- Back to homepage
- Related articles
  Related articles

Intel Core 2 on 45nm: Performance, Overclocking, Power Usage

CPU by piotke @ 2007-10-29

Intel is launching their successor the popular Conroe CPU, build on 45nm manufacturing process, it boasts reduced power consumption and has 50% more L2 cache. The first product out the door is a quad core beast dubbed QX9650. We take this new creation through its paces, comparing performance, power consumption and venturing into overclocking land, where sub zero cooling is the norm.

Overclocking

Overclocking

Before the official launch of the Yorkfield Intel had some samples out “in the wild” these showed promising overclocking results, some reaching up to 4.6Ghz with air cooling! (+53%) So our hopes were high when we received this engineering sample from Intel.

We mounted the Intel standard heatsink, left the vcore setting at default in the BIOS and raised the FSB until we noticed instability. Intel reports default BIOS for the QX9650 at 1.25v, the Gigabyte X38 board was more generous, providing 1.33v at default setting. Even with the extra juice the maximum overclock was 3.3Ghz. Time to raise the vcore, at 1.65v vcore the CPU ran stable at 3.85Ghz. Do note that we are still using the stock Intel cooler!

Madshrimps (c)

Overclocking on air

Time to take it a step further:

Madshrimps (c)

Gigabyte offers an impressive vcore range on their X38 DQ6 motherboard, but to be able to cope with the extra heat we need to go more extreme, air cooling won’t do, even water cooling will run into troubles. So we installed an Asetek VapoChill LS, this compressor based cooling system works pretty much like a refrigerator, but instead of cooling a large area we focus on a small surface, size of the CPU heatspreader. At idle the VapoChill LS can reach operating temperatures as low as -60°C.

A bit of info the voltage control in the BIOS of the Gigabyte X38:

CPU voltage, adjustable from 0,5 Volt up to 2,35 Volt , and even then there are some options to rise it a bit extra.

DDR2 voltage, can be raised in step from 0,05 V up to 1,55 V. this means that you can go as high as 1,8V (default) + 1,55 = 3,35 Vddr2. Deadly for most memory modules. So we didn't push higher than 2,5 Volt in total, which is already high.

Chipset voltage can also be raised with + 0,2 up to +0,35 volt.

We gradually raised the vcore in small steps until we reached 2v, which is insanely high for a 45nm CPU. At this voltage we could run stable at an impressive 4.9Ghz with all 4 cores enabled.

Madshrimps (c)

The maximum OC resulted in a SuperPi 1M calculation in less then 9,4 seconds, an impressive score without a doubt. The memory was running at only 452 MHz, so there was still room for fine-tuning.

Madshrimps (c)

Overclocking on Single Stage cooling

Since overclocking is limited to the slowest factor, with 4 cores, the “worst” core will determine the maximum stability; we disabled 3 cores and continued our overclocking tests with only 1 core. The results were the following:

Click for larger image.

Reaching 5191Mhz SuperPi 1M finishes after only 8.9 seconds, which put this score in TOP 10 bracket of the world’s fastest SuperPi 1M scores.

Let’s take a look at how the performance scales and how much power this overclocked system using ->

Comment from Sidney @ 2007/10/29

Reading other reviews, it would seem the engineering sample tested at [M] requires more vcore than others. 4Ghz quad would now be common speed; no bragging right unless you see 5Ghz.

Comment from jmke @ 2007/10/29

yes that seems to be the case, but no faulting engineering samples, they are supposed to run without fault at rated speeds with default vcore, which this QX9650 did

Comment from Kougar @ 2007/10/29

2.0v for 4.9GHz? Would have loved to see what that did to the power consumption figures for that CPU!

Regarding the discrepancy with your power chart, I think that has something to do with the physical properties of the chip design. I suspect that at very high frequencies there is a thermal threshold that once neared the leakage increases dramatically. After which it will quickly reach the point where the chip ceases to function or function stably since the increased leakage raises the heat, and the heat only further increases the transistor leakage in a self-fulfilling cycle. I don't have any real proof other than my own experiences with my Q6600...

I am curious, I notice from that CPUZ image the ES QX9650 uses 1.20v at 3GHz. My own Q6600 does the same... so how far can you drop the voltage and have the QX9650 remain stable at 3Ghz? I got a Q6600 down to 1.5v, but somewhere below that point my Q6600 will show errors. Gigabyte unfortunately lacks most of the FSB voltage tuning ASUS boards offer, as some members on the XS forums claim to have reached 1.10v for 2.5-2.8Ghz speeds for Kentsfields. Would be interesting to note what effect the smaller process size and change in transistor materials would have on this for Penryn.

Comment from jmke @ 2007/10/29

Quote:

Originally Posted by Kougar

2.0v for 4.9GHz? Would have loved to see what that did to the power consumption figures for that CPU!

You can

last chart on this page: http://www.madshrimps.be/?action=get...&articID=6 36
386W vs 210W stock

Comment from Kougar @ 2007/10/30

Ah, nice! I had completely missed that, staring me in the face.

I guess I need to take more power measurements since the data I have on my Q6600 includes my video card...

Why not extend that same chart a bit more to the right though, and undervolt that puppy?

Comment from CFKane @ 2007/10/30

I'm a little surprised that you're talking about a discrepancy in the chart while you're mentioning the changed cooling solution in the same sentence. The die temperature is one of the most important factors for the CPU power consumption and if you switch to a solution which removes the heat more efficiently, you should expect reduced power draw even with a higher clock and voltage.

That's also the reason why the maximum current in the electrical specifications for CPUs significantly exceeds what you would get from dividing the TDP by the core voltage. It's given for the maximum die temperature, which you will (hopefully) never reach in a real world situation.

Bear that in mind when testing or comparing CPU power consumption: The room/case temperature and cooling solution have a major influence and the die temperature at a certain load is an interesting figure to report along the power draw (sadly missing in most reviews).

Comment from jmke @ 2007/10/30

Thank you CFKane for you post and welcome to the forums

since the temperature was the only large difference between the two settings we were not doubting that it was indeed the lower temperature which was causing the lower temps; but I've not seen any article on the web discussing this aspect of the power consumption... hence were a bit hesitant to include that statement.

Comment from jmke @ 2007/12/07

we also got word back from Intel explaining the power usage at different temperatures:

Quote:

Matty @ Intel:

Yes, the power consumption is reduced when the temperature of the processor is lowered.

There are many things that happen in a CPU when the temperature is changed and to elaborate further on the processor specific causes we have to look at the origin of the power consumption. We can divide the total consumed power into two main parts, static power (Ps) and dynamic power (Pd).

The static power consumption is what we usually call the leakage. In an ideal transistor, it should completely shut off the channel between the source-drain, gate-source and gate-drain. Transistors are far from ideal, and the current leaks between these parts and the substrate of the processor, and this is heavily dependent on the temperature.
For example, going from room temperature to 85C (~60C difference) increases the leakage power by a factor of more than 50. Thus, reducing the temperature with the same amount will make a huge impact on Ps.

Dynamic power consumption is emitted during the short amount of time that the transistor switches. Lower temperature reduces the resistance in the processor which results in shorter delay/faster switching of the transistors. Shorter delays and less noisy signals also reduce Pd.

I hope this explanation give you some clarity to the relation between power consumption and temperature. This can even be seen with air cooling: The power consumption is lower just after a load is applied compared to after a while when the temperature has levelled out, even though the load is the same.

All information and graphics contained in Madshrimps are sole property of the Madshrimps crew and may not be reproduced or copied in any manner without written permission from us.