nForce2 6-way Motherboard Roundup - December 2002
by Evan Lieb on December 4, 2002 6:53 PM EST- Posted in
- Motherboards
Epox 8RDA+: Stress Testing
The abundant array of divider options available in the 8RDA+
BIOS allowed us to test this board in several different areas and configurations,
including:
1. Chipset and motherboard stress testing was conducted by running the FSB at 194MHz.
2. Memory stress testing was conducted by running RAM at 333MHz and 390MHz with as many modules populated as possible at the most aggressive timings possible.
Front Side Bus Stress Test Results
We were pleased (but not surprised) that the Epox 8RDA+ was able to handle such a high FSB with such exceptional stability. At 194MHz FSB, 24 hours straight of Prime95 torture tests was not able to bring the 8RDA+ to its knees. Running Prime95 is a great way to really stress your system, as we have very rarely encountered an issue with a system that was able to pass 24 hours or more of Prime95 torture tests. As Prime95 was running in the background, we reran our gaming suite as well as everyday apps like Word and Excel in addition some data compression here and there. We also managed to rerun SPECviewperf and XMPEG, and added in Sciencemark stress tests (even though we didn't include any Sciencemark numbers in this review). In the end, the 8RDA+ was able to easily pass all these stress tests at 194MHz FSB. Anything over 194MHz FSB did not sufficiently meet our stress testing requirements.
Memory Stress Test Results
Lets see how well the Epox 8RDA+ is able to handle three memory modules installed and running at 333MHz DDR. Here were the timings we were able to achieve:
Stable
DDR333 Timings |
|
Clock
Speed:
|
166MHz
|
Timing
Mode:
|
N/A
|
CAS
Latency:
|
2
|
Bank
Interleave:
|
N/A
|
Precharge
to Active:
|
2T
|
Active
to Precharge:
|
4T
|
Active
to CMD:
|
2T
|
Command
Rate:
|
N/A
|
These are great timings (in fact the A7N8X was able to achieve the exact timings in this exact same scenario). There shouldn't be a single user out there that isn't satisfied with these aggressive timings. We must again note the lower Precharge to Active setting, which other motherboards (not based on the nForce2 chipset) either cannot achieve or simply don't have the option of running at. Also, remember that these timings were achieved with the CPU Interface enabled, which boosts performance a bit.
This final memory test demonstrates the highest frequency the Epox 8RDA+ is able to handle with three memory modules at CAS2/2T/4T/2T. Here were our results:
Stable
DDR390 Timings |
|
Clock
Speed:
|
190MHz
|
Timing
Mode:
|
N/A
|
CAS
Latency:
|
2
|
Bank
Interleave:
|
N/A
|
Precharge
to Active:
|
2T
|
Active
to Precharge:
|
4T
|
Active
to CMD:
|
2T
|
Command
Rate:
|
N/A
|
While we weren't able to hit 400MHz, we did get to 390MHz. With timings like CAS2/2T/4T/2T, we certainly have nothing to complain about. Keep in mind though, if you want the best performance at DDR390 you need to run your FSB at 195MHz, or 390MHz DDR. We were able to reach 194MHz, so that's pretty close to what the 8RDA+'s potential clock frequency ceiling is with all memory banks filled.
The memory overclocking score for the 8RDA+ turned out to be about the average out of all the nForce2 motherboards tested.
Stable
Memory Overclocking Results |
|||
Memory
|
Memory
Clock
|
FSB
|
VDIMM
|
Corsair
CAS2.0 DDR400
|
412MHz
|
165MHz
|
2.63V
|
We had to ensure the various memory timings and frequencies
we were able to operate at were fully stable. Therefore, we conducted 24 hours
of Prime95 torture tests, ran various Super Pi stress tests, and reran various
benchmarks like SPECviewperf 7.0 and XMPEG. All in all, we didn't encounter
any issues with any of the listed memory timings and frequencies we were able
to reach during these stress tests.
SoundStorm Stress Test Results
We ran the same sound tests on the 8RDA+ as we did with all the other nForce2 motherboards tested here today. The following demos were run with SoundStorm sound enabled:
1. Jedi Knight II: Passed
2. Quake 3 Arena: Passed
3. Unreal Tournament 2003: Passed
4. Comanche 4: Passed
5. Serious Sam: Passed
There have been some minor, ongoing issues with NVIDIA's nForce1 APU drivers since the nForce1's introduction last year. Some users have experienced a Blue Screen of Death with nForce sound enabled in certain games. However, the latest nForce2 unified driver package (v 2.00) has seemingly fixed the sound issues that nForce1 users were experiencing. This is backed up by our own test results, which show that SoundStorm is able to smoothly play five of some of the most popular games out there.
1 Comments
View All Comments
c627627 - Monday, July 21, 2003 - link
MSI contradicts your reviewhttp://www.anandtech.com/mb/showdoc.html?i=1759&am...
You say:
"12/04/02 UPDATE: MSI sends word that the K7N2-L indeed does not have a PCI bus lock at 33MHz."
To this day, that was used as ultimate proof that the original MSI nForce2 mobo does not have a PCI lock.
Today MSI Senior Moderators said:
"they are wrong,and who ever told them it did not from msi"
(!)
Source:
http://www.msi.com.tw/program/e_service/forum/thre...
Won't you please settle this for us.