The Official X2 Overclocking Discussion Thread

[crazydiamond]

i have a 5000 black edition @ 3.2, 14x230, 1.425vcore. it will run orthos for hours at this setting. the ram is running 4x1gb 940mhz 4-4-4-10.

if i put it @ 3.2 only using multi (16x200) and ram 4x1gb 800mhz 4-4-4-10 it errors out in orthos right away.

now i know different combinations will work, like say you set it to 3.2 using all multi's between 10 and 16, some might be stable and some not, but shouldnt a 16 multi with no other overclocking be stable? maybe its gettin too much volts since im not overclocking anything else im not too concerned about it, as i would rather run at my settings than 16x200 but i was just curious. its makin me think i should go back and try all multi's to get to 3.3 and see if any can be stable.

[Power5]

Multipliers are not free either, just like FSB speeds. I don't really know what the multiplier affects though. What actually happens inside the CPU when you change the mutli from 14-16?

[Morais]

nvm

[BrokeDown]

One thing thats fishy about that is it says "Rated FSB" instead of "HT Link"

[JumpingJack]

Multipliers are not free either, just like FSB speeds. I don't really know what the multiplier affects though. What actually happens inside the CPU when you change the mutli from 14-16?

A phase lock loop consists of a phase detector, a filter, a divider and in most cases a voltage controlled oscillator. The phase detector works to match the input phase with the output phase of the reference clock to the output clock. The PLL is what sets the clock within the CPU (and can be used to set clocks with respect to the reference clock for any device as desgn needs). Phenom has 5 PLLs (5 differen clock domains), typically there is only one PLL in a CPU, AMD's K10 was the first to provide 5 different clock domains in an x86 chip.

When you change multipliers you are effectively changing the number of counts the counter sets for the number of oscillations within one period of the input clock.

The clock is then distributed as any sinusoidal wave through the chip. It does nothing more than change the timing to the clock distribution network within the chip. Nothing magical.

A google reveals all...
http://www.freescale.com/files/rf_if...note/AN535.pdf

Jack

[cdawall]

i have a 5000 black edition @ 3.2, 14x230, 1.425vcore. it will run orthos for hours at this setting. the ram is running 4x1gb 940mhz 4-4-4-10.

if i put it @ 3.2 only using multi (16x200) and ram 4x1gb 800mhz 4-4-4-10 it errors out in orthos right away.

now i know different combinations will work, like say you set it to 3.2 using all multi's between 10 and 16, some might be stable and some not, but shouldnt a 16 multi with no other overclocking be stable? maybe its gettin too much volts since im not overclocking anything else im not too concerned about it, as i would rather run at my settings than 16x200 but i was just curious. its makin me think i should go back and try all multi's to get to 3.3 and see if any can be stable.

mine is stable on the 16x multi @ 1.4v but takes 1.45v to get 3.28ghz on a 16x muli but if i can hit 3.44ghz on a 14x multi @1.64v (244*14) i'm still trying to get this one stable though for now 205*16 is solid

[ownage]

I had a M2A-VM and a 5600+ laying around, thought i give it a try. This matx board without pci/ide lock is really limiting my OC.

Max suicide is 3600mhz flat.

[NEC]

Just got an X2 5200. Never seen one of this before: AD05200IAA5D0 , any thoughts?

[GOLI@TH]

Just got an X2 5200. Never seen one of this before: AD05200IAA5D0 , any thoughts?

Thats new stepping BH-G2, similar to newest Black Edition 5000+.
FYI, i've tried X2 4600+ AD04600IAA5D0 and could run @3Ghz only with default vcore and stock cooling.
Just for example...


Here's the highest clock with stock cooler..
http://valid.x86-secret.com/show_oc.php?id=363087

[P4rD0nM3]

Hey guys, I have a Windsor 4200+ AM2. As of the time of this post, it's about to pass 8 hours of Orthos. Can I consider it stable now? Do I still need to wait 4 more hours? Or wait until a full 24 hour?

Here are my settings.

Bus/Multi @ 264 x 11 = 2.9GHz +/-
HT @ x4/x16
Volt = 1.35v + .225v = 1.575v

4GB DDR2 400 (200/266), YES! It's a value select from Corsair @ 322MHz STOCK VOLTAGE

Temp Readings

Ambient - 25c
Idles - 35/48c (+15 from Ambient)
100% load - 45/58c (+30 from Ambient)

I'm using a Swiftech 120 Compact to cool it. Let me know if there's something wrong with my settings or if I'm over doing some settings.

My mobo is a Biostar TF560 A+.







Are my temps too high? Should I remount it? Lap it? I'm not going to replace the radiator yet...not yet. I know the h2o 120 compact isn't a performance cooler too.

OH and by the way I have no idea what kind of Windsor I have, hehe. So I just used the 1.35v that I usually hear.

[fscussel]

Hello people, I have got a good deal in an 4200+ EE 90nm 65w, my question is: does it overclock as good as the 65nm brisbane brother??

