What can we expect from Phenom 2 965 C3?

[m411b]

I would like to just have a much cooler running chip than anything.

[mäggäri]

I would like to just have a much cooler running chip than anything.

So you're saying that a cpu which goes max around 60'C in load, with a "standard" Scythe Mugen2 for an example - isn't cool enough? What should it be then... -60'C on air?

What's in all the temperature fuss time after time... I mean, if the component X is designed to operate in like 300'C. So what, only thing that REALLY matters is power consumption (how energy efficient it is) and the real power in the chip. Aka how it performs against other contenders in the market.

FFS. Figure it out one time more, what-the-f00k does it matter if a temperature reading is 50 or 5000 or 300? It's not made out of water, so it doesn't boil =D Does a bigger reading make ppl cry? Not.

[Titan7171]

So you're saying that a cpu which goes max around 60'C in load, with a "standard" Scythe Mugen2 for an example - isn't cool enough? What should it be then... -60'C on air?

What's in all the temperature fuss time after time... I mean, if the component X is designed to operate in like 300'C. So what, only thing that REALLY matters is power consumption (how energy efficient it is) and the real power in the chip. Aka how it performs against other contenders in the market.

FFS. Figure it out one time more, what-the-f00k does it matter if a temperature reading is 50 or 5000 or 300? It's not made out of water, so it doesn't boil =D Does a bigger reading make ppl cry? Not.

lower voltage will give a lower temp and is going to give you a better chance at higher overclock.

My cpu will not run at 4.2ghz at 60c but it will run at 4.2ghz at 38c.

[freeloader]

So you're saying that a cpu which goes max around 60'C in load, with a "standard" Scythe Mugen2 for an example - isn't cool enough? What should it be then... -60'C on air?

What's in all the temperature fuss time after time... I mean, if the component X is designed to operate in like 300'C. So what, only thing that REALLY matters is power consumption (how energy efficient it is) and the real power in the chip. Aka how it performs against other contenders in the market.

FFS. Figure it out one time more, what-the-f00k does it matter if a temperature reading is 50 or 5000 or 300? It's not made out of water, so it doesn't boil =D Does a bigger reading make ppl cry? Not.

Overheating/overvolting the chip will also lower it's life expectancy. But who around here has a chip longer than two or three years anyhow?

[Titan7171]

Overheating/overvolting the chip will also lower it's life expectancy. But who around here has a chip longer than two or three years anyhow?

2-3 years? try 7-8 months

Phenom 2 965 BE ordered 8/25/09

Phenom 2 940 BE ordered 1/13/09

Phenom 1 9950 BE ordered 7/08/08

Athlon x2 6400 windsor ordered 12/22/07

AMD Opteron 165 Denmark ordered 4/19/07

[Mech0z]

Cant decide if this c3 or i5 is the right choise :s if I just knew if BD will be socket AM3 then I would gladly buy this and a 790FX and then in a few years upgrade to BD, but other than that the i5 just seems like a better choise with water as I would get 200-400 more Mhz from what I can see, is this totally wrong? Price is not that much different

[Oliverda]

AMD's Phenom II X4 965 BE C3 Revision minireview

[mäggäri]

Overheating/overvolting the chip will also lower it's life expectancy. But who around here has a chip longer than two or three years anyhow?

This is one of the most stupidiest claims in the universe. Who actually cares if the chip lasts 30years instead of 100years?

Example:
I've had socket-a Duron 1400 @ L2-cache mod (axp mod) and around 2360MHz on AIR, for about 6½years now in one computer. For the first 2years it was running on water, a little bit over 2500MHz. :P
And this sure is hardware which should be in the museum already.

Overheating/overvolting the chip will also lower it's life expectancy. But who around here has a chip longer than two or three years anyhow?

Yes that's true, I was sure some theoretic smartass will come up sooner or later I was earlier talking about the temperature just as a number. IF the X thingy works rock solid 24/7, who cares about the temperatures... It's just a temperature, nothing else.
I've noticed that ppl go crazy when they see a temperature near 100'C, and I bet it's due to the water boiling point in normal pressure... so ppl think instantly that OMFG 100 =DD, but they forget about the fact that it's kinda metal they're warming up. :b

And what comes to overheating the component X, there's no such thing as overheating, when we are under the value of the announced max temp. Between this range, even if the max is 5000'C, who cares if the reading is 50/892/4823'C - only thing that matters is being 24/7 stable. I just love being a realist, not some mumbo-jumbo professor. End of my opinion.

[richierich]

AMD's Phenom II X4 965 BE C3 Revision minireview

They were able to boot into Windows 7 RTM 64-bit @ 4GHz with air cooling but wasn't stable. Interesting.

