Core Duo - Intel's Best CPU?

Bender writes "How good is Intel's Core Duo mobile processor? Good enough that Apple chose to put it in the iMac, and good enough that Intel chose to base its next generation microprocessor architecture on it. But is it already Intel's best CPU? The Tech Report has managed to snag a micro-ATX motherboard for this processor and compared the Core Duo directly to a range of mobile and desktop CPUs from AMD and Intel, including the Athlon 64 X2 and the Pentium Extreme Edition. The results are surprising. Not only is the Core Duo's performance per watt better than the rest, but they conclude that its 'outright performance is easily superior to Intel's supposed flagship desktop processor, the Pentium Extreme Edition 965.'"

Having used a Intel Dual Core for awhile ...

I have to say the Intel Dual Core Processor is quite impressive. It's fast enough to run just about anything I throw at it, and still keep chugging, but I believe that the article negects the fact that the dual core processor runs extremely hot vs other Intel processor. My old Sony VAIO never got as hot as my MacBook Pro does, and it is something that should be considered.

Re:Having used a Intel Dual Core for awhile ...

They have pathetic battery life after all the bragging Jobs did at last years WWDC.

The impression I got was that Jobs was trying really hard to avoid mentioning the battery life; the MacBook Pro was still in development and all they had were prototype models, so they actually didn't know what the battery life would be; they were guessing it should be "about the same" (as the PowerBook G4).

They are slow. My old G4 laptops kick the shit out it for media type tasks, about the same for single thread performance, and of course are slower for multi-threaded tasks.

Are you running all native applications? If not, it's not a fair comparison (and if you really need apps that aren't available natively yet, maybe you shouldn't have bought one yet). If you are running native apps, your experience seems to disagree with most reports I've heard.

It seems the speed most people are claiming for the MacBook Pros is due more to the faster video cards and the silky smooth desktop acceleration people weren't use to with their old G4 machines.

I'm really looking forward to this.

It is depressing to think that if Apple hadn't pissed off IBM that we could be running much faster/cooler dual core 970 PowerBooks right now.

If Apple hadn't pissed off IBM? When the G5 was released, Apple announced that they had 2GHz then, but would have 3GHz in one year. What was Apple supposed to do when that never happened? Just wait and hope that IBM figured out how to make something work?

Instead dual boot AMD Windows/Linux systems are looking like the only option for people who don't want to pay twice as much for x86 hardware.

Show me a laptop with comparable specs for half the price of a MacBook Pro. I think you're trolling.

Re:Having used a Intel Dual Core for awhile ...

In a unit like the iMac, where there is plenty of air flow around the case heat is less of an issue, and you are not likely to spend your time with your hand on it. In a portable heat is a big issue, since the under side has zero air flow when on a desk, and on the upper side where heat is going to be noticed your hands are resting, for large amount of time.

Also, heat can actually reduce the life-span of components.

Re:Having used a Intel Dual Core for awhile ...

How important is heat, really?

Extremely important.

It's blindingly obvious why it is important in laptops - not only because of battery lifetime, but also because the cooling assembly size and weight depends on TDP, and of course for user comfort considerations. Intel started a mobile CPU revolution with the Pentium M, so it's a little disappointing to hear that its latest successor doesn't improve further.

It's just as blindingly obvious why heat is terribly important for servers, where rack heat and power density has long been the limiting factor to packing more servers into less space.

On desktops, to me personally, heat is a premier consideration when choosing any chip. I have no need for something twice as fast as my current CPU if it consumes twice the amount of power. I expect better.

Depends

I would argue that the 8080 was. If you normalize for date/speed that is...

The 8080? No way! The Intel 4004 was smokin' hot!

At the time it was introduced, there was no other microprocessor that came close to matching it.

It was indisputably not only the best microprocessor Intel had produced to date, but the best microprocessor on the market.

Simply no contest. No argument. It superlative in every way, the fastest, the cheapest, the lowest in power consumption, the most advanced in architecture, the widest path. It was king of the hill, the top of the tree, the Cadillac of microprocessors, the ne plus ultra, it bestrode the world of microprocessors like a colossus.

The world will never again see the day when one manufacturer so dominated the microprocessor market that a single product had a 100.0% market share.

