![[H]ardOCP](/images/blank.gif){kind=small}
- Date:
- Tuesday , March 19, 2002
- Author:
- Steve Lynch
- Editor:
- Sean Quinn
- Google +1
Iwill XP333-R Mainboard
I think we started this review several months ago and it has certainly been a journey. Have a look at our experience with ALi, Iwill, and DDR333.
Board Layout (cont.)
When you purchase a new motherboard, it's a good idea to remove the heatsink on the northbridge and check for thermal compound. Although many companies put a great deal of thought and effort into including a fancy chipset cooler, they often fail to use thermal compound (as is the case here), thereby greatly reducing the efficiency of the cooler. We suggest that anyone OCing their board simply check for this before tweaking. It might make your overall experience much easier.
Although the DDR memory slots seem a little cramped at first glance, they aren't. It's just the layout of the board that makes it seem that way. The DDR voltage jumpers are located just to the right of the 3 DIMM slots in somewhat of an unusual spot. This makes it hard to get to the jumpers once the board is installed inside of a case.
The Corsair memory we used for this review is the latest in their product line, the XMS PC2700 DDR memory. Something that has changed from their previous high end memory, the XMS PC2400, is the addition of stylish black RAM sinks or "heat spreaders". Although the PC2700 didn't generate a significant amount of heat, the addition of heat spreaders ensures lower, more uniform temperatures across the memory modules.
The heat spreaders are not only efficient heat dissipaters, they also look nice too. Although you may say looks don't really matter, remember that if your favorite memory manufacturer starts using hot pink heat spreaders.
BIOS
As far as layout and settings are concerned, the BIOS is standard Award SoftMenu III stuff. Iwill did include a decent set of overclocking features like VCore adjustments, which range from 1.125v to 1.85v, with an added feature allowing you to "increase VCore 10%" for a maximum of 1.95v. A MicroStepping FSB setting allows for a range of bus speeds from 100MHz to 233MHz in 1MHz increments.
Something that mainboard manufacturers are doing to achieve higher bus speeds is adding the ability to change the clock divider in the BIOS. The selection ranges from 1/3 to 1/6, although we found that the board would refuse to boot using the 1/5 or 1/6 options.
You also have the option to change the DRAM performance to one of six settings: Failsafe, Slow, Normal, Fast, Ultra, and Ultra2. It should be noted that the XP333 refused to operate at any setting above "Fast". Attempting to use the Ultra or Ultra2 setting resulted in the system refusing to boot.
Although the initial benchmarks were taken with the Default Optimized BIOS settings, we did test the XP333R in as many BIOS configurations as possible for compatibility issues. We adjusted everything from the DDR settings, timings, CPU voltage, bus speeds, you name it. We will cover the results of our performance testing independently in the overclocking section of this review.
Testing Platforms
Often we benchmark motherboards using another product in the same class for reference purposes. Our motherboard "reviews" are conducted as reviews which focus on the product at hand and not as "comparisons". We feel it benefits you, the reader, to know about the products features, highlights, and drawbacks. . .as well as how fast it is.
Iwill XP333R: AMD AthlonXP 1800+ clocked at 1.533GHz (11.5*133), Alpha 8045D, 256MB Corsair XMS2700, Abit Siluro GeForce3 Ti500, NETGEAR FA310TX NIC, 40GB Maxtor ATA133 HD, Enermax 550 watt PSU.
EPoX 8KHA+: AMD AthlonXP 1800+ clocked at 1.533GHz (11.5*133), Alpha 8045D, 256MB Corsair XMS2700, Abit Silur GeForce3 Ti500, NETGEAR FA310TX NIC, 40GB Maxtor ATA133 HD, Enermax 550 watt PSU.
**All 3 DIMMS are filled during compatibility testing with 3 sticks of 256MB Corsair XMS2700. For benchmarking purposes, the board was tested with only one 256MB stick of memory.
Benchmarks
Benchmarks. This is where the rubber meets the road. We have amassed a collection of industry standard programs that we feel test the product's stability and overall performance. While 3D games play a part in testing the overall performance, running a few timedemos and Sisoft Sandra does not constitute a review. Although we believe Sisoft Sandra to be a fantastic tool that allows you to get an idea of your system's performance, the program itself is most useful when used in conjunction with other testing procedures. We will talk about that more later.
3D Winbench has been around for a while now, and while I think the Frames Per Second portion of the test is no longer fully valid in all cases, the CPU test based on those 3D tests is very interesting to check out. The score here is derived from the amount of polygons drawn during the 128 3D tests that the benchmark utilizes.
Business Applications
Business Winstone measures a PC's overall performance by running today's top-selling Windows-based 32-bit applications. Latest release: Business Winstone 2001 version 1.0 (11/07/2000). The Business Winstone tests are "market-centered" tests. Business applications are the popular applications employed by most users every day. Their final list of ten business productivity applications includes: Five Microsoft Office 2000 applications (Access, Excel, FrontPage, PowerPoint, and Word); Microsoft Project 98; Lotus Notes R5; NicoMak WinZip; Norton AntiVirus; and Netscape Communicator.
Not an encouraging start. The XP333R scores lower in both the initial tests. I had expected the 3DWinbench scores to be slightly lower than the KT266A board, but I thought the Business Winstone tests would have favored the XP333R.