AMD Athlon 64 X2 Overclocking Results

[imported_PrinceGaz]

Originally posted by: Duvie
I agree but coming from my winnie where it wasn't flaky memory but a picky memory controller...I really think they have dialed the memory controller in much better...

yep I am going to take a shot at the AS5....may still look at getting a different fan...perhaps a 90 to 120mm converter and a 120mm fan....I can tell you it has done wonders for my memory...I believe I was overheating at 2.85v with my gskills but with the huge overhang of the XP90 and the downdraft of air I am running this memory much better...It may have something to do with those lack of BSOD as well....

You definitely got a nice chip...The last place I could use 1.44v was at like 2.5ghz and 1.34-1.35v was at 2.4xghz...

Yeah, I was really pleased that the XP-90C overhung the memory modules on my DFI board too, it's a free bonus on top of the already excellent cooling it provides the CPU. Of course the air being blown down onto the modules is probably warmer than the rest of the air in the case, but having some real airflow past the modules rather than them just sitting hot in what little case airflow reaches them more than makes up for that I'm sure, and makes the modules cooler overall.

Why would you want to stick a 120mm fan on the XP-90C through a converter? A 50cfm (2500rpm) 92mm fan is pretty quiet and any more airflow can at best only lower temps by a degree or two. Even the 92mm 4800rpm 120cfm screaming Vantec Tornado won't make much difference to the temperatures, though your eardrums will suffer

Yeah it seems like I have got a somewhat above average 4400+ judging from overclocking reports on various sites, which is nice as my last box dated from late 2001 (an XP 1700+) and I hope this will last at least three years as well so anything extra I get from it buys extra time. It's worth noting that the speeds and voltages I mentioned were those I set in the BIOS, the voltages reported for them both in the BIOS and by MBM5 were about 0.02V lower so it might really have been 2.55GHz with 1.33V, 2.60GHz with 1.38V, 2.65GHz with 1.43V, and 2.68GHz with 1.48V. Unless the voltage sensor was giving duff readings Either way I'm happy with a rock-solid 2.6GHz as that is to all intents and purposes an X2 5200+ which is quite a monster!

 

[imported_PrinceGaz]

Originally posted by: chinkgai
water cooling alone

black ice xtreme III
danger den tdx
swiftech mcp655

cpu is only thing in the loop

i run 2.8 instead of 2.9 daily and loads at 38C

Very nice. That makes effectively makes it a 5600+, or 5800+ if you push it! The A64 X2 2.6GHz parts aren't due to be released until Q1 2006, and the 2.8GHz ones aren't even on the roadmap yet, so you have a seriously fast machine that will probably be faster than anything officially available for at least a year.

I considered water-cooling but decided an XP-90C, 92mm fan, and AS5 paste, for £30 in total (roughly US$55) was more straightforward and less likely to need routine maintenance than a water-cooling solution). Unfortunately I think water-cooling might be mandatory in three or so years when I build my next box. but hopefully leaks will be a rarity. I know a properly put together water-cooled system today won't leak, but I'm a coward and the idea of water inside my computer scares me

 

[Pabster]

Seems kind of pointless to drop several hundred bucks on water cooling to gain 200MHz-ish doesn't it?

I'll take 2.6ish on AIR over 2.8ish on water any day.

That said, chinkgai, you've done very well

 

[chinkgai]

i was a coward but i took the leap of faith

i always was extreme air too till these x2's...just run so damn hot. i used to have a xp120, then i switched to a slk948u cuz my cpu is now running nude (ihs less). i finally changed to water when i saw i was approaching near 60's at not even extreme voltages of around 1.55. i'd get bsod after some stress testing.

now im set with the only thing to change from now on is the cpu block. the silence and extreme temp drops are well worth it. this is my first attempt at water cooling evaar too.

 

[chinkgai]

Originally posted by: Pabster
Seems kind of pointless to drop several hundred bucks on water cooling to gain 200MHz-ish doesn't it?

I'll take 2.6ish on AIR over 2.8ish on water any day.

That said, chinkgai, you've done very well

its a hobby and a learning experience

i dont need this much computing power at all but its just addicting to tweak and tinker!

much cheaper than my old car/rice hobby :laugh:

 

[changer]

Hey guys I just got my 4400+ and I've been putting it through it's paces. I can go up to 227FSB x 11 4HTT, 1.45V and 1:1 Ram (200mhz) 3,3,3,7 (crucial Value) but Windows gives me system errors "a machine check error has occured" I've lowered the fsb to 223 now and raised the voltage to the proc. to 1.45V and ram to 2.9V. So far running the machine for the last 10 minutes I'm not getting any windows errors. This is running a Shuttle SN25P. I'm getting 2453Mhz, not too shabby but hearing about 2.6Ghz + makes me jealous. Any tips for me out there guys, especially you Shuttle SN25P users????
Thanks in advance

 

[subzero813]

what are the stock voltage numbers?

how can you run your FSB at 300MHz on PC3200 ram (200MHz) and not have crazy heat issues and still run 2.7v? let alone the same issues with voltage/heat for the motherboard which is also meant for 200mhz FSB...

