We took a very methodical approach to configuring our test systems for this review. The first thing we did was enter the system BIOS and set the memory timings for optimal performance. On our Asus A7N8X v2.0, this meant running the memory in Dual-Channel mode, set to run synchronously with the processor's FSB (166MHz for the 3000+ and 200MHz for the 3200+), with the timings set to 5-2-2 with a CAS latency of 2. The memory was similarly configured on the Dual-Channel, i875 powered Intel test bed. After configuring the BIOS, we booted the systems, formatted the hard drives and installed a fresh copy of Windows XP Pro with Service Pack 1. When the installation was complete, we hit the Windows Update site and installed all of the updates currently available (Except for Windows Media Player 9 and the updates associated with Windows Messenger). We then disabled Auto-Updating, System Restore and Windows Messenger and then completely removed Messenger from the system. All of the necessary drivers for the mainboard and other hardware were then installed and the hard drive was de-fragmented. Lastly, we setup a 768MB permanent swap file, installed all of the necessary benchmarking software and ran the tests. Now, lets get ready to rumble...

SANDRA (the System ANalyzer, Diagnostic and Reporting Assistant) is an information and diagnostic utility developed by the good folks at SiSoftware. SANDRA isn't just for benchmarking though, it can also provide a host of other useful information about your hardware and operating system. We began our testing with four of the built-in tests that partially comprise the SANDRA 2003 suite (CPU, Multimedia, Memory and Cache sub-system). We ran these tests at the Athlon XP 3200+ processor's default clock speed of 2.20GHz and compared it to AMD's previous flagship processor, the 3000+. While we were at it, we threw a P4 3.06 and a P4 3.0C, both with Hyper-Threading enabled, in for good measure...

ATHLON XP 3200+ 2.20GHZ (11X200) CPU	ATHLON XP 3000+ 2.16GHZ (13X166) CPU	PENTIUM 4 3.0CGHz 3.0GHZ (15X200) CPU	PENTIUM 4 3.06GHz 3.06GHZ (23X133) CPU
ATHLON XP 3200+ 2.20GHZ (11X200) Multimedia	ATHLON XP 3000+ 2.16GHZ (13X166) Multimedia	PENTIUM 4 3.0CGHz 3.0GHZ (15X200) Multimedia	PENTIUM 4 3.06GHz 3.06GHZ (23X133) Multimedia
ATHLON XP 3200+ 2.20GHZ (11X200) Memory	ATHLON XP 3000+ 2.16GHZ (13X166) Memory	PENTIUM 4 3.0CGHz 3.0GHZ (15X200) Memory	PENTIUM 4 3.06GHz 3.06GHZ (23X133) Memory
ATHLON XP 3200+ 2.20GHZ (11X200) Cache	ATHLON XP 3000+ 2.16GHZ (13X166) Cache	PENTIUM 4 3.0CGHz 3.0GHZ (15X200) Cache	PENTIUM 4 3.06GHz 3.06GHZ (23X133) Cache

The SANDRA CPU results put the 3200+ ahead of the 3000+ by a few points, courtesy of its higher clock speed, but the Hyper-Threading enabled P4s definitely flex their muscles here, especially in the FPU tests where SSE2 is used (Note: Without SSE2, the Athlons take the lead). The Multimedia test tells basically the same story, with the Athlons playing second fiddle to the P4s. The trend continued in the Memory Bandwidth tests, which is an area where the Pentium 4s really shine. The 3.0C (800MHz bus) P4 / Canterwood combo posted some impressive numbers that were approximately 65% higher than the 3200+ / nForce 2 combo. When comparing the Athlons to each other, the 20% increase in the 3200+'s bus speed, up from 333MHz on the 3000+ to 400MHz, results in an approximate 15% gain in available memory bandwidth according to SANDRA.

We continued our synthetic testing with Futuremark's PCMark2002 benchmarking suite. Like other synthetic benchmarks, it's difficult to translate PCMark2002 scores into "real world" performance. However, because it is readily available, easy to run, and produces repeatable, comparable results, PCMark2002 has become a regular here in the H.H. labs. We ran PCMark2002's "CPU" and "Memory" performance modules on all four of the test systems. FYI, the CPU module incorporates the following tests:

PCMark's CPU performance module puts the Athlon XP 3200+ marginally ahead of the 3000+, but significantly behind either of the Pentium 4s. The P4 3.0C held and approximate 10% lead over the 3200+. It's interesting to note that system bus speeds had virtually no affect on this test. PCMark shows the P4 3.06 with its 533MHz bus outperforming the slightly lower clocked 3.0GHz CPU with an 800MHz bus. I'd be willing to bet system bus speeds influence the memory performance tests though...lets find out...

Raw read, write, and read-modify-write operations are performed starting from a 3072 kilobytes array decreasing in size to 1536 KB, 384 KB, 48 KB and finally 6 KB. Each size of block is tested two second and the amount of accessed data is given as result. In the STL container test a list of 116 byte elements is constructed and sorted by an integer pseudo-random key. The list is then iterated through as many times as possible for 2 seconds and the total size of the accessed elements is given as result. There are 6 runs of this test, with 24576 items in the largest run corresponding to a total data amount of 1536 KB, decreasing in size to 12288 items (768 KB), 6144 items (384 KB), 1536 items (96 KB), 768 items (48 KB) and 96 items in the smallest run corresponding to 6 KB of total data.

The Memory performance module definitely shows much larger performance deltas between the different platforms, but it becomes evident the Athlon systems simply can't offer the same amount of memory bandwidth as the Pentium 4's quad-pumped architecture. The 3200+ with its 400MHz bus posted a score about 12% higher than the 3000+, but that was still approximately 42% lower than the P4 3.0C.