Intel Dual Core Performance Preview Part I: First Encounter
by Anand Lal Shimpi on April 4, 2005 2:44 PM EST- Posted in
- CPUs
The Intangible Dual Core
The move to dual core is a bit of a "catch 22". In order to deal with the fact that a dual core die is twice the size of a single core die, AMD and Intel have to use higher yielding transistors. The larger your die, the more defects you have; so, you use higher yielding transistors to balance things out. The problem is that the highest yielding transistors run at the lowest clock speeds, so dual core chips end up running at slower speeds than single core chips. While the Pentium 4 could have hit 4GHz last year, we won't break the 4GHz barrier until late 2006 at the earliest.
In Intel's case, we're talking about 2.8GHz - 3.0GHz vs. 3.6GHz - 3.8GHz for the high end single core chips. In order to offset the difference, Intel is pricing their dual core chips within about $80 of their single core counterparts. Short of giving dual and single core chips a price parity, this is by far the best approach to assuring dual core adoption.
Why does Intel want to encourage dual core adoption? To guarantee a large installed user base, of course. The problem today is that the vast majority of desktop systems are single processor systems, meaning that most developers code applications for single processor systems. To encourage a mass migration to develop multithreaded applications, the installed user base has to be there to justify spending the added time and resources in developing such applications. As we just finished mentioning, Intel's approach is the quickest way to ensure that the exodus takes place.
So, with dual core CPUs priced very close to their single core counterparts, the choice is simple right?
On the Intel side of things, you're basically giving up 200MHz to have a dual core processor at virtually the same price. But things get a lot more complicated when you bring AMD into the situation. AMD hasn't officially released their dual core availability and pricing strategy, but let's just say that given AMD's manufacturing capacity, their dual core offerings won't be as price competitive as Intel's. Now, the decision is no longer that simple; you can either get a lower clocked dual core CPU, or a higher clocked single core AMD CPU for the same price - which one would you choose?
The vast majority of desktop application benchmarks will show the single core AMD CPU as a better buy than the dual core Intel CPU. Why? Because the vast majority of desktop applications are single threaded and thus, will gain no benefit from running on a dual core processor.
Generally speaking, the following types of applications are multi-threaded:
- Video Encoding
- 3D Rendering
- Photo/Video Editing
- most types of "professional" workstation applications
However, the vast majority of other applications are single threaded (or offer no performance gain from dual core processors):
- office suites
- web browsers
- email clients
- media players
- games, etc.
If you spend any of your time working with the first group of applications, then generally speaking, you'll want to go with the dual core CPU. For the rest of you, a faster single core CPU will be the better individual performance pick.
But once again, things get more complicated. Individually, single threaded applications will make no use of a CPU able to execute multiple threads. But, run more than one of these applications at the same time and all of the sudden, you're potentially dispatching multiple threads to your processor and thus, potentially, have a need for a multi-core CPU.
Facebook
Twitter
RSS
141 Comments
View All Comments
fitten - Tuesday, April 5, 2005 - link
#66: What he said about SMT (which HT is a brand name of) is kind of off. Of *course* having two cores is better than a single one acting like one. Heck, having 8 cores is better than a single core acting like 8. However, Intel's HT added about 5% to the total realestate of the chip to get arguable benefit. Adding another core adds around 100% more realestate.As far as as having a waste or resources, which is more wasteful: adding 5% logic or having execution units sitting around idle (and unused)? (Remember that just about any execution unit (especially an OOOE one) is going to be larger than 5% of the logic of a core.)
defter - Tuesday, April 5, 2005 - link
"Dont forget that socket 939 is dual channel it could be possible to give one memory channel for one processor and the other channel for the other"Actually you are propably right, memory controller can be build in a way that it can make two independent memory requests simultaneously. BUT the P4 chipset can be built in the same way, thus P4 chipset may also have the ability to allocate one memory channel for each core.
