Replies to post #15522 on Intel Corp (INTC)

[alan81]

Why dual core...
I may be different than most... well, I guess that goes without saying. But,
There are two cases where I wish my system was faster.
(1) Virus scan is running and all things slow down to a noticable level. It is not clear dual core will fix this as it may be a disk access bottleneck. I did notice a significant improvement in this situation when I switched to a HT enabled P4. I strongly suspect dual core will help out in this situation.
(2) Video editing software. Lucky for me, the video editing software I use is multithreaded.

Pretty much everything else I do is plenty fast on yesterdays processor anyway.
--Alan

[wbmw]

Tenchu, Re: I just don't see what "killer app" is demanding dual-core

I think Intel's recent reorg towards a platform orientation is the key. You can't just sell the CPU anymore and expect people to keep on demanding that kind of linear performance. One of Intel's new organizations is the Digital Home, and some of the new usage models includes converging the digital capabilities of DVR's and home entertainment centers with the PC. If you get past the difficulties of this convergence, you can imagine what possibilities might exist for PC's that can decode television streams, archive and record digital media, and manage everything from still images and music to home videos and movies, all on the same PC that handles everything else it once did.

Imagine the compute power of a single PC in the living room being shared between other clients elsewhere in the house, too. These clients don't have to be additional PC's, but perhaps wireless enabled televisions or appliances, cell phones, or even alarm clocks. It would be cool, for example, to have your alarm clock play from a list of favorite MP3's in the morning; or imaging turning on your television when you wake up and watch the 11 o'clock news that you recorded the previous night and archived on the living room PC.

These features may take time to integrate together, but it's all part of the Digital Home concept. I think the "killer app" vision is dead, but long live the "killer platform".

[drjohn]

Re: Smooth, I'm really afraid that the whole push to dual-core will fall flat on its face, as once again the critics bash it for not making a difference on "current" desktop applications. I just don't see what "killer app" is demanding dual-core,

It all depends what the user experience is like, the most system intensive tasks I perform are DVD editing, one of the two 2.5 year old 80gb segate SATA drives running in RAID O on my system finally decieded to pack it in: I took the opurtunity two upgrade to two 160 gb SATA Maxators and ugraded the 2.4c northwood to a 3.0E prescott. Well guess what my vedio editing times are down by 30-50 % so I think it was worth. Obviously the hard drives played a big part in the improved performance. Guess what many of the AMD gamers use for thier non-gaming desktops, thats right northwoods and prescotts because they find they have a much more pleasant multitasking experience (i.e. productive). So if a dual core setup will decrease my DVD editing times by another 50% then I think it will be worth my while to upgrade.

[SemiconEng]

Smooth, I'm really afraid that the whole push to dual-core will fall flat on its face, as once again the critics bash it for not making a difference on "current" desktop applications. I just don't see what "killer app" is demanding dual-core, at least in the consumer market where such Centrino-like platform strategies are targeted. It's a different story when it comes to servers, of course, and I suspect Otellini will try and craft a new kind of platform strategy targeted toward business and corporations. But as for the consumer market, I don't see anyone really screaming for dual-core at the moment.
Tenchu


WAG: I think that those who are "claiming" that Smithfield won't be a good performer are wrong. In any case, I think that intel should focus on promoting the power reduction of dual core. What reduction you ask? Well, although I don't have "hands on" acess to the data numbers, I've recently seen reports that indicate there is an unusual effect of linking 2 cores together. As I understand it, due to transistor physics, when dual cores are linked, a small reduction in clock speed, results in a huge reduction in power, as well as an almost doubling in performance.

