.33x) with +0.75v applied to the CPU and NB.

The only way to set a high FSB speed (greater than 105Mhz) is to switch the AHCI interfaces to IDE, and additionally connect the boot SATA device to the 5th port (labelled SATA3_4) on the motherboard, which will force the use of the PATA_ATIIXP interface.

Minimizing noise is a priority so all fans are at 5v, and the system is stable under nominal load (gaming & XMBC). However, several hours of prime95 increase overall case temperature and eventually one thread will fail, at which point the CPU temps fall back in line and the system continues on. Rock stable for hours with the case open.

Next step is testing at +0.50v at slightly reduced speeds, and zero in on a happy medium while maintaining a near zero-noise case.
 
Last edited:
I've dropped the CPU voltage adjustment to +0.5V and the FSB to 134Mhz (1.48Ghz CPU speed), and temperatures have stabilized, however still running a little hot for my liking.

I have a Gigabyte A75M-UD2H, and one of the thermistors is reporting in the low 70 Celsius range.

Soul_Keeper, do you know where the thermistors are located?
 
krc said:
I have an A6-3650, Gigabyte A75M-UD2H, Zalman CNPS9900MAX (5v), CX430 Builder PS, Kingston Blu 1600MHz DDR3 originally configured as an Windows HTPC but recently switched to Gentoo Linux for media (XBMC) and light gaming (Linux & Wine).

Operating at 3.510Ghz with a 135Mhz FSB and RAM at 1,440 Mhz (5.33x) with +0.75v applied to the CPU and NB.

The only way to set a high FSB speed (greater than 105Mhz) is to switch the AHCI interfaces to IDE, and additionally connect the boot SATA device to the 5th port (labelled SATA3_4) on the motherboard, which will force the use of the PATA_ATIIXP interface.

Minimizing noise is a priority so all fans are at 5v, and the system is stable under nominal load (gaming & XMBC). However, several hours of prime95 increase overall case temperature and eventually one thread will fail, at which point the CPU temps fall back in line and the system continues on. Rock stable for hours with the case open.

Next step is testing at +0.50v at slightly reduced speeds, and zero in on a happy medium while maintaining a near zero-noise case.
Click to expand...

Cutting the CFM on fans that aren't designed for low CFM is a fruitless venture (as you've already discovered), the faster the blade spins at default the worse the blades will do in a low power state so if you felt the need to reduce speed in the first place I doubt those fans are ideal for what you're trying to get done. I'd grab an all copper cooler, one or two of the largest PWM fans you can fit and be done with it. That or buy an open air case and use no fans at all other than the one on the CPU.
 
Here are the sensor readings.
Idle:

Code: 
it8720-isa-0228
Adapter: ISA adapter
in0:          +1.04 V  (min =  +0.00 V, max =  +4.08 V)
in1:          +1.49 V  (min =  +0.00 V, max =  +4.08 V)
in2:          +3.33 V  (min =  +0.00 V, max =  +4.08 V)
+5V:          +3.04 V  (min =  +0.00 V, max =  +4.08 V)
in4:          +3.07 V  (min =  +0.00 V, max =  +4.08 V)
in5:          +2.14 V  (min =  +0.00 V, max =  +4.08 V)
in6:          +2.14 V  (min =  +0.00 V, max =  +4.08 V)
5VSB:         +2.98 V  (min =  +0.00 V, max =  +4.08 V)
Vbat:         +3.17 V  
fan1:         485 RPM  (min =   10 RPM)
fan2:        1013 RPM  (min =    0 RPM)
fan3:           0 RPM  (min =    0 RPM)
temp1:        +36.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermistor
temp2:        +70.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermistor
temp3:        +14.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermistor
cpu0_vid:    +0.000 V
intrusion0:  OK

k10temp-pci-00c3
Adapter: PCI adapter
temp1:         +4.9 C  (high = +70.0 C)
and under sustained load:
Code: 
it8720-isa-0228
Adapter: ISA adapter
in0:          +1.44 V  (min =  +0.00 V, max =  +4.08 V)
in1:          +1.49 V  (min =  +0.00 V, max =  +4.08 V)
in2:          +3.30 V  (min =  +0.00 V, max =  +4.08 V)
+5V:          +3.04 V  (min =  +0.00 V, max =  +4.08 V)
in4:          +3.04 V  (min =  +0.00 V, max =  +4.08 V)
in5:          +2.14 V  (min =  +0.00 V, max =  +4.08 V)
in6:          +2.14 V  (min =  +0.00 V, max =  +4.08 V)
5VSB:         +2.96 V  (min =  +0.00 V, max =  +4.08 V)
Vbat:         +3.17 V  
fan1:        1503 RPM  (min =   10 RPM)
fan2:        1352 RPM  (min =    0 RPM)
fan3:           0 RPM  (min =    0 RPM)
temp1:        +40.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermistor
temp2:        +98.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermistor
temp3:        +66.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermistor
cpu0_vid:    +0.000 V
intrusion0:  OK

