STANFORD UNIVERSITY HAS joined forces with IBM, AMD, Sun Microsystems, Nvidia, Hewlett-Packard, and Intel to create innovative software that will let chips better process several simultaneously.
According to the New York Times, the partnership between the University and the six rival computer and chip makers will be formally announced this Friday, and the project will be dubbed the “Pervasive Parallelism Lab”.
In mid March of this year, Chipzilla and the Vole announced that they’d be funneling a combined $20 million into building specialised labs at Berkely’s University of California and Urbana-Champaign’s University of Illinois for parallel computing research, which would effectively tackle the same problem.
The massive amounts of funding and effort being channeled by the big players into such research programmes shows just how worried the software industry actually is about future microprocessors with 8, 16 or more cores on a single chip. The concern is that the software would not be able to work properly with the new hardware because without optimal programming, applications don’t profit from added chip power, and in some cases can even become slower because of it. This means that customers could just decide that it’s not worth their while to upgrade their system.
Clockspeed is no longer as important as performance per watt in computing, and hiking performance is now the domain of multicores, with most corporate server microprocessors and gaming machines having about eight cores.
To help the software take better advantage of the increased number of cores, the competing teams of boffins are going to have to experiment with new programming languages and tweaks in the hardware, as well as going back to the drawing board on things like operating systems and compilers (which translates programming gibberish into commands a computer can actually understand).
With all the effort going into it, it definitely seems that sequential programming will soon be a thing of the distant past, whilst parallel programming’s day is yet to come. µ
See Also
Microsoft
and Intel finally team up on multi-core programming
Parallel programmers not prepared for the glorious revolution
This?Software is mostly that within CPU.
Multitasking is Parsing data jobs by on/off selection of those swirling Low/High bits for each jibber. Say 80 Sections,MAX, as defined by Snippets as IE. Yet what if some jobs don't like others(Too Chewbacka hairy or maybe too BIG)? Well You Send them to Back Burner, err, i mean most unused core to fiddle around. Maybe L3 Crossbar controls flowback of incomplete work from back cores into front end L1 pareser agin, trying to expand gate openings needed in its stored libraries when possible or in Snippets of job till done. Not Simple task I'd think. Yet it is 2 way Crossbar.
CROSSBAR:
We are ALL Christians in this game, Like All Childrentians over world, starting out, even Ninja Warrior Bulldozer will have(OR Hod) its humble start up moment & soon enough Stanford will have Axe to Grind, BoBoes.
Drashek
Ps It might have something to do with Software Writing...
Those of us lucky enough to have participated in the Transputer fest and learning to programme in Occam none of this is news. 

Dividing up real world problems meaningfully into parallel and serial actions proved insoluble then and I wonder what's changed.

According to Donald Knuth

http://www.informit.com/articles/article.aspx?p=1193856
Maybe they will discover Ada. Tasks for message based parallelism. Protected types for controlling access to shared resources. It even comes with most linux distros as standard now (gnat).
"...with most corporate server microprocessors and gaming machines having about eight cores."

er.. what? Did I read that right? Eight? Poor you for only having a quad core machine.
"...most corporate server microprocessors and gaming machines have about eight cores."

I don't know about you guys, but my gaming rig gets by with just two... Maybe that's why I can't run Crysis, lol.

Seriously, though, except for video editing, video encoding, Photoshop, or other extremely intensive applications, two cores are all the average consumer needs. My C2D will serve me well for a while, and I have no plans to upgrade it to a quad-core anytime soon. However, increasing the number of cores is the simplest and most efficient way to increase raw power, and it is definitely the direction the industry is heading. Software WILL have to catch up sooner or later. I'm glad to hear about this proactive group.
According to me, there is a problem writing applications that use multiple cores well, but there is no problem using multiple processes. By using servers, and in particular, terminal servers, we can run many processes in parallel on the new CPUs, and make good use of that hardware. For example, my computer lab uses LTSP and I run over 700 processes when it is busy. If I had about 24 cores on my server, it would be a blast of a desktop experience. Only if you need raw number/graphic crunching power is harnessing multiple cores a problem. For most of us in the real world, starting applications, opening windows, or clicking icons, we can make very good use of all those cores now with no problems.
Parallel programming has been wth us for a while.

@Gerry - Dividing up real world problems into parallel and serial actions is not always the aim of the game.

