Intel's Dual Core Strategy Investigated
by Anand Lal Shimpi on October 22, 2004 3:09 PM EST- Posted in
- CPUs
Final Words
The best way to evaluate the impact of dual core CPUs on the desktop is to look at the impact by moving to a multiprocessor setup on the desktop. The vast majority of applications on the desktop are still single threaded, thus garnering no real performance benefit from moving to dual core. The areas that we saw improvements in thanks to Hyper Threading will see further performance improvements due to dual core on both AMD and Intel platforms, but in most cases buying a single processor running at a higher clock speed will end up yielding higher overall performance.
For the most part, it would seem that the dual core releases of 2005 are mostly to establish a foundation for future dual core CPU releases that will provide functionality such as power and thermal balancing across multiple cores. Next year Intel will be releasing a number of new processors, including the new 2MB L2 Prescott parts as well as the dual core x-series, but despite all of the new product launches, clock speeds will only increase by 200MHz in the next 14 months. If anything, the release of larger cache and dual core desktop processors is a way to continue to promote the "newer, faster, better" upgrades without necessarily improving performance all that much.
Today the slowest Prescott based Pentium 4s run at 2.8GHz and 3.0GHz - and a full year from now the slowest Prescott based Pentium 4s will run at 3GHz. This is the first time in recent history that the predicted roadmap for CPUs will remain relatively flat. It will take continued maturity in 90nm manufacturing, a smooth transition to 65nm as well as improvements in multi core designs to truly make the migration worth it.
The future of dual core doesn't lie in taking two identical cores and throwing them on the same die. The future and true potential is in the use of multiple cores with different abilities to help improve performance while keeping power consumption and thermal density at a minimum. The idea of putting two cores, one fast and one slow, in a CPU has already been proposed numerous times as a method of keeping power consumption low while continuing to improve performance.
Right now dual core is more of a manufacturing hurdle than anything else. Putting that many logic transistors on a single die without reducing yield is a tough goal. Intel will have a slightly harder time with the migration to dual core since their chips simply put our more heat, but in theory Intel has the superior manufacturing (although it's been very difficult to compare success at 90nm between AMD and Intel thanks to all of the variables Prescott introduced). Needless to say that we'd be very surprised if both companies met the current ship dates for dual core desktop chips simply based on how things have progressed in the past.
That being said, despite the end of 2005 being the time for dual core, the desktop world will be largely unchanged by its introduction. It will take application support more than anything to truly bring about performance improvements, but with an aggressive CPU ramp developers may be more inclined to invest in making their applications multithreaded as more users have dual core systems. The more we look at roadmaps, the more it seems like while 2005 will be the year of anticipation for dual core, 2006 may be when dual core actually gets interesting. Until then, we view dual core on the desktop as a nice way of getting attention away the fact that clock speeds aren't rising. It's a necessary move in order to gain more traction and support for multithreaded desktop applications but its immediate benefit to the end user will be limited. But then again, so has every other major architectural shift.
Facebook
Twitter
RSS
59 Comments
View All Comments
ChronoReverse - Saturday, October 23, 2004 - link
@48And Windows reports 10 threads for the UT2K4 demo. I still know that it's not really designed to take full (by full I mean special) advantage of multithreading.
Chuckles - Saturday, October 23, 2004 - link
#37According to 'top' on Mac and Task Manager on Win XP:
Escape Velocity: Nova - 7 threads
Robin Hood: Legends of Sherwood (Demo) - 5 threads
Airburst Extreme (Demo) - 8 threads
Homeworld 2 (Mac Demo @ 1st mission, 2nd save) - 8 threads
Homeworld 2 (PC Demo @ 1st mission, 2nd save) - 8 threads
All seem to be reporting multiple threads.
Sunbird - Saturday, October 23, 2004 - link
I have a question that just popped into my mind,Will the dual core processors from AMD and Intel technically be two 64bit cores?
If it is, man, you get dual core and 64bit all in one, seems it will be pretty cool (but not those dual core prescotts :P ) when you have both of those steaming away with software written for them.
