I think I just lost all of the remaining respect I had for you Jeff. Not that you care, I know you hear it a lot from many longterm disappointed readers -- and I thought about just quietly removing your blog from my iGoogle with a sad shake of my head, but I thought I'd at least first make mention that I thought you were a bit more reasonable than this.

Good luck with your new business venture. You're a smart person and I'm sure you'll be successful, but if you take any advice away from your blog, let it be humility.

I'm one of these desktop users with both an x64 OS and a quad-core CPU. I have 8 GB of DDR2-8500 too. I'm not going to waddle a finger at you and scream "you're wrong", because I agree that the returns are diminishing, but they are definitely there in a desktop environment, depending on how that desktop user uses his/her system.

Windows is a multithreading environment, Jeff. Why oh why do you consistantly look at benchmarks targeted at benefits of multiple cores and their performance in a single application when it comes to a much grander scale than that? What about hardcore gaming (which I do) and running Apache and MySQL in the background? What about DVD encoding while streaming music and working on source code? How about virtualization for the developer that wants to test his cross-platform application?

I'm not saying you can't do this with a dual core. Or even a single core, but it is hands-down significantly faster to do all of these *simultaneous*, CPU-intensive tasks with 4 cores than it is with 2. Or 1.

I think I'll also disagree with how soon the market in general will turn over to spending more time writing multi-threaded applications. It's just how the industry works.

Heck, less than two years ago, you said: "Upgrading from the slowest Pentium D to the fastest Core Duo (which is ~30% faster) is a giant waste of money for a gamer. Spend your money on a better video card instead!"

(Jeff Atwood, August 2006, http://www.codinghorror.com/blog/archives/000655.html)

And now you're saying: "Four cores on the desktop [...] provide almost no benchmarkable improvement in the type of applications most people use. Including software development tools."

Which at least implies that there may be a significant improvement that *isn't* benchmarkable, but wordplay really. You had already said that there's no point in a number of other ways.

I'd like to see any self-respecting hard-core gamer out there that would rather go back to a single-core Pentium D and give up his Core 2 Duo, though Jeff. I think you'll wind up eating humble pie on Quad Core's as well.

My two cents.

(P.S.: I'm also not entirely sure we all agree on what the 'average desktop user' does with his/her PC, but it's probably a safe assumption that your average internet-browsing, computer-illiterate user will likely never put the kind of strain that even a Core 2 Duo can produce. Then again, Quad Cores aren't marketed towards this group either.)

Something I forgot to add that is relevant:

All other things aside, newer CPUs will benefit *all* users in at least some of the following categories:

* bigger instruction set (SSE3, SSE4.1, etc.)
* more on-die cache
* support for higher bus speeds
* lower power consumption

Depending on the specific use of the system, some of this will matter and some of it won't. And for the bigger instruction set, that's usually something that comes up after a year or two when the latest development tools have compilers that write machine code with support for those instructions. Immediate benchmarks can't account for that (at all). Take a look at more recent benchmarks at TomsHardware:

http://www.tomshardware.com/charts/cpu-charts-2007/sisoftware-sandra-xi,391.html

The top three entries in this list are all quad core processors (the Core 2 Extreme QX6850 has four cores, despite its name). The rest of the list is filled with Athlon X2's and Intel Core 2 Duo's. Notice the difference in speed between quad core and dual core? They handily double and triple the scores of most of the dual-core processors. The insane Intel QX6850's score is nearly four times the lower-grade dual-cores.

Though as the name implies, PCMark05 is quite an old timer and not optimized for some of the latest instruction sets, you can find its scores telling roughly the same story:

http://www.tomshardware.com/charts/cpu-charts-2007/pcmark-2005-cpu,382.html

On that chart, based on the CPUs it's shown, the average quad-core scored 8,175 and the average dual-core scored 4,477 (I counted 12 on the list).

For the user that quad-cores are marketed to, I'd say they carry significant advantage.

@Weeble

It's probably a negligible cost, but because of the rarity of this kind of event and the extremely competitive computer industry, keeping costs down by shaving every penny possible is vital. I have seen boards though (Asus comes to mind) with automatic flash protection that revert to a failsafe BIOS if the primary fails. Gets you far enough to flash from a floppy or USB, but won't boot an OS.

@Izzy

I have to say it: what was the point to your post? I'm not looking for an argument (really, I'm not), so apologize in advance if that sounds offensive. To understand you better, though, you're essentially speaking up in a thread largely about processor upgrading and the diminishing returns of quad-core and dual-core computing by saying that you don't need to upgrade, so why should you?

Well. You're right: you don't need to. So don't ;) Your needs are aligned with the type of user who puts relatively low technological demand on their system and you will probably only *need* to upgrade when current software is no longer supported (old versions of the O.S., old anti-virus software that no longer functions, etc.). New hardware is often needed to support new operating systems, but that's a far cry from the top-of-the line hardware mentioned in this thread.

You're also not going to see 5-to-10-fold performance gain boosts in upgrading from just a few years anymore (and I don't mean to say that as in 'ever in the future'), but as Jeff pointed out, the technology required to make huge leaps past the silicone barrior haven't been breached yet, and that's essentially what's holding us back at the moment.

Another thought:

"Intel EDAT: the End of the Multi-core Clock Speed Disadvantage?

Intel also talked about its "Enhanced Dynamic Acceleration Technology" which is effectively integrated overclocking based on load. If you are running a single threaded application (or a multi-threaded application that's predominantly using a single thread), Intel's EDAT can power down the second core and increase the frequency of the working core to maintain the same thermal envelope at all times.

Intel's EDAT could spell the end of the clock speed differential between single and multi-core processors. With all cores running workloads, the multi-core system would be clocked lower, but when some cores are idle the chip could potentially run at the same speed as a single core solution would. Single core designs have pretty much disappeared from roadmaps already, but considering there are still applications that are single threaded in nature and benefit more from clock speed improvements, future processors will offer both options in a single package."

Summary? On unoptimized threads that really need clocks, the CPU will be able to power down one core and overclock another automatically. Completely blows the 'faster CPU clock speeds slower, but more cores' argument out of the water.

(source: http://www.anandtech.com/cpuchipsets/intel/showdoc.aspx?i=2955p=2)