I should have asked the question, if they meant 1.8x performance on "optimized" applications, but it didn't occur to me until later. I would guess that "optimized applications" was the case, but what if they meant "all applications"? In any case, the point is, it appears that 2 cores linked, with a slight decrease in clock speed, results in significant power reduction, while still increasing performance, on at least some applications. Perhaps that's why, as The Register reported, that Smithfield appears to be 2 cores, while being introduced at lower clock speeds, while at the same time, intel is claiming significant increased performance, "in the same thermal budget"..... And if that's true, It wouldn't even be too far fetched to decrease the clockspeed a little further, for Dual Core Mobile.


http://www.theregister.co.uk/2004/12/14/intel_dual-core/
What all that will do for performance remains to be seen. Smith claimed by 2008, multi-core chips will offer ten times the performance of an "initial Pentium 4" processor, compared to the threefold gain today's HyperThreading chips provide. Next year, dual-core will see that improvement increase beyond 3x, but from we'll be well into 2006 and the 65nm node before multi-core gives a significant performance boost. Only then will dual-core offer a greater gain than HT alone will do by 2008.

Smith would only say that Smithfield contains "two execution cores", not whether they're on the same die or not. Separately, he admitted Smithfield's clock speed range would be lower than the top end of today's P4 CPUs, to ensure the new chips "operate in the same thermal budget". Intel's roadmaps put those speeds at 2.8-3.2GHz, well below today's 3.8GHz P4 570.

[imho]

Tenchu,

I'm really afraid that the whole push to dual-core will fall flat on its face

You maybe right based on a technical point of view. However, Intel is no longer run by a technical guy, are they? I believe that Intel's reaction is as much a response to AMD as it is to addressing technical shortcomings of new processors.

I believe the technical argument you are making was also made about the issue of 64 bits on the desktop. AMD championed the idea, while Intel, perhaps correctly on technical grounds, ignored it. The result, IMO, is that AMD positioned itself as the innovator and leader in a new technology everyone expected Intel to be in. With dual-cores, the same thing was going to happen. AMD comes out championing the idea and boldly predicts it will be first on the market. The old Intel, argues on technical merits and watches as the groundswell of talking heads take AMD's position. Intel suffers from an image point of view, which is significant in the eyes of the market. This time, however, Intel is responding differently.

I believe Intel learned something from this. When it comes to advocacy, the size does not matter and the smaller you are the louder you get. Why argue on the basis of technical issues when your number one competitor will be making the same mousetrap that you (Intel) can easily make yourself? Intel now realizes that it has challenges on two fronts. Technical issues facing products it wants to champion, and products to counter what its competitor wants to champion. Intel, IMO, is simply taking on both challenges, head on, instead of getting into a public spat over technical issues.

IMHO

[smooth2o]

Ten: Smooth, I'm really afraid that the whole push to dual-core will fall flat on its face

***
I think not. We constantly hear the old argument about not needing more and more compute performance. This just doesn't seem to ever pan out. The new PC architecture is adding even more performance as we type. With the advent of the platform strategy in both the digital office and the home, I think you will find plenty of areas in which dual core performance will be needed. One thing for sure, more performance begets applications that need more performance.

Right now,virus scanning is minimal, most scheduling it to occur at off hours, or when it comes up, cancelling it. Most of my workers do, anyway. This is one area that needs much more development.

Also, in a earlier post, I pointed out that all applications are process based and sub-processes are capable of being scheduled on multi-cores. I think you will find significant speed improvement with dual cores.

Smooth

[Saturn V]

Ref dual-core will fall flat on its face.. for not making a difference on "current" desktop applications.

It may not be as bad as you fear.

Dual core will definitely lead to a greater performance improvement than Hyperthreading ever did. For the typical application the performance still improves:

A. The application and the OS will run in different threads, and so the computer will "respond faster". For example the application will run faster, because the OS functions like Screen Refresh, Disk IO, User Input Preprocessing, will run faster since these functions can be in a different thread from the application.

B. The OS have been rewritten to take advantage of multithreading, since HT has been around for a while.

C. The consumer does run multiple applications together, and the benefit will be plain to see.

The concept of dual core is easier to grasp than Hyperthreading. So the consumer will want Dual core since that will speed up the performance for future applications. A good marketing campaign like the MMX Ad campaign convinced everyone to buy MMX, long before any MMX software arrived. As a matter of fact the MMX ad campaign was too sucessful, and within two months destroyed the demand for non-MMX Pentiums, and Intel was caught with inadequate MMX units in the manufacturing pipeline. We should hope that something similar will happen with the Dual Core.