k10temp-pci-00c3
Adapter: PCI adapter
temp1:        +56.5 C  (high = +70.0 C)
Some observations:

1. it8720 temp2 and 3 are all over the board, and definitely aren't correct. At idle they read 70C and 14C, and no part of the motherboard or cpu are cool (impossible) or excessively warm to the touch.

2. in0 appearss to be CPU voltage; at idle it's 1.04 (and CPU's idling at 800mhz * 134/100) and under load it jumps to the full 1.44V. So it's good to see Vid adjustments working.

3. The k10temp CPU temperature also isn't an absolute value.
From /usr/src/linux/Documentation/hwmon/k10temp

There is one temperature measurement value, available as temp1_input in
sysfs. It is measured in degrees Celsius with a resolution of 1/8th degree.
Please note that it is defined as a relative value; to quote the AMD manual:

Tctl is the processor temperature control value, used by the platform to
control cooling systems. Tctl is a non-physical temperature on an
arbitrary scale measured in degrees. It does _not_ represent an actual
physical temperature like die or case temperature. Instead, it specifies
the processor temperature relative to the point at which the system must
supply the maximum cooling for the processor's specified maximum case
temperature and maximum thermal power dissipation.

The maximum value for Tctl is available in the file temp1_max.
Click to expand...
And pulling these values:
Code: 
./devices/pci0000:00/0000:00:18.3/temp1_input:63250
./devices/pci0000:00/0000:00:18.3/temp1_max:70000
So it seems the system is still roughly in bounds by about 10%. A little close for comfort, so I'm going to keep stepping down voltage (wikipedia: The power consumed by a CPU, is approximately proportional to CPU frequency, and to the square of the CPU voltage).
 
Last edited:
krc said:
I've dropped the CPU voltage adjustment to +0.5V and the FSB to 134Mhz (1.48Ghz CPU speed), and temperatures have stabilized, however still running a little hot for my liking.

I have a Gigabyte A75M-UD2H, and one of the thermistors is reporting in the low 70 Celsius range.

Soul_Keeper, do you know where the thermistors are located?
Click to expand...

In linux the temperature will be wrong on this board by default
Here is the fix I came across after scratching my head for awhile:

echo 3 > /sys/class/hwmon/hwmon1/device/temp2_type

the problem is it reads it as a thermistor instead of a thermal diode
likely because some other board/cpu uses it87 that way

Put this in one of your start scripts, I chose to put in in my /etc/rc.d/rc.fancontrol
and I have that setup nicely to adjust fan speeds

Also note that temp1 is the fch (accurate) and temp3 seems to be disconnected (you can add an ignore in /etc/sensors3.conf if you want).

My current sensors3.conf
This gives idle sensors output like this:
HTML: 
it8720-isa-0228
Adapter: ISA adapter
VCore:        +1.09 V  (min =  +0.00 V, max =  +4.08 V)
DRAM:         +1.50 V  (min =  +0.00 V, max =  +4.08 V)
+3.3V:        +3.39 V  (min =  +0.00 V, max =  +4.08 V)
+5V:          +5.03 V  (min =  +0.00 V, max =  +6.85 V)
+12V:        +12.54 V  (min =  +0.00 V, max = +16.32 V)
+5V Stdby:    +5.08 V  (min =  +0.00 V, max =  +6.85 V)
VBatt:        +3.15 V  
CPU Fan:     1415 RPM  (min =   10 RPM)
SYS Fan:     1173 RPM  (min =    0 RPM)
FCH Temp:     +34.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermistor
CPU Temp:     +29.0°C  (low  = +127.0°C, high = +127.0°C)  sensor = thermal diode
intrusion0:  OK

load properly scales into the 55-65 range depending on fan speed setting
 
Last edited:
BD - thanks for the tip. I checked out the case you're using, and it's practically an open air model; definitely something to keep in mind for my next build. Also found some interesting self-made open air syle cases (with lots of zip ties 🙂. Plenty of good stuff.. but would have to pass the wife-test!