AdamK47 3DS - Saturday, October 23, 2004 - link
I envision more marketing behind the dual core CPUs than performance numbers to sell these. Most people (as is proven by the responses here) will think their getting double the performance simply because they are "dual cores". Dual cores require multithreading capabilities to truly take advantage of them in a single app. If you're a gamer there aren't very many games out there that are multithreaded. Even most games coming out in the next couple years won't be multithreaded. Multithreading can be cumbersome to programmers.AdamK47 3DS - Saturday, October 23, 2004 - link
Reflex - Saturday, October 23, 2004 - link
Just to make one thing clear: I like dual core, I think it is a good move, I can't wait to get mine...However, you will not see 100% improvement on a dual core system, not even close, even with multi-threaded apps. At best you will see 80% in *some* situations, in most circumstances its more like 40-50%. The thing to remember here is that those dual cores are sharing the rest of the system with each other, so a straight 100% improvement is impossible due to the fight for system memory and resources. This is exacerbated on shared bus designs like the P4, but even in the case of the Athlon64 there are some shared resources..
xsilver - Saturday, October 23, 2004 - link
#42 I think you answered your own question"Within a year or two, you will be buying (essentially) 6 to 8 GHz CPUs, instead of 3 to 4 GHz CPUs."
Essentially this 6ghz cpu is no better than the 3ghz cpu on one program -- its all fine and dandy to be able to run 2 programs just as fast with no performance hit but when advertisers say it runs X fast, that X would not have changed; only you can run X and Y at the same time.... Its not double performance, its more like HT overdrived.... you mention MMX, sse etc. they are good features cause they add no cost/ heat into the equation whereas dual cores may double the processor on both counts here...
The INTEL spin doctors though will spin it so it sounds good until software can use the multithread properly...
theprofessor - Saturday, October 23, 2004 - link
I don't know why everyone seems so against dual core CPUs. I have been waiting years for this (as I was for 64-bit CPUs). Most people will see an increase in performance using a dual core CPU. I don't care if the program is single threaded or not. Most people run more than one program at a time. All modern Operating Systems (including XP) will allocate time on both cores for different processes. So, while you’re playing your single threaded game on one core, you can run your encoding software, anti-virus software, im software, e-mail software, download software, whatever software on the other core, with no hit to performance. If you don’t play games, include whatever single threaded software you like. There will be a decent boost to performance no matter what you are doing.Dual core is the best upgrade a processor can get. Why? Because with MMX, SSE, SSE2, 3D Now!, and, especially, 64-bit, there needs to be re-programming and/or re-compiling involved to notice any difference at all. With dual core you will notice a difference in almost all modern computing paradigms the day the chip is released. And as the technology becomes more mature and prevalent, you will notice even more performance as developers re-program/re-compile there software.
Computer enthusiasts should be looking at this as a doubling of computing performance. Within a year or two, you will be buying (essentially) 6 to 8 GHz CPUs, instead of 3 to 4 GHz CPUs. As far as pure performance increase, I think this could be the greatest technology ever introduced to a processor line. No other technology (listed above – MMX, etc.) has been able to give almost 100% performance increase (theoretically) across the board in all applications.
If nothing else, think of it this way. With the ramp in CPU frequency drastically slowing over the last two years, going forward it will now be at least double what it would have been. (i.e. If AMD and Intel can currently only handle 200 – 400 MHz frequency increase in a year, with dual core that becomes 400 – 800 MHz.)
It’s a win almost every way you look at it. So please, try not to be so critical of this great technology.
Thank you
Reflex - Saturday, October 23, 2004 - link
#37: If that were true then every ported PC game would take years to rewrite and bring to the Mac since they are not designed multi-threaded in the first place(and most Mac games are PC ports).Don't confuse the fact that you can multi-task while playing a game with the idea that the game itself is multi-threaded. They are not the one and the same, and you can multi-task while playing a game on Windows as well...
thermalpaste - Saturday, October 23, 2004 - link
Intel could have launched the dual core using the P6 architecture as in the pentium-M processor. If they still love high clock speeds, maybe they can deepen the pipelines a bit for the p6. With prescotts touching 65 degrees plus, the dual cored pentium-4s may need car radiators in order to overclock;). Its obvious that the dual cores only come into play for multi-threaded apps, so perhaps intel can shift back to p6 when majority of apps support multi-threading.