• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

OC'ing an AM2: An update to Easy 1, 2, 3, Overclocking ...

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

bowman

New Member
Joined
Apr 8, 2008
That setting would be the bus width (8 byte = 64-bit, 16 byte = 128-bit) but I believe if you look farther you'll see a "1000 MHz" value on that page as well.

Like many manual settings in BIOS you may have to turn an "Auto" setting to "Manual" to get more options ... ;)

I finally found this, after wandering around like a drunk in there for a while :beer:

When I did, I followed your guide and found out that the highest stable system clock was 230MHz, that the highest adjustment possible was 425, and that the clock and the lowest multiplier would hit stock speeds at the utterly unreachable 600MHz. I thought, safe then.

After going to the next step and adjusting the multiplier back to standard, HTL back to original setting and system clock to 210 MHz, I got a large red blinking STOP and some very unpleasant beeps from Orthos. This is the message:

Type: Small FFTs - stress CPU Min: 8 Max: 64 InPlace: Yes Mem: 8 Time: 15
CPU: 3149MHz FSB: 200MHz [203MHz x 15.5 est.]
11.04.2008 14:11
Launching 2 threads...
1:Using CPU #0
1:Beginning a continuous self-test to check your computer.
1:press Stop to end this test.
2:Using CPU #1
1:Test 1, 800000 Lucas-Lehmer iterations of M172031 using 8K FFT length.
2:Beginning a continuous self-test to check your computer.
2:press Stop to end this test.
2:Test 1, 800000 Lucas-Lehmer iterations of M172031 using 8K FFT length.
2:FATAL ERROR: Rounding was 0.5, expected less than 0.4
2:Hardware failure detected, consult stress.txt file.
2:Torture Test ran 0 minutes 50 seconds - 1 errors, 0 warnings.
2:Execution halted.
I can't find any 'stress.txt' file in my orthos folder, unfortunately, but I think it's pretty obvious that this MB is not suitable. OCCT and CPU-Z both report my vCore to jump erratically between a few numbers between 1.392 and 1.320, even though I have a 700W power supply. This is both when it's set to 'Auto' and to 1.4V. OCCT reported my load temperatures to be lower than stock when I ran orthos, which I suppose is because it never 'got cracking' before orthos stopped and nearly gave me a heart attack with the case speaker.

A shame, really. I have just ordered a thermalright ultra extreme to really squeeze this CPU, I guess it'll collect dust until I buy a new processor.

A word of advice to anyone searching the forum or randomly reading this thread, then: all other reports about the MSI K9N SLI Platinum (and I'd assume other K9Ns) are correct. It is useless for anything but housewife computers.

I have ordered a Gigabyte PS5 which I hope will work as a temporary fix to let me tinker with this setup until I buy the next one.
 
Last edited:
OP
Q

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
I didn't understand all that for some reason - maybe becuase it's 5 AM. ;) Could you fill in the "?" below for me:

System clock top end: 230 MHz
CPU top end speed, clock, and vCore: ?
RAM top end clock, timings, and vDIMM: ?


Please note, the guide isn't finished yet ... ;)
 

bowman

New Member
Joined
Apr 8, 2008
I didn't understand all that for some reason - maybe becuase it's 5 AM. ;) Could you fill in the "?" below for me:

System clock top end: 230 MHz
CPU top end speed, clock, and vCore: ?
RAM top end clock, timings, and vDIMM: ?


Please note, the guide isn't finished yet ... ;)

Well, I don't understand what you want me to say, so we're in the same boat. :p

The 'CPU top end speed' - I don't know..? Clock was 3000 and increased to 3150 when I set the system clock to 210 according to the guide. vCore was 1.4 as that was what it seemed it was trying to accomplish when it was set to auto (although it was always slightly below 1.4 and jumping erratically between values).

I didn't touch the RAM, I didn't get that far. In any case, here are the values: 1.9V and 400MHz.

I've ordered a Gigabyte MA790FX DS5 card, which I hope will help me in this situation.