[Oliverda]

They were able to boot into Windows 7 RTM 64-bit @ 4GHz with air cooling but wasn't stable. Interesting.

Whereas we had to resolve to somewhat arctic temperatures to squeeze 3.8GHz stable operation out of the previous revision, the C3 revision manages 4GHz at only a minor core voltage increase and air cooling at ambient room temperature.

We were able to get into Win7 at 4.1 GHz but it would be exaggerating to claim stability at that speed. At 4.2 GHz, the CPU still reached the Windows7 desktop but crashed.

If I'm not mistaken it was stable at 4GHz.

[ajaidev]

Whereas we had to resolve to somewhat arctic temperatures to squeeze 3.8GHz stable operation out of the previous revision, the C3 revision manages 4GHz at only a minor core voltage increase and air cooling at ambient room temperature.

We were able to get into Win7 at 4.1 GHz but it would be exaggerating to claim stability at that speed. At 4.2 GHz, the CPU still reached the Windows7 desktop but crashed.

If I'm not mistaken it was stable at 4GHz.

+1

4Ghz stable on 64bit win is good to say the least. 955 based on the C3 would be my choice 3.2Ghz would be good enough.

It is either that or wait 3-4 months for the x6 to come. If priced similar to i7 920 that could be a wonderful buy...

[FlanK3r]

yes, its good think, 4 GHz is possible stable with this revision. Example, most reviewers had x4 955 stable only at 3.7 GHz in x64. But we are tweakers , my 955 C2 is 3850 MHz x64 stable . So, if lostcircuits gewt about 4 GHz, think, itsd possible for us to get 4 GHz stable :-)

[crazydiamond]

yes, its good think, 4 GHz is possible stable with this revision. Example, most reviewers had x4 955 stable only at 3.7 GHz in x64. But we are tweakers , my 955 C2 is 3850 MHz x64 stable . So, if lostcircuits gewt about 4 GHz, think, itsd possible for us to get 4 GHz stable :-)

exactly! i think we will get the magic 4ghz this time around

[chew*]

+1

4Ghz stable on 64bit win is good to say the least. 955 based on the C3 would be my choice 3.2Ghz would be good enough.

It is either that or wait 3-4 months for the x6 to come. If priced similar to i7 920 that could be a wonderful buy...

Ahhh but most likely the 955 would not be a BE which means you will be limited in your OC configuration trying to match ram, NB and cpu

[FlanK3r]

hm...a bit yes, 955 is not so whorse with 3.2 GHz and 16x multiplier. NB multiplier is open think donw+up for non BE.

[Smartidiot89]

hm...a bit yes, 955 is not so whorse with 3.2 GHz and 16x multiplier. NB multiplier is open think donw+up for non BE.

Nope NB multi is also locked, so you can't increase it on a non-BE part.

[chew*]

Yep NB multi is locked at 10.

At about 270 to 280 fsb......you have 2 options........2700 NB or 2500 NB.......now you also have 2 options for ram dividers.......about 1400 or 1800.......how many of you have sets of ram that will do 1800 24/7 guess your stuck running what the divder says you have to run. Of course you can always run 300 fsb.....I doubt you will see 3000 NB for 24/7 use....so you settle for 2700NB and 1600 ram. Problem is how many "devices" are going to cooperate that high, USB, sata, ethernet....

I know my little micro atx asus will not boot if they are enabled at 300 HTT.

So much for tthe 24/7 OCe'd rig.

I like my freedom personally.

[Titan7171]

everything in this thread was done on regular air in the summer up to post #8 then I started using the ac to cool the cpu. Ambiant results are whats important here though.

http://www.xtremesystems.org/forums/...d.php?t=233002 but on 32bit windows.

It seems pretty close. I guess at this point all I could do is try it out and see what it does myself in my particular setup.

That link showing the mini review is not working so I didnt see what they ran and at what speeds?

[G.Foyle]

My first sample does ~3950 stable on Win 7 x64 (I would only use x86 if it was WinXP and I was benching super pi).
Currently testing the second one. Idle power consumption seems to be lower than C2 but I don't have the numbers just yet.

[crazydiamond]

awesome man keep us updated! =)

[Kampele]

Asmola testing PII 965 C3

(First boot)

http://murobbs.plaza.fi/1703841856-post46.html

I hope i post this correct thread.

[FlanK3r]

whau, congr. Asmola, must watch at overclock.net forum

[richierich]

It seems that 4GHz is very possible w/ 1.5vcore

[Oliverda]

1.50V is too much.

[Titan7171]

1.50V is too much.

Yea its a little high We need to get these in our hands though its almost a given we can do better than the mini review