[techtrancer]

X2 4200+ * 3.41 Ghz @ 1.472V * G2 CAABG 0805BAA * Watercooling @ 42°C CPU Temp * without HS * 2,5h DualPrime * Techtrancer Link




X2 4200+ * 2.45 Ghz @ 0,992V * G2 CAABG 0805BAA * Watercooling @ 30°C CPU Temp * without HS * 2 h DualPrime * Techtrancer Link

[blackwarriors]

X2 4200+ Toledo @ 3.88Ghz, 1.90v réel, Zalman 9500Led, 50° Idle

Valid CPU-Z as dead

[carajean]

I was recently thinking about going back to AMD from my E7200, and was wondering when overclocked to lets say 3.0ghz will ANY of these procs perform the same as a same clocked Intel proc??? Im not counting any superpi calcs in this. I just want to know if my computer perform slower or lets say so slow I will notice.

[Fottemberg]

It's my result.

X2 4800+ EE @ 3,0 ghz vCore def on Biostar TF560A2+.

[AncientSword]

Ok, so I am new to the forum and new to OC again (it has been a long time.) I used to work as a network admin about 4 years ago and got into another business. Now I am back and catching up.

I have a bunch of systems running in my house. My traveling gaming rig is E7200@3178 FSB1334, 4GB Gskill PC8000@999, ATI 4850 @750/1188. All perfectly stable. So I have a little figured out.

BUT, my main gaming rig (kinda started out budget then grew a little) is MSI k9Nsliv2 with AMD 5000+BE. System has 2x2GB OCZ1066, Corsair TX750 ps, 2x9800GTX SLI, Zalman 9500LED, single 500GB WD sata.

I am trying to overclock it and I am having problems. I am assuming it is board related. I set it manually in bios to 16x and it accepts it fine. I leave fsb at 200 stock. Set voltage to max mb will allow (which is low from what I have seen in limited reading here). Restart and bios reports 3200mhz. Go into Vistax32 and CPUZ reads stock multiplier (x13) and reports it as 2612 mhz. I have tried several multipliers with same result. What gives here?

Sorry for such stupid questions. I know I need a better MB. Spent too much $$ already trying to catch up but I might have to just upgrade board. Is the 5000BE worth keeping anyway? My e7200@3178 seems WAY faster.

Any help appreciated.

[Dee]

I was recently thinking about going back to AMD from my E7200, and was wondering when overclocked to lets say 3.0ghz will ANY of these procs perform the same as a same clocked Intel proc??? Im not counting any superpi calcs in this. I just want to know if my computer perform slower or lets say so slow I will notice.

I think an overclocked X2 is fast enough for general everyday use and you probably wouldn't be sat there thinking "damn, this thing is so slow...", but basically AMD cannot compete with Intel. Simple as that. I'd probably take even an E4xx over an AMD X2 chip right now, taking overclocking into account.

[Dee]

Is the 5000BE worth keeping anyway?

If money isn't an issue, then the answer is no. A relatively cheap Intel dual core will kick the crap out of any X2 as long as the cache size on the Intel chip isn't too small.

[KeZzZu]

If money isn't an issue, then the answer is no. A relatively cheap Intel dual core will kick the crap out of any X2 as long as the cache size on the Intel chip isn't too small.

This is amd section, dont turn this on flame bait thread thanks, what comes with am2 cpu's those are working with new mobos too which is better than by it's rival intel. and i dont mean in overclocking situations.

[Dee]

I'm not turning anything into "flame bate threads" blah blah blah, or whatever the f it is you are talking about. All I did was state a fact. If that offends you, you have issues and you need to deal with them. Look at my sig. I'm an AMD user. I'm not pro-Intel or anti-AMD, or vice versa. All I am saying is that a cheap Intel dual core will beat any X2, as long as the cache size of the Intel chip is large enough. That's just a fact.

But anyway, I'm done here. Bye!

[techtrancer]

X2 6400+ * 3,62 Ghz @ 1,568V * F3 Winsor * CCB8F 0809 BPMW * Watercooling @ 44°C * mit HS * 2,2 h Prime * Techtrancer Link

[[XC] Synthetickiller]

Another n00b here to bug you guys. Its been forever since I've OC'd an AMD chip (think mobo jumpers). I have really no experience with AMD as of late, so I need some help.

I bought a 5000+ (non black version) from the egg http://www.newegg.com/Product/Produc...82E16819103211

I also have an msi k9a2, both in the mail, on the way to my house in 4 or 5 hours.

I wanted to know, what's the max temp / voltage you can push these things to for 24/7 usage? I'm on stock cooling for now, till I upgrade. Any general tips are appreciated or even a link to a walkthrough for basic bios options.

[techtrancer]

with watercooling 1,5V should be the maximum

with air cooling, i think the maximum ist 1,4V

[64dragon]

dang guys, some nice OC's on the x2 4200's.

i've got an am2 4200 90nm Windsor 65W F2 rev and currently at 2.45 and 1.325v in bios, my board is an m2n sli deluxe and in OS i'm seeing 1.39v with a multimeter. my temps at load are core 1: 43C and core 2: 36C. i've hit 2.55 but wasn't stable at the vcore i had.

since my temps are low do you think i'm safe to push more vcore? i crunch 24/7 so the chip is always under load

[stunned_guy]

X2 4200+ Toledo @ 3.88Ghz, 1.90v réel, Zalman 9500Led, 50° Idle

Valid CPU-Z as dead

What a fake !!!!
That can never be true with stock cooling and such few voltage...never !!!!