CoreDuo != Core Microarchitecture

It's not obvious from the article, but you can find it elsewhere on the internet (such as Intel's comment that the Core microarchitecture will provide 20% boost over CoreDuo). It is hinted at in the article with the following quote (emphasis mine).

If you've been hanging around here for a while, you may have heard us referring to Core Duo by its code name, Yonah, long before Intel decided to give it a somewhat confusing official name. ... In the case of the Core Duo, those CPU cores are massaged and tweaked versions of the Pentium M processor, familiar as part of Intel's Centrino mobile platform.

The new core microarchitecture, if you read the Ars Technica article in the previousl /. posting linked, was designed from the ground up and is similar to PentiumM in many respects, but is much more different than the CoreSolo and CoreDuo are.

Re:CoreDuo != Core Microarchitecture

The new Merom-based products (Conroe is the desktop version) were *NOT* designed from the ground up. The Ars Technica article repeated some Intel marketspeak that overstates the case. Merom is a major revision of Yonah, but is derived from the same code base. In fact, it is still technically a derivative of the P6 family that began with the Pentium Pro 10 years ago.

This is more than just a matter of semantics. The major micro-architectural features that defined the P6 are still present in Merom. The P4 architecture (may it rest in peace) was a brand new architecture -- Merom is not.

Re:CoreDuo != Core Microarchitecture

Yeah, they use both HDL coding and EDA (cad-like) tools to design most microprocessors. The designs are too massive to design them by placing each wire manually - they haven't done that for _several_ generations (1980s? - not sure really)

That's not to say there isn't a small army of design engineers at Intel and AMD who work with nothing but schematics - there are. Its just that most of the logic design work is done on the HDL coding level (with either VHDL, IHDL, Verilog, or some other tool). You only start dealing with schematics at a much later stage of development. Until then your designs are constantly changing and its infinitely easy/faster to change a few lines of HDL code than to re-write hundreds/thousands of wires and transistors.

I've worked at both Intel and AMD in the past and in both cases you could take the entire codebase for a processor (HDL, microcode, ROM, etc), compile it with the right HDL compiler and run the entire thing with small test programs as a simulator. Thats how much of the validation/verification work is done before they make the masks.

As for using the old code bases... That's done a lot. There's just too much complexity and too little time for them to re-write every processor from scratch. You also have countless hours invested in making sure previous designs work. If you're only doing small changes it would be hard to justfy building something from scratch since you'll have to do all of that validation work again.

Even more reviews

More reviews here [vr-zone.com] and here [extremetech.com].

Common Knowledge

I thought this was commonly known or assumed. Is this news to many people?

I thought that the only reason the P4 had not been totally abandoned already was that it takes time to switch directions in such a massive company. (and with so many partners that design around your product)

If those figures...

...actually show ANYTHING really well, then it's the absolute neglibility of recent synthetic benchmarks. Looking at the numbers SiSoft Sandra spills out, the clocked-to-the-brim Netburst-cores should take the performance-crown with ease in FPU and ALU-applications alike. In reality though, said CPUs hardly matter at all when it's about uncompromising peak-performance. I fail to understand why benchmark-suites this far away from reality still matter in reviews like this.
Sad, in an awkward way.

Benchmarks

I already posted some benchmarks of a Core Duo Mac Mini running Windows (http://slashdot.org/comments.pl?sid=182379&cid=15 077120 [slashdot.org]) and to be honest I was fairly impressed. The gaming benchmark was obviously miserable, the "general purpose" benchmark (zipping files, encoding audio/video, etcdid very well. The Apple zealots may say "it's because it's a Mac", but really the hardware is almost identical to your average Intel laptop. The only major difference is the Core Duo, which not many laptops have (although that's increasing all the time), and that's what I'm putting my money on. Can't wait to see a benchmark with this thing in a gaming rig.

Take note! Many of these features inside AMD too.

"But Yonah also supports the group of 13 new instructions known as SSE3, handles some SSE2 instructing like Shuffle and Unpack up to 30% faster, and is capable of using its instruction-grouping abilities (known as micro-ops fusion) on some SSE instructions, improving overall throughput."

SSE3 has some very nice hardware thread synchronization instructions. These are important (and AMD has them now). As for the instruction grouping, that sounds rather suspiciously like the double dispatch operations [chip-architect.com] that were added to Opteron:
"Appendix C of Opteron's Optimization Guide specifies to which class each and every instruction belongs. Most 128 bit SSE and SSE2 instructions are implemented as double dispatch instructions. Only those that can not be split into two independent 64 bit operations are handled as Vector Path (Micro Code) instructions. Those SSE2 instructions that operate on only one half of a 128 bit register are implemented as a single (Direct Path) instruction."

Assuming AMD can tune Turion64s to be more power friendly, they'll be able to best Intel's fancy new Core Duo. If they can't, then Intel may be the best game in town for the first time in a decade (assuming they price competitively).

Maybe per watt performance is the best but...

Looks like AMD still has them beat. From my take on this, on pure performance, the 3800+ X2 is going toe-to-toe and the 4800+ X2 is beating it every single time. So again, not that impressive. Now the per watt performance is important in some applications, so I can see why it would be a better, say, mobile platform than the AMD chips. But let's not pretend that Intel is winning the benchmarks with this quite yet.

Re:Maybe per watt performance is the best but...

But let's not pretend that Intel is winning the benchmarks with this quite yet.

'Yet' is now.

Merom/Conroe defeats AMD-AM2 hands down, and AMD has nothin' on the roadmap for the next two years, because AM2 slipped a full 12 months.

Go surf around Anandtech.com

AMD is in deep doo doo.

Keep in mind that

Intel's lead is mostly a manufactoring one -- 65nm process. AMD still uses 90nm. Not to discount Intel's advantage, but AMD doesn't need a new core design to continue their dominance -- merely a new manufactoring facility (which is hard, but not as hard as the design).

Re:What?

The reason for going to 64-bits is to increase the amount of physical address space, not for speed. The majority of applications, especially integer, do not benefit from bigger registers and wider ALUs.

Re:What?

Actually, x86-64 does have some speed benefits over standard ia32 for smaller programs and data sets in that it doubles the number of exposed registers. Most other archs were not register starved on the 32 bit version, so going 64 bit generally slowed the system down a bit because the pointer size doubled, taking more memory bandwidth to store pointers.

Re:What?

The "more registers" with x86-64 has been massively overhyped. There's very little real world benefit.

For example: AMD's claims about UT2004 being 20% faster in 64-bit mode turned out to be bogus (more like 2%).

Re:What?

The Core Duo cannot do the same things the Athlon 64 X2 can. Largely because (gasp) it cannot run 64bit code.

What the hell is the point of this comparison?

You're correct, of course. However, many of us don't need to run 64-bit code. You can completely ignore this, because any 32-bit CPU doesn't fit your needs, but please try to understand that other people need different things.

Re:What?

The Core Duo cannot do the same things the Athlon 64 X2 can. Largely because (gasp) it cannot run 64bit code.

I drive an 18 wheeler, and I can't imagine why anyone would want a passenger car. You can't haul near the same amount of goods!

Re:What?

Hate to say this, but there are not that many uses for 64 bit processors yet. Manufacturers do not provide 64-bit drivers for their products. The drivers that exist are buggy. To the average Joe, 64-bit is useless. He doesn't need the extra horsepower for his Internet browser or word processor. Well, unless Vista comes out.

Re:Load of Crap

Our QA department is testing my universal application right now (AppKit based). They've recorded a 20 to 30 percent increase in performance of a 1GB MacBook Pro over a 3GB 2Ghz Dual G5 doing a particular operation (mostly mathematics based done in cross-platform C++). It's single threaded, I might add, since OpenMP isn't here yet. The *ONLY* difference in the XCode settings between the two architectures that I made was to enable SSE3 for the Intel build. I can't believe that it's that alone, of course, and suspect it's just better code gen for the Intel architecture coming out of GCC.

Re:Load of Crap

It is not faster than the G5 period!

It sure the hell is. I have a 2.0x2 G5 desktop machine and one of the new 1.66 GHz Core Duo Mac Minis. Running Handbrake [m0k.org], the mini is easily twice as fast.