- By Jeff Field -
It was only a month ago that AMD released the 1 GHz Duron, the first Duron to be based on the Morgan core. AMD has again raised the bar for the low-cost computing market, moving the Duron up to 1.1GHz with its latest release. Again, we are forced to wonder just how much of an improvement it really is.Clock speed and price/performance
Long gone are the days where mere clock speed determines speed; if that were still true, a 1.4GHz Thunderbird wouldn't beat out a higher clocked P4 with ease. However, in the same class of CPU, such as Durons, clock speed still is a deciding factor and so we wonder just how much 100MHz matters in the Duron line.
Another consideration is the price to performance ratio -- right now a 1 GHz Duron costs you about $68, according to Pricewatch; a 1GHz Athlon is available for $70. AMD is killing its own value segment -- with the availability of high-end CPUs, there really isn't enough of a difference between the value and power options to fuel value purchasing. Despite its disadvantages the Duron still deserves fair coverage, which is what I aim to do.
As for the chip itself, nothing has changed since the release of the 1GHz Duron. The Duron 1.1, like the 1.0, is based on the newer Morgan core, which I covered in my Duron 1GHz Review. Other than the increase in clock speed, this chip is exactly the same.
AMD Duron 1.0GHz or AMD Duron 1.1GHz
256 Megs PC2100 DDR SDRAM from Crucial.com
Gigabyte 7DXR motherboard
Western Digital 7200 RPM 10.2 Gig Hard Drive
3Com 3C905TX-C 10/100 NIC (PCI)
400 Watt ATX 2.03 Power Supply
Gigabyte GF3000 GeForce 3 64MB AGP
Slackware 8.0 with Kernel 2.4.9 and XFree 4.1.0
In order 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.
|Kernel 2.4.6 Compile Times (Minutes:Seconds)
Lower numbers are better
This shows roughly a 5% increase in performance, from a 10% increase in clock speed. No revelations here, really; the Duron 1.1 is slightly faster than the 1.0, as you would expect with a chip clocked slightly higher.
POVRay is a multi-platform raytracing program. It is a very 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. The command to run for this benchmark, once you obtain POVRay, you run povray -i skyvase.pov +v1 +ft -x +mb25 +a0.300 +j1.000 +r3 -q9 -w640 -H480 -S1 -E480 -k0.000 -mv2.0 +b1000 from the command prompt. Results are in seconds.
Lower numbers are better
|Duron 1.0GHz||21 seconds|
|Duron 1.1GHz||19 seconds|
In povray, we see a difference of about 10%, directly in line with the difference in clock speed.
Distributed.net Client Benchmark
Distributed.net is a distributed computing network that works on various distributed computing contests. The contests use primarily integer numbers while performing their tasks, and therefore serve as an excellent benchmark for overall integer performance of properly optimized software.
|Distributed.net Client Benchmarks
Higher numbers are better
|CPU||RC5 Core 6||OGR Core 0|
|Duron 1.0GHz||3,544,709 keys/sec||7,541,546 nodes/sec|
|Duron 1.1GHz||3,903,400 keys/sec||8,280,414 nodes/sec|
Distributed.net depends on the CPU in much the same way POVRay does, requiring it for calculations that benefit directly from increases in clock speed. Here we see this as the 10% increase in clock benefits Distributed.net by 10%, as you might expect.
If my conclusions were reached merely on the technological merits of a CPU, the Duron would do well here -- it is, all in all, an excellent CPU. In the current market, however, the Duron has no real place. It is a bargain CPU in a bottomed-out CPU market. Same-clock Athlons are selling for a few dollars more in most cases, and in the case of the Duron 1.1, it's actually more expensive than AMD's 1.4GHz offering, although that is only temporary.
Still, with the exception of the very low clock speed Durons, which can be had for about $25 on Pricewatch, the Duron line doesn't offer any real value compared to the Athlon, at least not at this juncture. As AMD readies a new line of CPUs, the Duron may indeed find its place, but for now, I recommend the Athlon over the Duron, as the Athlon is worth the extra few dollars.