Corsair fired off a pair of their XMS2 PC4300 modules for VL to cook in our labs. As the PCxxx rating tells you, these modules are clocked a little slower than their first batch of DDR2 modules (rated at PC5400) but to make up for it, Corsair tightened the timings to 3-3-3-8.
Overclocking

A will be used to test for reliability as well as stability during our overclocking tests. The card is quite expensive, but if you're testing a lot of ram, or you're a technician troubleshooting systems, this card is well worth the money.
We proceeded with overclocking, and tested the ram via the memory tests on the R.S.T. Pro2 until we got a failure. In order to maximize our OC potential, we used a Pentium 4 3.6 ES, whose multiplier can be adjusted from 14-18 will allow a bit of FSB headroom on our ASUS P5AD2-E motherboard. We've hit ~270FSB regularly with this board (albeit, not for long periods), so it should give us some headroom in overclocking. The power supply used is Cooler Master's RealPower 450W.
Timings
|
Max DDR2 OC
|
Pass R.S.T. Pro2
|
3-2-2-8
|
533
|
Yes
|
3-2-2-8 |
556*
|
No
|
3-3-2-8 |
580
|
Yes
|
3-3-3-8 |
642
|
Yes
|
4-4-3-8 |
718
|
Yes
|
* Denotes the ram failed the R.S.T. Pro2 tests, but still managed to boot into Windows.
While we did not include every overclocking test/result in our results above, we did include some that were worth noting. At DDR2-556, the modules failed our R.S.T. Pro2 tests, yet they did run fairly stable nonetheless. Note that I said "fairly". While we were able to perform benchmarks, occasionally we would lock up the system, even though it succeeds in completing the failed benchmark in later attempts.
At 3-3-3-8, the ram was able to come close to the rated DDR2 speed of the 5400C4PRO modules, and surpass it with tighter timings at DDR2-718. At the 4300C3PRO's rated DDR2-538, we managed to improve the timings slightly at 3-2-2-8.
Test Setup
ASUS P5AD2-E: Intel P4 3.6, 2 x 512MB Corsair TWIN2X1024-4300C3PRO, ATI X600 XT, 120GB Seagate SATA 7200rpm, Windows XP SP1, ATI Catalyst 5.1.
Testing software will consist of the following:
SiSoft Sandra 2004 Memory - Our standard synthetic test to establish a baseline.
PiFast - A good indicator of CPU/Motherboard performance is version 4.2, by Xavier Gourdon. We used a computation of 10000000 digits of Pi, Chudnovsky method, 1024 K FFT, and no disk memory. Note that lower scores are better, and times are in seconds.
Quake 3 - While it's old and moldy, it still has some value as a system level benchmark.
The comparison memory will be Corsair's 5400C4PRO clocked at DDR2-667 but with tweaked timings of 4-4-3-10, which were the best we were able to manage while passing the R.S.T. Pro2 tests. The memory in the charts below are to be read: Model - Timings - DDR2 speed.
SiSoft Sandra 2004 Memory

No surprises here as we see the performance scales proportionately with the clock speed and timings.
PiFast

Some improvements can be seen in PiFast as we go up in clock speed. With the frequencies being equal, we can see marked improvements when we tighten the timings.
Quake 3: Arena Demo Four, Min Detail @ 640

Quake 3 shows that memory timings can make quite a difference in performance, but as we've observed in the past, memory frequency plays a larger role once we reach DDR2-533 and beyond.
Final Words
While the Corsair TWIN2X1024-4300C3PRO had a very good showing here today, there are a few things to note if you're in the market for performance DDR2 ram. While our modules ran great at DDR2-667, we began to hit some road blocks at DDR2-680 and up. The ram continuously failed our R.S.T. Pro2 tests, and the system exhibited some quirky behavior. In contrast, the TWIN2X1024-5400C4PRO kit was able to reach above DDR2-700 without too many problems. Granted, if you own a 915P or 925X setup, unless you can hit 1066FSB, your money is somewhat wasted on high frequency ram.
Of course, Corsair isn't marketing this kit as a high frequency replacement, and while we wouldn't call a stable DDR2-667 overclock low speed, if you're looking at overclocking to 1066FSB and beyond, this kit isn't for you. Owners of 800FSB setups will be satisfied with the 4300C3PRO's abilities to run at tight timings (tighter than anything we've managed with the 5400C4PRO) at DDR2-667 and lower. As we've seen today, there is an improvement in performance in those scenarios, and if you fall under this category, the Corsair TWIN2X1024-4300C3PRO will be well worth the purchase.

Pros: Very fast, great overclocker, tight timings.
Cons: Didn't reach the same frequencies as overclockable DDR2-667.
Bottom Line: Unless you're planning to heavily overclock your ram or system, high frequency DDR2 kits are a bit of a waste of money considering their conservative timings. The 4300C3PRO is fully suited for overclocking, and so long as you stay at around DDR2-667 and lower, the lower latencies will make a bigger impact on performance.
If you have any comments, be sure to hit us up in our forums.