Illissius - Tuesday, April 5, 2005 - link
I second the request for Linux benchmarks. Primarily because it's what I use :), but also because I've heard rumors that Windows's task scheduler sucks at multithreading, and it'd be nice to see if they have any grounds in reality (my suspicion is it's just a relic from the win9x era, but you never know)...re: repeatable multitasking benchmarks. couldn't you use the task scheduler / at / cron for that? or are those not fine-grained enough?
Also, benching game + other intensive task isn't as dumb as it sounds -- especially as the game would be the one that has the focus, so dual cores might actually make that a viable scenario (remember that whole 'enables you to do entirely new things' part?).
defter - Tuesday, April 5, 2005 - link
#71"I am not really sure about that, amd always said the processor was being done dualcore since day one that must mean something."
Yes and Intel always said that P4 makes internet faster that must mean something :)
"Dont forget that socket 939 is dual channel it could be possible to give one memory channel for one processor and the other channel for the other"
Actually it isn't possible: K8 based CPU has a CPU core and an integrated northbridge that has a memory controller and a HT link. K8 based dual core CPU has two CPU cores and one northbridge. So in dual core K8 CPU both cores are using the same memory controller.
Crassus - Tuesday, April 5, 2005 - link
Great preview Anand! I enjoyed especially the the multitasking tests and they confirmed essentially what's been known in the community since the days of the MP Celerons and Athlons. I liked also the fact that the tests were run with a whole lot of background tasks running, since that is the reality of present-day computer usage by the vast majority of users (who in most cases don't have the slightest idea of threads and schedules and happily install everything that can create a pop-up to ask ;c)My suggestions for future testing: Although you correctly mentioned the performance of games won't change due to being single-threaded in nature, it would have been interesting to see the impact of a number of general-purpose background tasks (Antivirus, Antispam, ICQ etc.) both on single- and dualcore chips. Enthusiasts know which threads to kill before gaming, but does the enthusiast's 'The Sims' playing family?
More about the general move to parallel processing of data: Didn't you promise a series of RAID-related articles back in last year? What became of them, considering that the demand for a fast supply of data now essentially doubled? Maybe it would also be feasable to look into RAMDISK solutions again?
And one point of criticism at the end:
You did mention that games recieve no benefit from the move to dualcore, but to weight the picture the reader gets from the review, a representive number of pictures does wonders, even if they proove nothing really new. After all, one picture says more that thousand words - and a couple of benchmark diagrams a lot more than a sentence at the bottom of one page!
Keep up the good work! Take care - Crassus.
stevty2889 - Tuesday, April 5, 2005 - link
Also what about running 2 games at the same time? I play MMORPG's and at times with my dual monitor setup, I'll have one game running on one screen and another running on the other..it's a bit sluggish, but do-able with hyperthreading, but I would think a dual core would allow this to run more smoothtly.L3p3rM355i4h - Tuesday, April 5, 2005 - link
lol. 50 cent as a l337 computer h4xx0rz.Impressive, but how can the TDP be only 130 watts? Unless Intel has some kind of magic, a prescott at the same voltage wouldn't be able to run a mere 65 watts. NFW.
Anand Lal Shimpi - Tuesday, April 5, 2005 - link
nserraI was skeptical that gamers would have things like this running in the background while they played, but given that a handful have requested the tests be created and run I have no problem doing just that. I'm working on Part II right now and I hope to finish it late this afternoon.
Take care,
Anand
michael2k - Tuesday, April 5, 2005 - link
#72You want to use Doom3 as a benchmark while DVDShrink runs in the background?
I suppose that's a benchmark, but I doubt it's a valid one... but if Anand has time to try it, I suppose what you'll see is that the performance of Doom3 will be LESS than the performance on a similarly clocked P4 running Doom3 WITHOUT DVDShrink.
I suppose you want to know what the performance penalty is, though :)
nserra - Tuesday, April 5, 2005 - link
#70 Yes you are right and wrong. So i can listen MP3, zip files, record an dvd and do word processing. But i cant play a game while the PC is doing other things?