AMD Phenom II In-Depth Performance Scaling Analysis

[Final8y]

Introduction:

Madshrimps (c)


As you probably already have noticed, Madshrimps did not publish an article regarding the Phenom II's performance in comparison to competitive products. There are two main reasons: on one hand, we only received our Phenom II sample a few days after the product launch, which means that most hardware reviewing websites already had an article regarding the Phenom II. Ours would just have been 'yet another' comparison and would only serve to confirm everything we already know. On the other hand, we had the possibility to test the widely hyped overclockability of the processor and check if the rumors about 6GHz were indeed true.

We were actually surprised by the excellent overclocking scaling of the Phenom II X4 940 on LN2! As you may or may not know, our retail (but sent by AMD) sample went beyond 6GHz quite easily and there was absolutely no sign of any coldbug. Unlike its predecessor, the Phenom II (also referred to as Ph-II) is indeed meant for overclocking ... well, extreme overclocking. In 24/7 configurations, it's still very difficult to reach 4GHz stable on normal air cooling, which doesn't match the current C2D overclocking capabilities.

These were our findings of our performance scaling article on AMD’s Phenom I:

"If you're planning to spend a couple of minutes on overclocking, you should go for the raw cpu megahertz overclocking ... which is fairly obvious. Everything else gives a nice boost, but that's it; no real extreme increases in performance. Note that it's important how you interpret the graphs: this is a performance scaling investigation, not an overclocking scaling investigation. Please understand that the given increases are expressed in comparison to a very low base frequency, not at stock speeds. For instance, the performance scaling of the HT Link frequency may seem very impressive in the 3D benchmark, but if you look more closely, you'll notice that 1GHz is almost enough to maximize performance."
(~ AMD Phenom In-Depth Performance Scaling Analysis)


We already had a peek at the overclocking performance scaling of the Phenom I. In the article, you were able to find out that the processor frequency had the biggest effect in performance, which is pretty obvious, and that overclocking the IMC frequency is actually quite interesting to get the absolute most out of your machine. Since we got a lot of positive feedback about the first article, we did the exact same for the Phenom II. In this write-up, you will find out what's the impact of each overclocking variable performance-wise.

Madshrimps

 

[lukext]

Thats a really good read. Interested to find out what AMD are going to offer in the way of new AM3 chips as Ill be upgrading in about sept/october time

Cheers for the link dude

 

[deshepherd]

Thats a really good read. Interested to find out what AMD are going to offer in the way of new AM3 chips as Ill be upgrading in about sept/october time

Cheers for the link dude

ditto ... at least it tells me that the way my MB refuses to set HTT to higher than 1.6GHz (even if I set it to 2GHz in BIOS) isn't an issue.

One point the article didn't highlight was that in the IMC scaling, which I assume is the NB scaling, then the dominant effect is probably that this is increasing the speed of the L3 cache

 

[dagwoood]

ditto ... at least it tells me that the way my MB refuses to set HTT to higher than 1.6GHz (even if I set it to 2GHz in BIOS) isn't an issue.

I was playing with the BIOS this morning and noticed that, even if you change the HT/NB frequency, when you open CPU-Z, it says the link is still running at the default of 1600MT. I found out though that if you change the NB frequency and open Everest, under the Overclock tab, it displays the NB running at the frequency I'd set in the BIOS (2000), but the HT speed still running at the default value.

So the NB frequency can be changed, but I haven't figured out how to change the HT speed to the corresponding value.

 

[Final8y]

I was playing with the BIOS this morning and noticed that, even if you change the HT/NB frequency, when you open CPU-Z, it says the link is still running at the default of 1600MT. I found out though that if you change the NB frequency and open Everest, under the Overclock tab, it displays the NB running at the frequency I'd set in the BIOS (2000), but the HT speed still running at the default value.

So the NB frequency can be changed, but I haven't figured out how to change the HT speed to the corresponding value.

That can be down to the motherboard & BIOS error not allowing it.

 

[deshepherd]

I was playing with the BIOS this morning and noticed that, even if you change the HT/NB frequency, when you open CPU-Z, it says the link is still running at the default of 1600MT. I found out though that if you change the NB frequency and open Everest, under the Overclock tab, it displays the NB running at the frequency I'd set in the BIOS (2000), but the HT speed still running at the default value.

So the NB frequency can be changed, but I haven't figured out how to change the HT speed to the corresponding value.

Yes, that's exactly what I see ... I've managed to adjust NB speed (got it up to 2400 after nudgign voltage upto 1.3) but HT remains fixed at 1600 .... actually, I have managed to make it lower - i.e. if I set it to 1000 in BIOS then CPU-Z/AMD Overdrive report as being at 1000! Maybe there's something in the BIOS that has the AM2 limit fixed?

 

[drunkenmaster]

generally a bunch of the am2+ mobo's won't allow changing the htt or nb multipliers(some mobo's more the htt but fine on nb) unless the cool and quiet type options in the bios are disabled.

Which is a shame truly and something that needs working on, as the downclocking and power saving a P2 can do are truly impressive. the power saved tbh is pretty significant especially if you've upped the voltage and speed.

I guess its because they change the multiplier and set the voltage to a predone setting. If you've overclocked though the lowest setting which is stable at default is now at a higher cpu speed and the voltage it "normally" drops to might of been stable at 800mhz, but unstable at the new speed of 1Ghz.

WHich is why someone needs to do a mobo that allows manual C&Q where you can set the speed and voltages for idle, low usage and max usage speeds which would negate the need to disable it.

 

[Bleek]

My mobo allows manual cool and quiet settings in the bios.

 

[deshepherd]

My mobo allows manual cool and quiet settings in the bios.

So does mine ... but it doesn't seem to be doing anything as its enabled but clk speed is still reported as full speed when cores are idle.

Is there a setting in windows that is needed as well? ... I remember with previous Athlons there was a CnQ driver that was needed

 

[ecat]

I don't think CnQ needs a driver but I did need to enable Windows power management.

 

[deshepherd]

I don't think CnQ needs a driver but I did need to enable Windows power management.

I'll take a look tonight at my power management settings. For now (while experimenting with overclocking) I'm quite happy with it being disabled!

 

[deshepherd]

I'll take a look tonight at my power management settings. For now (while experimenting with overclocking) I'm quite happy with it being disabled!

Sorry to resurrect this thread but I eventually got round to putting in some effort to sorting out CnQ on my new PhII 720 last night.

Firstly discovered that my "processor driver" was, I assume, still the default Microsoft WinXP driver dating from ?2005 ... doing an "update drivers" pulled in an AMD driver v1.0.0.0 from 2007. Via AMD site I found latest driver was v1.3.xx but while this claimed to install no change showed up in system hardware processor properties.

Then searching for CnQ on the AMD site I found a document on how to install it ... I originally thought this might be irrelevant as it dated back to the original CnQ. However it did contain the key bit of info ... Windows power management needs to be set to "Minimal power management" ... once I'd done that the 720 would throttle back to x4 multiplier (800MHz) and 1.15V.

N.b. I'm only doing multiplier overclocking so it should be "safe" to drop back to 4X and 1.15V. I suspect that if you're FSB overclocking then CnQ may well need to be disabled as if 4X is then substantially faster than 800MHz then 1.15V may well not be enough and I'm not sure that you can change this.