The “Break it Down” Blog

or

Introduction

I’ve been in the market for a new laptop recently and have been trying to decide between purchasing a new Core Duo system now, or waiting a few months for all the Core 2 Duo laptops to be out and then take my pick. A usage requirement I realized recently that I had no idea was important to me before was battery life. I want a fast machine (for software development and maybe a casual game a few times a month, nothing serious) with long battery life (around 8 hours), which seems like an impossible feat.

Luckily I ran across the ThinkPad series of laptops from Lenovo (used to be IBM). Not only can you purchase a high capacity 9-cell battery, but you can also buy what is known as an Ultra-Bay batter, meaning you pop the CD drive out of the laptop, and in it’s place you put another batter (3-cell). All together, with normal computer usage, integrated graphics and the 14″ screen, you can crack 12 hours of usage on a single charge.

The cost to the user here is a fairly heavy laptop with a huge battery sticking out the back.

With the ATI Mobility x1400 and the 15″ screen, that usage drops to around 8, right in the ballpark I wanted. In the T60p model, which can have the 1600×1200 15″ monitor and the ATI FireGL V5200, run times drop to around 4 hours, which is too little, so I backed off and went with the T60 model.

Core Duo versus Core 2 Duo Comparisons

Anyway in an attempt to better understand the differences between Core Duo and Core 2 Duo, I’ve been digging through some new comparison articles from PC Magazine and ExtremeTech that cover the differences between these old and new laptops. I read these articles trying to learn the chip performance/power differences between them, I didn’t care about the overall system performance (like high resolution gaming) because that has too much to do with other computers in the computer, not the CPU. Here are my conclusions:

First, off the comparison article is is tilted by the marketing team at Intel because they sent PC Mag a Core Duo T2600 to review against a Core 2 Duo T7600, basically a 2.16Ghz chip versus a 2.33Ghz chip. I would point out that there is a 2.33Ghz Core Duo chip called the T2700 that they could have included to do an apples-to-apples comparison, but instead, they sent along obviously two uneven chips to make the visual difference on the benchmark charts look better for the Core 2 Duo chip, and assuming you missed the note at the beginning of the article saying that there is a 8% Mhz difference between these chips to consider, you would think that the Core 2 Duo is faster, cause all the charts show it.

NOTE: The Core 2 Duo chip is faster, if only slightly, but not as much as the charts in these reviews suggest.

So given that information and scaling back the differences by about 8%, when you look at the charts for processor intensive tasks and battery life, the chips are very similar. Sometimes near identical, other times about 15% difference (Which I would attribute simply to the 4MB L2 cache). The Core 2 Duo chip may eek out a few higher points because of the double L2 cache size, but in the grand scheme of things, I would strongly doubt you will see any difference in performance between the two chips when compared Mhz to Mhz and using two laptops side by side for work, not running synthetic benchmarks.

Second, the power consumption is higher for the Core 2 Duo chips, again this is misleading because it’s a faster chip and the difference isn’t that huge. If we were comparing, again, Apples to Apples, the results I imagine would be near identical except for the minor difference in consumption the extra 2MB of L2 causes (and 64-bit instruction processing if you want to count that).
From these results the nearest I can tell is that these two processors, the Core Duo and Core 2 Duo, are very similar. The Core 2 Duo brings a 4MB L2 cache and 64-bit support. But if you remember the hub-bub when the MacBook Pro launched, it turns out the Core Duo chip supports 64-bit already but has a control code turned off on the chip to disable it. So I think these two chips are infact a lot more similar than Intel wants us to believe.

Not Comparing Apples-To-Apples

What solidifies my belief of this even more, is the fact that Intel conveniently sends along a Core Duo 2.16Ghz chip to compare to a Core 2 Duo 2.33Ghz chip. Marketting teams aren’t dumb, they know 85% or more of folks don’t read every word in a benchmark article, they just skip directly to the charts and graphs so they can visualize the speed increase. I would also go so far as to say, have a look at every one of the charts in the PC Mag review, it never once refers to the chip by their speed (2.16Ghz and 2.33Ghz) but instead refers to them by their names. If you didn’t read the 2nd page, Merom, you wouldn’t know that there was a Ghz speed difference between these two, which seems to be the point:

What bothered me so much about this presentation of the chip “comparisons” was that it is almost intentional misrepresentation. This is like Car & Driver magazine getting two Honda Accords from Honda to test against each other: 2005 model and 2006 model. Except, only mentioned in a single paragraph at the beginning of the article, is that the 2005 model is a V4 Sedan and the 2006 model is a V6 Coupe. Assuming you miss that part, as you skim through the comparison of the cars you’ll think that the 2006 model is say 60% faster or 100% more powerful, then the real difference when comparison the same models, just different years is more like 8%. Then at the end of the article PC Magazine raves about the Core 2 Duo instead of restating the obvious difference of why the numbers were so screwed, so now people can run off talking about how amazing Core 2 Duo is with 1/2 the correct information.

I don’t mean to be ragging on PC Magazine for this review, it’s just very frustrating to me because I feel, as a consumer, that I’m being actively ‘duped’ in a very subtle way so I can buy into some Intel marketing machine. I don’t necessarily mind that, but don’t try and sneak a fast one by me, grrr!

Conclusion

I would have to say for most normal people, using a laptop for day to day work, these chips are going to feel identical and waiting or buying now doesn’t matter. For crazy hard-core gamers, there can be a different benefit for waiting for a Core 2 Duo machine, and that is the simple fact that it may be packaged with a more advanced mobile graphics processor like the ATI x1800. So while the chip difference doesn’t realize much of a gain for you, the GPU will make a big difference.

If you are curious, I ended up purchasing a ThinkPad T60, with the 15″ Flexi screen, 7200RPM drive and 2GB of ram along with the 9-cell and Ultra-Bay battery. My goal is to reach roughly 8hrs of run time during software development, and then I will be happy.

In closing I would have to say that you are fine buying a mobile computer using either series of these chips. If you don’t need to buy now, sure wait and get a Core 2 Duo system. If you are in the need of a laptop, take solice in knowing that there isn’t much different with the current Merom generation of Core 2 Duo chips, so don’t sweat it and buy your laptop. (Digg this)

Update #1: Here’s a much better comparison over at Anandtech.