Why You're Not Getting Your Money's Worth....

[Capper]

Why You're Not Getting Your Money's Worth Out of That New CPU


Multicore CPUs like Intel's Core 2 Extreme and AMD's dual-core Athlon 64 have brought about better performance, better power management, and a way for the industry to free itself from a slavish devotion to sheer clock speed.
But multicore CPU architectures are creating a nightmare for programmers, particularly those who want to take full advantage of the new chips' power. The upshot? Much of your brand-new CPU's potential, like an uneducated brain, is going to waste.
That's quite a change from the days when the industry was fixated on clock speed. In those days, developers got a "free ride," says Jerry Bautista, director of technology management for Intel's Microprocessor Technology Lab.
"Even if (programmers) did nothing and the clock speed doubled, their software would run significantly faster," Bautista says of the days when the megahertz wars raged. "When we go down the path of parallelism, that free ride is over."
Help is finally on the way. The chip industry is sending out the cavalry ... in the form of new development tools.
In general, "multicore" chips include two or more cores -- the central processing units of a chip -- on a single piece of silicon. This allows properly coded software to break computing tasks down into separate pieces, known as "threads," and process the threads simultaneously, in parallel, instead of sequentially, as older single-core chips require.
Although multicore platforms have been around for some time in academia and research, it's been just over two years since the chips were commercially introduced by the likes of Sun Microsystems, IBM, Intel and AMD. Now, as core counts are poised to take off with eight, 32 and even 64 cores, the software that will run on them is seriously lagging. With the exception of the gaming industry, the vast majority of software publishers aren’t programming for multithreaded chips.
Indeed, the potential benefits of multicore chips are rendered obsolete if the software itself isn't coded to take advantage of its primary selling point: namely, parallelism.
Put another way, for the software to run at maximum speed, programmers will have to develop multithreading applications that take advantage of them. As Alan Zeichick, president and principal analyst for Camden Associates notes, that's hard -- hard, the way earning a Ph.D. in computer science is hard.
Typical 9-to-5 programmers, who are used to programming single-thread apps, are ill-equipped to handle things such as memory locks or calculation delays. What's more, there are added complications like scalability: Code written for eight-, 16- or 32-core systems won't necessarily scale up to work on systems with 64 cores -- like Tilera's recently announced Tile64 -- or even more.

Why You're Not Getting Your Money's Worth Out of That New CPU

[stinger608]

Its always the same story, just like the 64bit problem, you have the Chip, mobo, ram, and gaming community building all of the killer technology, and then the damn app developers are still trying to figure out how to build something for a pentium 133

[Dread]

I'm, for the most part pretty happy with my multi-core systems.
I have only one app that can take advantage of multi-cores running at any given time, but the rest of the apps show improvement with multi-tasking.
4 cores? 4 apps running without a slow down.

It would be nice to have some more apps take advanatge of the additional cores available, but I won't be holding my breath for it.

[Quakindude]

This is why I'm satisfied with my Core 2 Duo for now. I don't need Quad cores and rarely have more than two apps open at any given time any way.

[One4yu2c]

Originally Posted by Bryan Gardiner

That's quite a change from the days when the industry was fixated on clock speed. In those days, developers got a "free ride," says Jerry Bautista, director of technology management for Intel's Microprocessor Technology Lab.
"Even if (programmers) did nothing and the clock speed doubled, their software would run significantly faster," Bautista says of the days when the megahertz wars raged. "When we go down the path of parallelism, that free ride is over."

There's a key point that this article (and others like it) overlook, and it's that mutlicore CPUs haven't rendered a stand still among single core application performance. Programmer's software are still getting faster with each new processor release, because each new release represents a faster CPU than previously. The single core Pentium 4 played the GHz race, but gets summarily thumped in every conceivable benchmark, single or multi-threaded, when pitted against its successor, the Core 2 Duo. New processor lines aren't just adding more cores, the underlying architecture is getting improved/faster too.

Originally Posted by Bryan Gardiner

Much of your brand-new CPU's potential, like an uneducated brain, is going to waste.

That's true, but 'waste' is perhaps too strong a word. Untapped might be better suited, as the former wording makes it sound like one might as well opt for a single core processor. But not only are there a bevy of multi-threaded programs out there (and not just content creation, but even some games are getting in on the multi-threaded action), there's also really no single core CPU that can compete with the current multi-core kings of the hill (C2D), and today's multi-core processors don't cost more than what Intel and AMD used to charge for single core parts when that's all that existed. In fact, Intel's pricing structure these days is much more wallet friendly than year's past. Kind of calls into question the article's title...

[gvblake22]

There's a lot of room for argument either way on this subject. First, let me start by saying that hardware technology has always outpaced software. I cannot think of a single instance when there was new software available that had a special feature that didn't yet exist somewhere in the computer hardware.

When it comes to the "multitasking" argument, it could easily go both ways. Let's say for example you have a dual core processor and you want to run two single threaded apps. Theoretically each application will use nearly all of one core (about 50% of the processing power). Suppose it takes the dual core processor 1 minute to complete the two tasks.

Now think that you have that same dual core processor but it now runs the same two tasks that have been patched for multi-threaded operation. Again (theoretically) assuming you start both tasks at the same time, each task will use all of the CPU's processing power (about 100%) and complete the task even faster. For simplicity's sake, lets say the same single task takes 30 seconds instead of 1 minute. But given that the processor has to finish the first task, then start on the second because each task now uses 100% of the processor, it will still take about 1 minute to complete both tasks.

Please correct me if I'm wrong, but isn't it essentially "six one, half-dozen another"? Meaning the two tasks are still completed in the same amount of TOTAL time whether they are multi-threaded or not?

[Elysium]

i like my C2D, it's a lot faster than my old single core. But programmers should still keep up with the times.

[One4yu2c]

Originally Posted by gvblake22

Please correct me if I'm wrong, but isn't it essentially "six one, half-dozen another"? Meaning the two tasks are still completed in the same amount of TOTAL time whether they are multi-threaded or not?

No, multi-tasking is noticeably smoother on a dual (or quad) core machine, than one equipped with a single core CPU. Whereas it's fairly easy to bring a single core machine to its knees (heavy video encoding, etc), the extra cores keep the system from getting bogged down, freeing the user to focus on other tasks.

But again, it's a dying argument, as single core processors are quickly being phased out, and the architectural advantages of the C2D series aren't present in any single core parts. A perspective one could argue is that you're buying a faster processor, and getting an additional core to boot. The single vs dual core debate was much more applicable about two years ago, when the s939 X2 procs first hit the scene. For the same debate to take place today, it'd make more sense to discuss dual for quad core parts.

[Elysium]

Actually, I think I read something about some of the newer Celerons having Core2 single cores...... Celerons with Conroe cores.

[One4yu2c]

The author (and my responses) are geared towards the mainstream processors, from budget all the way up to enthusiast. Gimped Celerons don't really factor in the the single vs dual core debate, not the least of which is that the Celerons would 'give' an unfair disadvantage to the single core comparisons (they're handicapped from the start). Even the comparatively priced E2140 comes with double the L2 cache, and 4x as much when moving on up the line (as well as a frontside bus boost), not to mention the GHz disparity.

Single core is all but dead, and already deceased among the mainstream.