The guide may not be finished but I didn't even get halfway through it until my PC decided he had enough, so that's that. :beer:
 
OP
Q

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
Cheat Sheet

Cheat Sheet
NOTE: These tests must be done in order

System Clock Max
System Clock: 200 MHz
CPU Multiplier: 6X
vCore: 1.35v
RAM ratio: 1:1 (DDR2-400)
vDIMM: stock manual
RAM timings: stock manual
HT Link: 3X

1. Increase system clock by 10 MHz
2. Test for stability by running Prime95v25.1/Orthos or SuperPi for a few minutes
3. Repeat steps 1 & 2 until stability test fails
4. Increase NB voltage by one step (if not possible goto #6)
5. Repeat steps 1 & 2 until stability test fails
6. Record last stable system clock and NB voltage


CPU Max
System Clock: 200 MHz
CPU Multiplier: stock manual
vCore: 1.40v
RAM ratio: 1:1
vDIMM: stock manual
RAM timings: stock manual
HT Link: 3X

1. Increase system clock by 5 MHz (do not exceed system clock max!)
2. Test for stability by running Prime95v25.1 or Orthos for a few minutes
3. While stability testing run CoreTemp to check CPU temps. CPU temps in excess of 55°C is failure
4. Repeat steps 1 thru 3 until stability or temp test fails
5. Record last stable system clock, vCore, and CPU temp
6. If the last test was a temp failure goto 9
7. Increase vCore to 1.45v
8. Repeat steps 1 thru 3 until stability test fails again or temp test fails
9. Record last stable system clock, vCore, and CPU temp
FX/BE ONLY
If system clock max is reached before stability or temp failure:
set CPU multiplier to stock +1
set all other values as shown above and retest


RAM Max
System Clock: 200 MHz
CPU Multiplier: 6X
vCore: 1.35v
RAM ratio: 2:1 (DDR2-800)
vDIMM: stock manual
RAM timings: stock manual
HT Link: 3X

1. Increase system clock by 5 MHz (do not exceed system clock max!)
2. Test for stability by running Prime95v25.1 or Orthos for a few minutes
3. Repeat steps 1 & 2 until stability test fails then decrease system clock by 5 MHz
4. Record system clock, RAM timings, and vDIMM

5. Increase vDIMM by 0.1v (exceed 2.1 vDIMM at your own risk!)
6. Repeat steps 1-4
7. Increase vDIMM by 0.1v (exceed 2.1 vDIMM at your own risk!)
8. Repeat steps 1-4

9. Decrease vDIMM to stock +0.1v (exceed 2.1 vDIMM at your own risk!)
10. Relax RAM timings one step (usually +1 +1 +1 +3 but see post #4 RAM Timings)
11. Repeat steps 1-4
12. Increase vDIMM by 0.1v (exceed 2.1 vDIMM at your own risk!)
13. Repeat steps 1-4

14. (If possible) Repeat steps 9-13

If the system clock max is reached before RAM stability failure:
1. Increase RAM ratio one step upward (8:3)
2. Adjust system clock as needed to return RAM to last stable speed tested (see post #5 RAM Ratios)
3. Continue testing


HT Link
Keep the HT Link below 1000 MHz if possible and not to exceed 1100 MHz
 
Last edited:

hokiealumnus

Water Cooled Moderator
Joined
Oct 14, 2007
Just starting to read through the updates. Bump so it doesn't get lost and double-bump for STICKY! :beer:
 

dark bishop

Member
Joined
Jan 19, 2008
Location
Folding the PWN, Salem Oregon
are we supposed to put the HT Link back to normal?

ive gotten to stable at 12x 225mhz(2700mhz). when i first tested at 5x i got up to 245mhz. its not the BE so i guess this is pretty good so far. i havent figured out how to overclock the ram yet and when i have little to no control over vcore.
 
OP
Q

QuietIce

Disabled
Joined
May 7, 2006
Location
Anywhere but there
ive gotten to stable at 12x 225mhz(2700mhz). when i first tested at 5x i got up to 245mhz. its not the BE so i guess this is pretty good so far. i havent figured out how to overclock the ram yet and when i have little to no control over vcore.
That's a good OC if you have no vCore control! :thup:
are we supposed to put the HT Link back to normal?
HT Link
Keep the HT Link below 1000 MHz if possible and not to exceed 1100 MHz
So at 225 I'd use 4X (= 900 MHz) ... :)
 

dark bishop

Member
Joined
Jan 19, 2008
Location
Folding the PWN, Salem Oregon
thanks, didnt see that last ht link part. of course now ive had to go back to stock because its up to 89*F. with the overclock it idled at around 40*C, never checked it under load but i could tell it didnt like the heat. ill give it another go if it cools down, hopefully i get my WC in soon, this chip seemed very stable while it was still cold.\


stock settings
 
Last edited: