Posted on 08/18/2002 8:35:35 PM PDT by HAL9000
--Boris
I steadfastly refuse to make off-color jokes about this.
Not from me will you receive dry witticisms about the "3" referring to length, or possibly circumference.
No, I shall remain aloof from the gutter and make no comment whatsoever.
--Boris
I had the exact thought when I read it...
However, I am not altogether certain that the perceived lack of performance might not be attributable to the fact that her machine runs Windows XP.
So the P4 should have BLOWN by the AMD processor?????? What was/is their advertised speeds? Is one of the points of the article that Intel is fudging a little?????? Confused in Texas.
FGS
(*)Scaling up speeds is not as simple as one might think. Although the newer chips have many more transistors in them than older ones, signals actually pass through fewer transistors each clock cycle. Even though an 8088 does much less useful work per clock cycle than a P4, it would probably be difficult if not impossible to engineer a version of the 8088 which didn't use any more transistors but which could go at 100MHz. Such a chip would be less than a 50th as fast as a 2GHZ P4, even if one attached zero-wait-state memory, but getting an 8088 to run that speed would still be just about impossible. If Intel hit the speed limit with their P3 design, then a new design which offered 10% fewer MIPS per MHz but could run at a 50% higher clock rate could be a net improvement. I suspect that's what happened here.
BTW, one advantage of the P4 versus other chips is what happens when the heat sink fails...
Another issue I've seldom seen considered is that in many cases what really counts is how well the system can perform when it's bogged down and thrashing. If I'm not waiting on my computer, I don't care whether it takes 1ms or 10ms to process my keystrokes. But when it's going into one of it's "big think" times, even a 20% speedup would be welcome.
No joke. Drop the sink on an Athlon and you can get out your checkbook, 'cause that baby's dead and gone...
One of my coworkers trashed one by forgetting to remove the adhesive-backed paper from the heat sink before putting it on the chip. Why they make those chips so touchy I have no idea. [It would seem like the logical thing to do would be to have the chip factory-bonded to a medium-sized piece of metal; heat transfer between the chip and the metal could be better than what's possible with conventional packaging, and heat transfer between the metal and an external heat sink could be better than that between the chip and a heat sink due to the larger surface area of contact. Am I missing anything in my thinking?]
I put the machine together, and it wouldn't start. I rechecked and double checked every connection. Nothing. I went back to the store and bought a new power supply and a new board. First the supply, then the board. No help. I pulled out the memory and put the memory from another machine. No help. I couldn't figure out what was wrong. I checked and replaced almost everything.
I figured the only thing left I hadn't checked was the CPU. So I went back to the store and bought a new CPU. While standing there, thinking about all I've been through, I had another thought. What if the problem is that the button on the case was shorted out? So for good measure I bought a new case, too.
I got home and put the MB into the new case. I pushed the button, and it started up immediately. It was the damn "on" button on the case that was defective.
Thermal expansion comes to mind. The chip carrier is ceramic, and the metal bonded to it will expand or contract at a different rate than the ceramic carrier, so it would probably pop right off just due to the heat stress before too long. Unless you had some sort of mechanical attachment, like bolting it directly to the carrier.
But that would still increase the size and weight of the thing. Aside from purely engineering problems, the reason I suspect they don't do something like that is because it leaves system manufacturers free to find a heat solution that fits their needs - a big block of aluminum works fine in a nice, roomy desktop, but you want something cleverer and more streamlined in a laptop, for example.
Heat dissipation is basically a function of the surface area of the object, so all a heat sink really does is increase the surface area - how it's actually configured really doesn't matter much. So maybe something thin and wide for a laptop or rackmount, versus big and blocky for a desktop. And anything you attach beforehand just increases the size and makes it less attractive to folks with space constraints.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.