 

[imported_PrinceGaz]

Originally posted by: changer
Hey guys I just got my 4400+ and I've been putting it through it's paces. I can go up to 227FSB x 11 4HTT, 1.45V and 1:1 Ram (200mhz) 3,3,3,7 (crucial Value) but Windows gives me system errors "a machine check error has occured" I've lowered the fsb to 223 now and raised the voltage to the proc. to 1.45V and ram to 2.9V. So far running the machine for the last 10 minutes I'm not getting any windows errors. This is running a Shuttle SN25P. I'm getting 2453Mhz, not too shabby but hearing about 2.6Ghz + makes me jealous. Any tips for me out there guys, especially you Shuttle SN25P users????
Thanks in advance

There are lots of good A64 overclocking guides about, theres probably one here one the AT forums, and there are definitely good ones on the Guru3d and HardOCP forums.

The basic rule is to do one thing at a time, find out what your mobo can reach (HT speed) when the CPU and RAM are both at much lower multipliers to remove them from the equation.

Then leaving the HT multiplier low (say at x3), and the memory at DDR266 (2:3), see how high you can raise you CPU to find its limit. Rather than have Windows BSOD on you, run a Prime95 Small FFT Torture Test for just five minutes at each speed before raising the FSB another 5MHz. Prime95 will report problems long before Windows goes funny. Be sure to have something like Motherboard Monitor 5 (MBM5) or whatever software that came with your mobo that allows you to monitor CPU temperature in Windows, and don't let it go above 55C, or 60C as an absolute max. When Prime95 gives problems, edge the voltage up slightly, say from 1.35V to 1.40V and see if that helps. Rinse and repeat, raising the speed, and voltage as needed, until you find the limit, but watch those temperatures as raising the voltages pushes the temperatures up quite a bit, and if you hit 65C you could damage your CPU beyond repair (what I like to call "meltdown temperature"

Finally once you've found the limit of your CPU, only then find what your RAM can do. It is far less important than the CPU speed and there is also the additional fun of memory timings to contend with, so feel free to just run the memory at a suitable divider that isn't overclocking it. So if the FSB is anything up to 240MHz, set it to DDR333. Memory overclocking and tweaking timings and voltages is something a book could be written about so I won't go into details. It'll only gain you 1-2% anyway so if in doubt just run at DDR333 if your system isn't stable at the DDR400 speed.

 

[imported_PrinceGaz]

Originally posted by: subzero813
what are the stock voltage numbers?

how can you run your FSB at 300MHz on PC3200 ram (200MHz) and not have crazy heat issues and still run 2.7v? let alone the same issues with voltage/heat for the motherboard which is also meant for 200mhz FSB...

Not sure who you were replying to, but running the FSB at 300 is fine on some of the better motherboards available these days. Good enthusiast oriented mobos are designed with overclocking in mind (my DFI nF4 SLI-DR can do 300), even if the manufacturers do not officially condone it. Depending on your RAM, you'll probably need to use a divider but that's fine as there is no penalty for not running memory 1:1 on an Athlon64 (I've verified that myself with the ScienceMark 2 latency test).

 

[changer]

Hey PrinceGaz, I went through the guide they had on Hocp. I found that my mobo can do up to 278Mhz FSB. My cpu can go up to 228FSB wiht stock multiplier with the memory and HT set really low. I'm running at 227FSB right now memory at 1:1 3,3,3,7 and 4x HT multiplier. The thing I'm worried about is that I had to raise the voltage to 1.475. I'm going to try running Prime95 and see what temps I get under load.

 

[Polish3d]

Originally posted by: PrinceGaz

Then leaving the HT multiplier low (say at x3), and the memory at DDR266 (2:3), see how high you can raise you CPU to find its limit. Rather than have Windows BSOD on you, run a Prime95 Small FFT Torture Test for just five minutes at each speed before raising the FSB another 5MHz. Prime95 will report problems long before Windows goes funny. Be sure to have something like Motherboard Monitor 5 (MBM5) or whatever software that came with your mobo that allows you to monitor CPU temperature in Windows, and don't let it go above 55C, or 60C as an absolute max. When Prime95 gives problems, edge the voltage up slightly, say from 1.35V to 1.40V and see if that helps. Rinse and repeat, raising the speed, and voltage as needed, until you find the limit, but watch those temperatures as raising the voltages pushes the temperatures up quite a bit, and if you hit 65C you could damage your CPU beyond repair (what I like to call "meltdown temperature"

Can you source the basis of the comment about 65 being 'meltdown temperature'? After all, in the OCing guide here on this forum the fella says don't worry as long as it doesnt go ABOVE 65 degrees but that 65 itself is not a major concern.