Doing several of them at the same time, with no performance hit, is also very useful. (Think webservers, transaction servers etc)
Remember when the geeks got bored with single CPU's and got themselves the amazing 760MPX chipset and then threw in a couple of Athlon MP's or hacked a couple of XP Athlons only to fins out that the only thing that really took advantage of parallel computing was the older versions of Quake? We all waited around for someone to make use of the cheap server/gaming machine but it never came true. I wonder if they take it serious this time if I can get out my old dual Athlon system......Hum....probably not.
Wow, perhaps Windows12 will be multi-threaded! 
Or at least all the DRM code, which no doubt will be 85% of the OS.
"most.. gaming machines having about eight cores." Hello?! Is he for real?
If IBM would simply bring OS/2 SMP back to life all problems willbe solved.
So far OS/2 is the only OS that can really multitask and does it very well, no matter the number of CPU's or Core's or both....

Windows is unable to even multitaks the most simple task good.
...but get lambasted every time I point out that the whole multi-core thing is really kind of a cop-out on CPU makers' part. Virtually no one can benefit from more than 2 cores, and in virtually all cases even those people would be better served with 1 4Ghz core instead of 2 3Ghz cores.

At some point the world is going to realize that the emperor has no clothes, and stop buying into multi-core mania. What a waste.
" most corporate server microprocessors and gaming machines having about eight cores."

So would that be 51%? The top of range 3D workstations at use where I work are dual core dual processor systems, so only 4 cores in those. The file and web servers are a bit dated, and are only dual processor single core Xeons running at 3.2Ghz. The render farm for all the 3D work is currently an aging bit of kit, with the top end of the lot being an Opteron 2Ghz dual processor 1U server. Guess that puts this company in the 49% that aren't running about 8 cores. :)

Cheers,
John
I can't believe nobody mentioned this. Remember the Big original rumours about phenom?

Now, if only that could be true!
Depends what type of work you're doing but for Multitasking, 3D Rendering and Visualisation multicore is a godsend.
Even if you're not in the graphics industry mutlicore just makes the whole windows environment flow better with multiple apps open especially video. If I used it for gaming I'd probably stick with a CoreDuo x8xxx oc'd to 4Ghz. Its all up to the software guys to optimize for multicore.
Hopefully, I won´t offend The Inquirer by ¨advertising,¨ but if you go to Google, and do a search on Distributed Computing, you will find many, MANY ways to utilize your cores.

Distributed Computing is, basically, a way to use underutilized cpus(the average cpu only uses about 15% of it´s power, on average, during time that it´s actually on) to do research. The type of research varies widely, from protein folding, to proving mathematical conjectures, to simulating subatomic landscapes, to all kinds of other stuff.

If you don´t like to get your hands dirty, but are interested in getting involved, I suggest you do a little research on the platform called BOINC. Just enter boinc into Google and see what you think. (Be sure that you remember that it´s spelled with a c, boink.com is a sex site for horny college students.)

If you have any questions at all, there are plenty of pages that can help you out. If you want to join a team, google Distributed computing teams.

If you have any questions that you specifically want to point at me, email me at

rieselsanta at gmail dot com.
It's unfortunate to hear multi-core is useless from the well-respected Donald Knuth. I thought programmers are the first group to support multi-core simply because compilations are the easiest to parallelize:
1: compile each file on a different cpu/core
2: combine object files to lib / exec

in case of Java it's even simpler
1: compile each file on a different cpu/core
2: end

I've worked on a number of projects where a full build takes hours. Here, double the number of CPU = half the compile time = doubling the productivity.

Also, running regression tests in the background while editing document / code is second nature.

Can't wait for the Oct- / Hex- cores to arrive.
I have no idea why Donald Knuth (of all people!) can't figure out how to parallelise a parser. Then again, TeX notation is a bit of a train-wreck compared to XML-based declarative syntaxes, which are more parallelisable especially if stored in a database.
>> Wow, perhaps Windows12 will be multi-threaded! 

Windows has been multi threaded since NT and Windows 2000. If you run two programs at once, they can both use separate cores. Since Windows runs a dozen services, you already have programs running in the background using whatever core is most free.

So you've been able to benefit from multi-cores for the past 8yrs. The only issue today is getting more programs (besides AutoCad, video encoding and Photoshop) to use more than 1 core at a time.

Games aren't really an issue, the bottleneck in the most demanding games (ex: Crysis) is video power, not CPU power.