Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home	.	.
Prices: Mainboards ABIT ASUS Chaintech Shuttle Soyo Tyan CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ Search Prices:

LOSTCIRCUITS	SHORTCUTS: Another $64-bit Question PanoramaFactory Revisited Test Configurations Wizard and Stitch Alignment and Geometry Transformations Redering and Blending The Final Crop Conclusions Give Us Some Feedback on this Review
Intel Pentium4 600 64-bit Performance PanoramaFactory Revisited
(Review by MS March 2, 2005)

Test Configurations

We used the following test configurations:

Test Procedures

Time Demo (Four Pictures)

Panorama Factory was run on a fresh installation of Windows XP-64, Windows Server 2003-64. We did not find any differences with respect to performance on the AMD platform between the different beta releases. We already mentioned that only the latest version of WindowsXP64 installed on the P4 platform. We ran the 32-bit and 64-bit time demo of four pictures for timing comparison using either the native 64-bit support of WinXP64 or else the WoW emulation of a 32-bit environment. All benchmarks were run 4 times and we took the average of all runs for comparison.

Full Panorama

It is easy to claim that there is full 64-bit support for memory addressing on a 64-bit processor and, quite frankly, we have no reason to doubt it. On the other hand, if there is a benchmark out there to prove the claims, there is no excuse for not running it. As we showed already on the last page, the P4 ExtremeEdition had no problem addressing more than the 32-bit memory address space, however, at its native operating frequency of 3.73 GHz, the ExtremeEdition did not finish the benchmark. Instead, at the beginning of the cropping procedure, that is, at the point where the system apparently uses all available resources, the system crashed every single time, regardless of whatever variables were defined in the mainboard's CMOS setting - including relaxation of memory latencies to 5:5:5:15 (to make sure that we were not running into an issue with overall system memory density and the associated load and clock skew that could negatively affect timings).

However, the mere fact that at the onset of the cropping even the Task Manager came to a total stand-still indicates that the entire system, including CPU, was running under full load, which could result in either overheating or else bogging down of the VRM (since the CPU might draw as much as 145W under these extreme conditions). Note that we were using the more powerful Intel cooler rather than that supplied with the 600 series. One easy way to separate the rest of the system from CPU issues is to clock down the processor, that is, we were running the ExtremeEdition at 14 x 200 MHz (2.8GHz) and at that speed, the system did not crash.

next page: => Benchmark Results =>