PDA

View Full Version : Mobile CPU Wars: Core Duo vs Core 2 Duo


MacBytes
Aug 20, 2006, 04:22 PM
http://www.macbytes.com/images/bytessig.gif (http://www.macbytes.com)

Category: Processors
Link: Mobile CPU Wars: Core Duo vs Core 2 Duo (http://www.macbytes.com/link.php?sid=20060820172219)
Description:: Unlike its desktop predecessor, Core 2 Duo comes from the same genealogy as the Core Duo. Despite the similarities in name and in architecture, there are some fairly major differences between the two CPUs, some of which won't become apparent until next year.

Compared to the desktop Core 2 Duo (Conroe), the mobile version is architecturally no different. Obviously clock speeds (both CPU and FSB) are lower because these things will be going in notebooks where power consumption is more of a concern, but other than that the architectures are identical.

Compared to Yonah, Merom has some very clear advantages; on the surface the larger L2 cache is responsible for the 140M increase in transistor count, but architecturally the improvements extend far beyond that. You can get the details from the table above or from our previous articles on Intel's Core 2 processors, but simply put Merom is wider and slightly deeper than Yonah. The slightly deeper pipeline helps increase clock speeds on Merom (which will bump performance a bit), but the added decode and execution width will increase overall performance.

Not listed in the table above are the improvements to the cache subsystem and memory accesses on Core 2 Duo. Merom features more aggressive prefetchers than Yonah, as well as Intel's Memory Disambiguation technology that allows for out of order loads. In other words, not only is Merom able to operate on more data at once, at a faster speed, but it can also get access to that data quicker.

Posted on MacBytes.com (http://www.macbytes.com)
Approved by Mudbug

DMann
Aug 21, 2006, 12:57 AM
http://www.macbytes.com/images/bytessig.gif (http://www.macbytes.com)

Category: Processors
Link: Mobile CPU Wars: Core Duo vs Core 2 Duo (http://www.macbytes.com/link.php?sid=20060820172219)
Description:: Unlike its desktop predecessor, Core 2 Duo comes from the same genealogy as the Core Duo. Despite the similarities in name and in architecture, there are some fairly major differences between the two CPUs, some of which won't become apparent until next year.

Compared to the desktop Core 2 Duo (Conroe), the mobile version is architecturally no different. Obviously clock speeds (both CPU and FSB) are lower because these things will be going in notebooks where power consumption is more of a concern, but other than that the architectures are identical.

Compared to Yonah, Merom has some very clear advantages; on the surface the larger L2 cache is responsible for the 140M increase in transistor count, but architecturally the improvements extend far beyond that. You can get the details from the table above or from our previous articles on Intel's Core 2 processors, but simply put Merom is wider and slightly deeper than Yonah. The slightly deeper pipeline helps increase clock speeds on Merom (which will bump performance a bit), but the added decode and execution width will increase overall performance.

Not listed in the table above are the improvements to the cache subsystem and memory accesses on Core 2 Duo. Merom features more aggressive prefetchers than Yonah, as well as Intel's Memory Disambiguation technology that allows for out of order loads. In other words, not only is Merom able to operate on more data at once, at a faster speed, but it can also get access to that data quicker.

Posted on MacBytes.com (http://www.macbytes.com)
Approved by Mudbug

Wonder if Core 2 Duo Extreme will eventually be worked into Merom chips for MacBookPros.

After G
Aug 21, 2006, 02:38 AM
Read the article. Not much of a war, when Core 2 is only 15% max ahead for the same power consumption. Not worth upgrading for now, but might be nice to drop in my Mini when prices go down and Yonah is slow (which it isn't, for now).

DMann
Aug 21, 2006, 02:48 AM
Read the article. Not much of a war, when Core 2 is only 15% max ahead for the same power consumption. Not worth upgrading for now, but might be nice to drop in my Mini when prices go down and Yonah is slow (which it isn't, for now).

I did....... I agree, it would not yet be worth the upgrade, however,
from what I have read about Core 2 Duo Extreme, more speed and
more cache just might perk the Merom line up a bit.

freiheit
Aug 21, 2006, 05:32 PM
15% faster at the same clock speed, though. The Core 2 Duo goes to higher clock speeds then Core Duo (granted, at the cost of battery life). Drop a 2.4GHz Core 2 Duo into a system currently running a 2GHz or 1.83GHz Core 2 Duo and you're going to see a nice speed improvement.

This makes me very eager to see the next revision of the iMac -- a system which will not be battery-life dependent but will make good use of Core 2 Duo's improvements in all other areas. The current crop of iMacs seem to be pretty much on par with my 2GHz G5 PowerMac (when both are running native code). A 2.33GHz or faster Core 2 Duo iMac ought to blow the door off my PowerMac.

Come on Apple, make the announcement!

daveL
Aug 21, 2006, 07:13 PM
Read the article. Not much of a war, when Core 2 is only 15% max ahead for the same power consumption. Not worth upgrading for now, but might be nice to drop in my Mini when prices go down and Yonah is slow (which it isn't, for now).
As the author noted, the application benchmarks test the whole system, not just the CPU. Since disk I/O comes into play, you can't use the benchmark results as a CPU performance comparison. Having said that, I think the interesting thing to note is just how little CPU performance comes into play during most day to day computer usage. In CPU-only benchmarks, the Merom is a good 20% faster than Yonah, but that doesn't translate into a 20% boost for a majority of real world applications.

I really think the significant feature for Merom is 64bit. When Leopard comes around, you'll see some additional performance gains. None of the benchmarks used in these tests were 64bit. Although 64bit doesn't get you much on many CPU architectures, it does on x64.

Just my take.

bryanc
Aug 22, 2006, 09:48 AM
I really think the significant feature for Merom is 64bit.

Absolutely! This is the *key* to the importance of this update. Going to 64 bit could reasonably be expected to result in slower (clockspeed), more power-hungry chips *cough*G5*cough*, but the merom is neither. This is a *major* advance in CPUs.

And you're absolutely right, until Leopard ships, the significance of this upgrade will be lost on many. But, if I buy a new laptop this year, I want to be able to run the latest software on it next year, so I don't want to be stuck with a 32 bit CPU when the transition to 64 bit software starts to happen.

Most of you are too young to remember the shift from (8 bit) 8088 to (16 bit) 80286 CPUs, and then the shift from 286s to (32 bit) 386s after that. But it wasn't long after either of these transitions that we started seeing software that required the current CPUs. The only reason we haven't seen a requirement for 64 bit CPUs (outside of scientific computing) is that the costs (both financial and technological) of 64 bit systems have not made them a compelling option over existing 32 bit CPUs. This has been especially true on the Windows side, where a different version of the OS is required to run in 64 bit mode, and drivers for this version of the OS are much less widely available.

With Leopard about to launch, with full 64-bit capabilities, including support for all existing drivers (how did they *do* that?!?), the value of 64-bit CPUs on the Mac side is about to take off.

So I eagerly await merom-based MBPs, and I'd buy one even if it was *slower* and *more power hungry* than the yonah-based systems. The increased speed and battery life are gravy.

Cheers

freiheit
Aug 23, 2006, 07:24 PM
With Leopard about to launch, with full 64-bit capabilities, including support for all existing drivers (how did they *do* that?!?), the value of 64-bit CPUs on the Mac side is about to take off.


Just a shot in the dark, but they could have a form of Universal binary containing 32-bit and 64-bit forks, and depending on which CPU is detected one or the other runs.

Or they could be using a form of "thunking" which was comming in the 386-Enhanced days of Windows 3.x and 95 as well as OS/2 which translates 32-bit code into a 64-bit address.