Sign in with
Sign up | Sign in
Your question

Phenom II X3 720 Black Edion on AMZON _ it's can OC to 4.2GHz on Air

Last response: in CPUs
Share
February 20, 2009 3:41:02 PM
February 21, 2009 1:58:52 AM

Nasty speeds. Given the price, performance and overclocking, this chip has basically killed any legit reason to buy a 45nm Intel dual core. Im sure ill be attacked by all the rabid fanbois who think the E8400 is the best chip ever.
February 21, 2009 2:25:24 AM

Voltage please?

I'm wondering how long it could last with this high clock speed.
Related resources
a b à CPUs
February 21, 2009 2:26:17 AM

Yeah but the triple has a lumpy idle due to the high rise cam lobes.

You gotta put an extra leaf spring on the front right hand side of the box or they shake a bit under load ...

Has anyone got some comparison data on these compared to the E8400 yet?

Maybe both running some of the games at 4Ghz??

Might be interesting ... I think there would not be much difference.

Trouble is the E8400 die is much smaller than the Phenom II die (all of the triples and quads use the same die).

Intel wins on manufacturing cost one would assume ... all other things being equal you can get more useable silicon from the same wafer.

So AMD has to flog more of these.

Should claw back a bit of market share under these tough times.

a b à CPUs
February 21, 2009 2:43:02 AM

I'd be curious to see a comparison of temps and voltages between this and the Intel duals. If these consistently hit 4+, it does mostly kill the reasons to buy an E8000 series, but I'm not convinced that they hit 4+ reliably at a reasonable voltage yet.
a b à CPUs
February 21, 2009 3:18:31 AM

1.65V is uncomfortably high for a 45nm part, IMHO, although it's hard to say for sure. I can't find voltage specs on AMDs website, but I would be surprised if 1.65V were within spec.
February 21, 2009 4:18:56 AM

I think I saw 1.55v is the top of AMD 45nm spec.
February 21, 2009 4:49:39 AM

Hmmm...those voltages are absurd. Considering Intel 45nm quads "shouldnt" be taken past 1.4v for daily usage, I cant believe the AMD 45nm parts can survive at 1.65v.
February 21, 2009 4:54:35 AM

I'm pretty sure I've seen that the HK/MG that Intel uses at 45nm doesn't stand high(er) voltages well, which would explain the higher spec on AMD's. Though they're both 45nm, their processes are very different right now.
a b à CPUs
February 21, 2009 4:56:01 AM

That's why I was unsure, but I'd be curious to see some official spec sheets.

I'd also be curious to see power usage and heat for 4GHz or more for the AMD triples vs the Intel Duos. Even if they can take the 1.5v and more without damage, I'd be surprised if they used as little power as the Intels.
February 21, 2009 12:35:20 PM

I think 1.475 is the top of spec.

Damn, 4.2ghz on air, just another reason to get water cooling, Now if only they could make a 12mb L3 cache phenom II, damn that would be nice.

Is he using one of the AM3 mobo's with the SB750 because he might make it to 4.5 or 4.6 using a SB750.
February 21, 2009 1:25:52 PM

Quote:
4.319Ghz at 1.65v, 3.8+ghz at 1.55v, 3.6ghz at 1.44v (stock is 1.325) on what he says is a mobo that needs higher volts.


sounds a bit higher than required to overclock the Core2's. i think the cpu can handle it as most cpu's can if you keep them cool but i see short lived motherboards. wouldnt discourage me from doing it though lol
a b à CPUs
February 21, 2009 2:19:56 PM

Quote:
http://www.legitreviews.com/article/905/2/

4.6ghz on dry ice at 1.9v

They also got the 810 to 3.84ghz on only 1.46v on air, which is a good deal higher than most have managed and pretty incredible for a 2.6ghz quad core.

http://www.legitreviews.com/article/898/14/


I really honestly could not care less what a CPU can do on dry ice and insane voltage.

It's a matter of the maximum realistic, day to day overclock on air or relatively mild water cooling that matters.
February 21, 2009 2:29:59 PM

Well I don't think people that overclock their CPU's and have high end video cards worry to much about the power a system would use. Unless it would be alot more like enough to have to get alot larger power supply.
February 22, 2009 12:37:12 AM

3.8ghz is still a good OC considering the chip is 2.8ghz stock.
a b à CPUs
February 22, 2009 1:28:49 AM

Water cooling is the most reliable cooling for the price ... other than a big aftermarket heatpipe cooler.

What are people getting these up to with a decent water cooling rig?

Forget ice and LN ... that tells me nothing useful.

February 22, 2009 2:26:58 AM

Are these tests being done on AM2+ boards or new AM3 boards?
February 22, 2009 9:00:27 AM

What I"d want to know is how well it would overclock with a Zalman cooler without changing the voltages one bit. I haven't tried overclocking since the Williamette days and all I'd be interested in is getting 3.4 to 3.6 with a nice air cooler.

I'll have to wait on the Phenom II 720 though. My Gigabyte 780G board died on me and I ran out to Fry's and spent money that could have gone to the 720 on a new board with SB750. At least that should overclock well when I do get one.
February 22, 2009 11:38:12 AM

I hope it's on the am3 boards since most of them use the SB600 chipset instead of the SB750. Which would mean you'd get better oc's on a 790 and SB750 AM2+ board.
a b à CPUs
February 22, 2009 8:52:08 PM

Quote:
I agree completely - the dry ice and LN benchmarks are simply...benchmarks for people who want to benchmark cpu's on dry ice or LN.


On the other hand, if you are oc'ing a system to 40% over stock, you sure aren't worrying about a few extra watts - so i dunno why you are so interested in that. Heat is a non-issue from what i've heard. The Phenom II's have such a huge surface area I haven't seen anyone complain about temps yet with air, it's usually running out of volts. From the reports I'm seeing, the most reported wall with 720 BE overclocking is failing to boot windows even with higher voltages that are POST'ing.

It's probably a variety of stuff and it's almost certainly going to improve over time. Of course very few will hit 4.3ghz with a 720 BE, but 3.8 seems to be standardising itself as the marker for what should be expected.


I care about wattage because lower wattage means lower temps and not as hard on the board. It isn't a matter of power bills or anything like that.
a c 127 à CPUs
February 23, 2009 4:10:14 AM

I for one am not comfortable with those voltages. Even if AMDs process is a bit better for taking that high a voltage. I worry about the mobo really with that as the mobo also needs to come into play and if it overheats and dies it could take the CPU and everything else with it.

Thats the only reason I am interested in the Core i7 920 even though I want to wait till the 32nms hit. Just because at stock it will hit 3.8GHz and on air which is plenty of power for me.

But still. 4.2GHz on air is much better than Phenom I did. At least makes Phenom II a threat to Core 2 and Core i7 which means I may be able to pick up a leway CPU upgrade (want the Q9650) for even cheaper in May.
February 23, 2009 5:14:54 AM

So would you say even if the temps are low higher voltages would be bad for the CPU?

(this is a question not a provocation (:< )
a b à CPUs
February 23, 2009 6:14:41 AM

bonanzaguy said:
So would you say even if the temps are low higher voltages would be bad for the CPU?

(this is a question not a provocation (:< )


Correct - above a certain level, voltage will damage the CPU regardless of temps.
!