• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** Official Opteron 170 Ocing Thread ***

Hesky82 said:
with regards to your overclock?

what frequency did your cpu top out at with the stock cooler then? and how much mhz did you gain when you put the zalman on?

I could do 2.9ghz but with major temp issues when priming with the stock cooler, i'm still at 2.9ghz but the temps are still a little high for my liking when small fft’ing in prime95.

it helped my temps, but not so much my overclock. I'm thinkin water for 3ghz

Just to note I used as5 on the stock cooler also.

I never used my stock HSF.I'm comparing the temps to my mate pc who has the same case and case fans, but he is using the stock HSF.

My temps are are around 5-7c cooler using the zalman.
 
you cant compare cpu coolers used in 2 different machines simply on the basis of similar case cooling

does he also have the same mb, cpu, ambient room temp, vcore, thermal paste etc.....?

and there was me thinking you had done some back to back tests
 
Last edited:
of course it would be silly to compare if the two pc's where different.

He has the same setup as me.Bar the optical drives

I also have a akasa pro fan controller and monitor case temp with a temp sensor.
Guess what last week he got one.

ambient temps are roughly the same and yet the zalman keeps the cpu cooler.
I could put the stock one on mine but there would be not much point, the variables are to minute for me not to claim the zalmans superior cooling over the stock HSF:)

and yes I forgot he stole my as5 a few weeks back the readings were tested at stock vcore idle and on load.
 
so you can guarantee the on die temperature diode is identical in both cpu's, the ihs is properly seated over the core (we all know how variable amd are with this), there are still factors that make a difference even if the systems are "identical" test it on your own system and then I will trust your findings.
I'm trying not to be pedantic here but the only way you can get significant improvement with your clocks is to watercool, a questionable 5c increase just wouldnt cut it.
dont get me wrong I believe the zalman cooler is better just not good enough to make a difference, temps aside here.
 
Hesky82 said:
so you can guarantee the on die temperature diode is identical in both cpu's, the ihs is properly seated over the core (we all know how variable amd are with this), there are still factors that make a difference even if the systems are "identical" test it on your own system and then I will trust your findings.
I'm trying not to be pedantic here but the only way you can get significant improvement with your clocks is to watercool, a questionable 5c increase just wouldnt cut it.
dont get me wrong I believe the zalman cooler is better just not good enough to make a difference, temps aside here.


of course I can't ..and you know I can't ;) we both know this.
All that I am saying is that the zalman is superior to the stock HSF.


If you can shave a good few degrees off your load temps it gives you more headroom when increasing vcore etc...

Water cooling is an expensive option for 100 mhz gain.
I will eventually be fitting a 120 MM fan to my zalman for noise and better temps.When... I dont know, but the zalman is a fine cooler and cooling on the stock HSF is just not a feasible option IMO. For the price you can get a better after market solution.
 
Last edited:
if you forget about the scythe ninja, the zalman is the 'best' :)

yeah, a modded 9500 is the way forward

120mod7ek.jpg
 
you dont have to mod it, I'd only bother if you think the 9500 could do with being better at cooling, also the noise levels with the larger 120mm fan will be reduced slightly.
are you in fact asking how to mod this cooler?
 
Hi mate, as the fan blades are significantly larger it kind of requires you to hack away at the outer edges of the fan blades, easily done with a pair of metal cutters as copper cuts like butter. You then replicate the mounting holes of the existing fan onto the back of the new fans chassis with the outer frame cut away. It’s a fiddle and doesn’t look fantastic but the 120mm makes a big difference to noise/performance levels.
 
I'm still sitting here with my finger on the *buy* button, I'm just wondering how much these chips costs before Opteron-Mania pushed the prices?

How much were the 170's at their cheapest?
 
I'm reading conflicting reports on steppings?

Lopri@extremeoverclocking said:
CCBWE 0530TPMW
CCBWE 0546XPMW[Edit Big.Wayne] I have option to buy this one?

CCB1E 0550VPMW

The above steppings will most likely reach 2.8GHz and possibly beyond. (3GHz!!)

And the next best steppings:

CCBWE 0550UPMW (170, not 165) [Edit Big.Wayne] Or buy this one? (Overclockers.co.uk Fave!)
CCBWE 0546MPMW
CCBWE 0547TPMW
CCBWE 0544XPMW

Now trying to find out which is the better stepping between:

CCBWE 0546XPMW
CCBWE 0550UPMW

[EDIT] Probably go with the same one as Hesky82 and Agent WD40 are using, can't find out this 408L bit till Monday, dunno if I can wait that long as it will be just my luck that they all sell out before then!

CCBWE 0550UPMW hmmmmm. . . . .
 
Last edited:
Agent WD40 said:
Can you buy the CPU and then if it isn't the stepping your after send it back?

Agent.

If you haven't opened the package, then you have the right to return it, and it is usually possible to see the stepping from the outside of the package. I think most retailers would prefer it if you didn't, of course, and many would rather you asked them to check before hand.

Big.Wayne said:
I'm still sitting here with my finger on the *buy* button, I'm just wondering how much these chips costs before Opteron-Mania pushed the prices?

How much were the 170's at their cheapest?

I think they were under £300 at one point - but don't let that put you off too much. At that time they were a fantastic bargain. At the moment they're essentially the same price as the 4400+ for essentially the same chip. My 170 was bought from OcUK very recently, was an 0550 stepping (one of the best) and clocked to 2600 MHz (ie. FX60 speeds - a chip costing £750) with no trouble at all - I haven't tried any further yet. So, while they're no longer quite the bargain they were, they can still be pretty impressive!
 
mercyless said:
I'd like to throw in another stepping which i have now running at 2.9ghz using 1.48 volts stable , CCB1E 0550 VPMW

check steppings here

i'm using dfi expert and water cooling for testing . like the CCBWE 0546 XPMW stepping i reached 2.7 ish using stock volt, 2.8 with 1.4ish, but with CCB1E 0550 VPMW stepping 2.8 was reached using stock volts, booted 2.9 with
1.4ish volts but needs more for stability, prime95 etc
Hi mercyless,

thanks for that, you bugger! :p Now I have three times the problem picking one :)

CCB1E 0550VPMW
CCBWE 0546XPMW
CCBWE 0550UPMW

I'm reading up on all the forums now, my eyes are gonna melt soon, you guys do realise that I only just built my first A64 system about three weeks ago! (146 @ 2.9GHz) Now I'm picking through steppings of dual-cores, what the heck has happened and is there a cure!!! :eek:
 
Bito said:
PinkPig....OT, but what do you think of your G.Skill HZ m8 ?

What timings etc. are you running pls ?

I haven't really had a proper play with it yet - it does 260 MHz easily at stock timings, will have to have a proper go at it sometime soon. I'm possibly thinking I would have been better with the ZX, but not something of particularly concern really. Sorry not to be of more help!

I'm loving the system by the way, all working very happily! :D
 
Back
Top Bottom