 

[Diogenes2]

The AMD specs call for max temp of 65..

Interestingly, it is 71 for the 3800+ x2

 

[Bona Fide]

Originally posted by: Diogenes2
The AMD specs call for max temp of 65..

Interestingly, it is 71 for the 3800+ x2

Very interesting...I guess they're using a new revision of the Manchester cores.

@OP : Isn't 1.55vcore a BIT too high?

 

[Darkstar757]

Well folks I got my 3800+ up too 2.2ghz working on more

 

[Diogenes2]

Originally posted by: Bona Fide

Originally posted by: Diogenes2
The AMD specs call for max temp of 65..

Interestingly, it is 71 for the 3800+ x2

Very interesting...I guess they're using a new revision of the Manchester cores.

@OP : Isn't 1.55vcore a BIT too high?

Not necessarily, if the PSU will handle it and the temp doesn't max out..

Remember, these are Venice cores.. I've personally had a 3500+ at 1.7 for several months..

Not that I would recommend 1.7 for the X2..

I wouldn't worry about the volts, just keep the temp below the max at load...

 

[Mickey21]

Originally posted by: dguy6789
You wouldn't want something like water cooling. If you are trying to overclock, watercooling will not give you much more than a top end air cooling setup. W/C is good for high end air cooling performance without the noise. But on the downside, there is a ton of upkeep and maintanance required, expensive, and even a slight leak can fry your pc.

I have done water cooling for about 8 months running an Athlon XP 1700+ tbred at around 2.4ghz. To be honest, all the trouble it was, it wasn't worth it. I would rather have had just some good air cooling and not worry about refilling the water every month and algae growing and such.

No offense, but correct watercooling is nothing like you mention. I think you need to work with it more. Filling water every month? Algae growth and such? You got to be kidding me. You did use good tubing right? Hence likely water loss rate. You did use growth inhibitors right (ie radiator fluid, etc.)? Hence likely algae growth. You did leak test the setup (48 hour run with no cooling)? Hence the risk of water leaks. Watercooling is very safe when done correctly. Distilled water also does not conduct electricity hence no risk of shock but that wouldnt matter with a properly fitted and tested setup. I have watercooled for over 5 years and have never had a leak or problem I can attribute to watercooling.

If you need help, I would be glad to offer pointers, but air cooling is not much compared to proper watercooling. It is much quieter and more efficient on average. Removal of heat from components and pushed out of the entire system is greatly enhanced over air. And with cleaning out the system less than once a year, maintenance is right on par with performance. I have had systems run more than 2 years on the same water loop.

 

[Mickey21]

Originally posted by: jpeyton

So with mid-end air cooling and a safe VCore of 1.52V, we have the equivalent of a AMD Athlon 64 X2 5200+, or two AMD FX-55s in one socket. Not bad.

It all looked good until that statement. I have personally taken FX-55's with a single 850XT PE past 7400 in 3DMark 2005. To say that your setup is like two FX-55's in one socket and only get less than 6300 in 3DMark 2005 is very skewed. Good scores, yes. Better than high end gaming processors, not necessarily...

 

[Duvie]

In theory if the clock speed is the same and you could get all the other tangibles like memory speed and timings the same an X2 at 2.6ghz should outscore an FX55...Why?? cause most FX55's are the 130nm version and dont have the core revisions that came with the E6 and the San Diegos. That is why if you look at reviews in some games a 4800+X2 beats the FX55 and/or beats the 4000+ (non san diego...)...There is no other differnec between the cores of an FX57 and the current X2's...if same speed can be achieved and ram held to same they should be the same, period...

 

[Xed]

Got my 4400+ today

0528 ccbwe

let's hope she clocks well

 

[seanp789]

Originally posted by: PrinceGaz

Also I have had some better like Oc'iong in the past day. These 4400+ badly need to be burnt-in i went from 2.4 to 2.5Ghz stable without actually changing any other settings. I jsut had to wait a day. How weird is that? The day I got I could only get 2.2 Stable. so im not sure what my max is but i might give it a week and then try to up it again.

I don't know what thermal compound you are using, or how long you have been testing your box, but apparently the Arctic Silver 5 I'm using suggests it takes several days and the computer being powered-down for a while every so often to become fully effective, and maybe other compounds are similar. The info for AS5 reads

"Due to the unique shape and sizes of the particles in Arctic Silver 5's conductive matrix, it will take a up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity. (This period will be longer in a system without a fan on the heatsink or with a low speed fan on the heatsink.) On systems measuring actual internal core temperatures via the CPU's internal diode, the measured temperature will often drop 2C to 5C over this "break-in" period. This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired."