Soul_keeper -- thanks for the help! I replaced my sensors3.conf with your updated version, and the numbers look right. The CPU is now down to +0.25V at 134Mhz bus (3.484Ghz) and stable.

Here's the load temperature:
Code: 
VCore:        +1.41 V  (min =  +0.00 V, max =  +4.08 V)
DRAM:         +1.49 V  (min =  +0.00 V, max =  +4.08 V)
+3.3V:        +3.30 V  (min =  +0.00 V, max =  +4.08 V)
+5V:          +5.11 V  (min =  +0.00 V, max =  +6.85 V)
+12V:        +12.16 V  (min =  +0.00 V, max = +16.32 V)
+5V Stdby:    +4.97 V  (min =  +0.00 V, max =  +6.85 V)
VBatt:        +3.17 V  
CPU Fan:     1513 RPM  (min =   10 RPM)
SYS Fan:     1457 RPM  (min =    0 RPM)
FCH Temp:     +41.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermistor
CPU Temp:     +58.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermal diode
58 Celsius is definitely more reasonable, and somewhat expected given the Zalman CNPS9900MAX is quite a monster cooler.

idle reading:
Code: 
VCore:        +1.04 V  (min =  +0.00 V, max =  +4.08 V)
DRAM:         +1.49 V  (min =  +0.00 V, max =  +4.08 V)
+3.3V:        +3.33 V  (min =  +0.00 V, max =  +4.08 V)
+5V:          +5.08 V  (min =  +0.00 V, max =  +6.85 V)
+12V:        +12.29 V  (min =  +0.00 V, max = +16.32 V)
+5V Stdby:    +5.00 V  (min =  +0.00 V, max =  +6.85 V)
VBatt:        +3.17 V  
CPU Fan:      511 RPM  (min =   10 RPM)
SYS Fan:     1040 RPM  (min =    0 RPM)
FCH Temp:     +37.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermistor
CPU Temp:     +29.0 C  (low  = +127.0 C, high = +127.0 C)  sensor = thermal diode
 
Last edited:
I don't think i've ever had a CPU that idled under 30C
I really like the powernow on these llanos

krc - if you have two pwm fans lemme know, I can post my /etc/fancontrol
I really like the fancontrol script.
 
man, i am having a tough time with this little bastard. i have tried 2 different sets of ram, all kinds of different timings, different voltages, yet still cannot prime for longer than 15 minutes before core one fails, even at stock speeds. i really think i might have a bad chip, or maybe a flaked out board.
 
MrVeedo said:
man, i am having a tough time with this little bastard. i have tried 2 different sets of ram, all kinds of different timings, different voltages, yet still cannot prime for longer than 15 minutes before core one fails, even at stock speeds. i really think i might have a bad chip, or maybe a flaked out board.
Click to expand...

I was lucky with mine, but you wouldn't be the first on these forums to get a DOA llano.
Time to RMA if everything is indeed stock and it's not working imo.
 
Soulkeeper said:
I was lucky with mine, but you wouldn't be the first on these forums to get a DOA llano.
Time to RMA if everything is indeed stock and it's not working imo.
Click to expand...

This will be the first doa chip in my hardware history, and there's been a lot! Wish i had another chip to try, but this is my one and only fm1.
 
Soulkeeper said:
I don't think i've ever had a CPU that idled under 30C
I really like the powernow on these llanos

krc - if you have two pwm fans lemme know, I can post my /etc/fancontrol
I really like the fancontrol script.
Click to expand...

Would definitely like to try it out.

I'm currently using the BIOS auto-control settings, which drops the CPU speed to ~500 RPM (pretty much silent), and then ramps it up to 1500 RPM during full load.
 
New BIOS for FM1 boards (checked both Gigabyte and ASUS, both have new bios released this week). My Gigabyte changes are:
Beta BIOS
Update CPU AGESA code

I wonder what's this Beta BIOS?
 
well, it indeed was a bad chip. new one primes fine at stock speed, right now im sitting at 3600/900, need to check stability. wheres the beta bios? what boards?
 
dont be too envious, it isnt stable! priming at 3500/800 now, its looking like these things like a bit of voltage, which sucks for htpc usage. still not seeing any beta bios for the asus. latest release i see is 1903 released in december.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top