August 6, 2001

Are top-of-the-line CPUs worth the premium price?

Author: JT Smith

- By Jeff Field -

Although they may come from different manufacturers, there's one thing the Pentium 4 1.8, the Athlon 1.4, the Duron 950 and the Pentium III 1.13/1.2 all have in common -- a high price premium for being top-of-the-line CPUs. This pricing isn't just limited to CPUs, either, it happens with all sorts of hardware, including hard drives and monitors. Is having the top-of-the-line worth the premium price?
If you look on Pricewatch for the Athlon 1.4GHz and 1.33GHz CPUs, you'll see something interesting - an Athlon 1.33 will cost you, roughly, $133 with shipping. On the other hand, the top-of-the-line Athlon 1.4 will cost you somewhere around $166 with shipping. That's a 25% increase in cost from the 1.33 to the 1.4 model. A 1.33GHz CPU is 1330MHz, and a 1.4GHz is 1400MHz, and if you do the math, the 1.4GHz has only a 70 MHz, or 5%, increase in clock speed. This is why price-to-performance ratios are so important when purchasing a CPU: to find out what kind of bang you'll get for your bucks. You wouldn't pay 28% more for an extra egg in a dozen, why should you pay the equivalent when buying a CPU?

It isn't just Athlons, either: Almost every CPU you can find on Pricewatch is sold this way. The Duron 950 costs 40% more than the 900, the Pentium 4 1.8GHz costs 78% more than the 1.7GHz model, and the Pentium III 1.13 costs 75% more than the Pentium III 1GHz. And I don't want to leave the impression that the price differences are just for CPUs either -- a 100GB hard drive will cost 45% more than a 80 gig, but only give you 25% more room. A 22-inch monitor with 20-inch viewable costs $685, while a 21-inch with a 19.9-inch viewable screen costs $415. In that case, you are paying a company 65% more money, for 0.5% more screen. Somehow, that doesn't seem "cost effective" to me. To illustrate my point, I am going to take the aforementioned Athlon 1.33GHz for a spin, comparing it to the 1.4GHz.

Performance comparison
System specifications
Gigabyte 7DXR Motherboard
256 Megs DDR PC2100 SDRAM from

Western Digital 7200 RPM 10.2 Gig Hard Drive

3Com 3C905TX-C 10/100 NIC (PCI)

400 Watt ATX 2.03 Power Supply
Gigabyte GeForce 3 64MB AGP

Mandrake 8.0 with Kernel 2.4.3

For performance comparison purposes, similarly configured systems are used, where only the memory type (PC133), the processor type and the motherboard are different.

Kernel compiles

To test both the board's stability and speed, I ran three sets of Linux kernel compiles on this board. One is a normal, "uniprocessor" make, or make -j1, which is the default. This uses one process, and does not always maximize system usage. I then did make -j2, which spawns a second process. The last test I run is with make -j3, spawning two extra processes. I do this for several reasons -- to find the "sweet spot" for the board/CPU, as well as to stress the system as much as possible when trying to rate its stability. Also, the kernel is extremely useful as a measure of integer performance. In order to compile the kernel, I untarred kernel 2.4.6, ran "make config" and used the default values. (In other words, I held down the "Enter" key.)

Kernel 2.4.6 Compile Times (Minutes:Seconds)
Board -j1 -j2 -j3
Athlon 1.4 - DDR 4:29 4:20 4:21
Athlon 1.33 - DDR 4:32 4:28 4:27

In this case, your $33 buys you six entire seconds in a kernel compile. Thats $5.50 a second.

POVRay Benchmarks

POVRay is a multi-platform raytracing program. It is a floating point intensive task and serves well to help measure the floating point performance of a CPU. For more information on this benchmark, head to the official POVBENCH homepage. Once you've obtained POVray, run this command to perform the benchmark povray -i skyvase.pov +v1 +ft -x +mb25 +a0.300 +j1.000 +r3 -q9 -w640 -H480 -S1 -E480 -k0.000 -mv2.0 +b1000 from your command prompt. Results are in seconds.

POVRay (Seconds)
Board Result
Athlon 1.4 - DDR 15 seconds
Athlon 1.33 - DDR 16 seconds

In this case, you're actually paying $33 a second, because POVRay gets a whopping one second increase in performance.

Quake 3 Arena Timedemos (Frames Per Second)
Board 640x480 800x600 1024x768 1200x1024 1600x1200
Default Quality
Athlon 1.4 - DDR 183.6 182.6 172.8 126.1 90.7
Athlon 1.33 - DDR 180.6 178.7 169.2 123.7 90.3

Here, we find a gain of three frames a second, and at the very high end only 0.4 frames a second. It is even worse on slower boards than a GeForce3 -- on those boards resolutions above 800x600 are barely affected by CPU speed.

What's the moral of the story here? If there is one, it is to shop smart. That 1.33GHz Athlon will perform nearly as well as the 1.4, and it will leave some extra money in your pocket, perhaps for another good upgrade. When buying components, always factor in what you need, as well as how much you're willing to pay, and do not let yourself get caught up in that silly game where you must have the highest speed grade, even if it means exorbitant premiums. Do your research, find out what advantages higher models really have over lower models -- or you may end up paying more for a sound card that's only benefit is some additional Windows software, or a CPU that costs a hundred dollars more but doesn't really offer superior performance.


  • Unix
Click Here!