I have also heard that CPUs themselves can get slightly better when pushed towards their limit for several days, but I've no idea why. Maybe it's a combination of those two factors contributing to your better overclocking success. I'm assuming of course that your room temperature hasn't fallen dramatically over the last few days

Wow, that explains like everything. Cause I tried regreasing my HSF and I had to start all over again but I had no idea that the arctic silver itself would actually change with age.

I couldnt understand why my HSF ould be cool to the touch and yet my cpu temps were uncomfortably in the 60's C.

I read the instructions, alot fo good stuff for anyone overclocking.


http://www.arcticsilver.com/arctic_silver_instructions_big2.htm

 

[seanp789]

Originally posted by: Diogenes2
The AMD specs call for max temp of 65..

Interestingly, it is 71 for the 3800+ x2

I had my X2 4400+ running oc'd at 68C for serveral hours, it didnt melt but im sure thats got to be way to hot to be running it normally. I try to keep it at 60 or below to find my ideal OC. oddly enough its 39C at idle, so that is a HUGE jump at load temps. Considering watercooling, but cant justify the additional price/performance at this point. If I have to spend $300 on watercooling, I may as well spend $800 on a vapochill LS and get subzero temps.

 

[Tijean66]

Hi, yes you should go watercooling and let me suggest you the EXOS 1 or EXOS 2 from koolance, very easy to setup. I have mine (Exos 1) since one year. At first I just tried the stock fan with my new AMD X2 4400+ and the GeForce 7800GTX, in idle mode it's ok but as soon as you load the CPU and the video card, the noise is freaking crazy.:shocked:

Exos almost noiseless, with my new case (Thermaltake Tsunami) and his 2 120mm silent fan , my computer is very quiet.

X2 in idle mode 31-35 (sensor of koolance 31, asus probe 35)
X2 in load mode 39-43 dpending of ambient temperature

The 7800GTX is 38-40 idle and 41-43 load mode....

Watercooling is the way to go !

 

[BlackDragonX]

Hey guys, I have a problem and need some help. First let me tell you about my system:

Asus A8N-SLI Premium
AMD X2 4400+
2Gb of Corsair TWINX1024-3200LLPRO (512mb X 4)
XFX 7800 GTX Overclock(only 1 card for now)
WCD 74Gb Raptor

The system works great , but I of course want to push it. I did some reading on overclocking the X2 4400 and decided to overclock mine. So here is my problem

iF i leave the memory settings on AUTO I can push the CPU as far as I want, but the system defaults the memory to 333Mhz so my 3dmark05 scores drop by almost 1000 points. As soon as I try to set the memory manually the system wont boot. The best I got it to do was

CPU = FSB 212 @ 1.375 Volts
Memory = 2.3.2.6 @ 2.9 volts (at lower voltages the memory is not stable)
which gives me 2.33Ghz, but this is nothing compare to what others have gotten.

Any ideas? I had the same menory in my oder system and I was able to get it to 220Mhz at 2.8volts and timings of 2.3.2.5.

Thanks in advance for your help guys.

 

[Mariosalice]

Here my problems are worse with my 4400+. I can only go up to 208X11=2288 on an Asus A8N-SLI Dlx. With my previous Winchester 3500+ on the same system I could easily go up to 230X11=2530MHz rock stable. Any ideas?

 

[Markfw]

Originally posted by: BlackDragonX
Hey guys, I have a problem and need some help. First let me tell you about my system:

Asus A8N-SLI Premium
AMD X2 4400+
2Gb of Corsair TWINX1024-3200LLPRO (512mb X 4)
XFX 7800 GTX Overclock(only 1 card for now)
WCD 74Gb Raptor

The system works great , but I of course want to push it. I did some reading on overclocking the X2 4400 and decided to overclock mine. So here is my problem

iF i leave the memory settings on AUTO I can push the CPU as far as I want, but the system defaults the memory to 333Mhz so my 3dmark05 scores drop by almost 1000 points. As soon as I try to set the memory manually the system wont boot. The best I got it to do was

CPU = FSB 212 @ 1.375 Volts
Memory = 2.3.2.6 @ 2.9 volts (at lower voltages the memory is not stable)
which gives me 2.33Ghz, but this is nothing compare to what others have gotten.

Any ideas? I had the same menory in my oder system and I was able to get it to 220Mhz at 2.8volts and timings of 2.3.2.5.

Thanks in advance for your help guys.

More voltage for the cpu !