![]() |
Cellshock DDR2 1066 MHz CL5 2 GB Kit |
|
Test Setup
| Test System | |
|---|---|
| CPU: | Intel E6300 Conroe 1.8 GHz, 2 MB Cache |
| Motherboard: | ASUS P5B Deluxe BIOS 1004 Intel P965 C1 |
| Video Card: | PowerColor X800XL Pro 16 PCI-E |
| Harddisk: | Samsung P80 80 GB |
| Power Supply: | Ultra V-Power 450W |
| Software: | Windows XP SP2, Catalyst 7.6 |
Performance & Overclocking

Once the memory was installed, the SPD programming of the memory is correct and reflects the two possible settings. It also featurs EPP, so that you may let the BIOS set the correct speed and timing for you.
Testing the memory is done by a specific method. First we run the kits at the intended settings - in this case 2.3V and 1066 MHz with CL5-5-5-15. This worked flawlessly. Then the timings were set at 3-2-2 with 667 MHz. The memory did not boot at these settings. When relaxing the timings a bit to 3-3-3, the system became stable once more at 1.95V. This is where the smile on my face became bigger and bigger after each test run. The first milestone is 800 MHz at CL3-3-3, the second came in form of 1000 MHz CL4-4-4 and then even 1066 MHz at the same setting with 2.35V. The kit even manages 1066 MHz CL5-5-5-15 with default timings of 1.85V. The maximum this kit managed was 1170 MHz stable at 5-5-5-15 and 2.35V. Cellshocks warranty only coveres an operation up to 2.3V, but even at 2.45V and CL5-5-5-15, the memory did not manage 1200 MHz.
| Cellshock 1066 MHz CL5 2GB Kit | ||||||||
|---|---|---|---|---|---|---|---|---|
| CPU Clock & Memory Ratio | Memory Speed | Memory Timings | Everest Read | Everest Write | Everest Latency | Quake 3 Timedemo | 3DMark 2001SE | SuperPi Mod 1M |
| 7 x 320 3:5 | 533 MHz | 5-5-5-15 1.85V | 8064 MB/s | 5814 MB/s | 57.7 ns | 439.8 fps | 27799 | 24.01 s |
| 7 x 324 3:5 | 540 MHz | 5-5-5-15 2.15V | 8365 MB/s | 6041 MB/s | 55.2 ns | 456.5 fps | 28392 | 23.07 s |
| 7 x 351 3:5 | 585 MHz | 5-5-5-15 2.35V | 8752 MB/s | 6391 MB/s | 53.4 ns | 478.5 fps | 29179 | 21.96 s |
| 7 x 294 2:3 | 442 MHz | 4-4-4-12 1.85V | 7262 MB/s | 5373 MB/s | 65.6 ns | 404.8 fps | 26373 | 26.15 s |
| 7 x 333 2:3 | 500 MHz | 4-4-4-12 2.15V | 8207 MB/s | 6070 MB/s | 58.1 ns | 453.9 fps | 28450 | 23.11 s |
| 7 x 355 2:3 | 533 MHz | 4-4-4-8 2.35V | 8768 MB/s | 6461 MB/s | 54.7 ns | 471.6 fps | 29238 | 21.64 s |
| 7 x 277 4:5 | 347 MHz | 3-3-3-10 1.95V | 6341 MB/s | 5059 MB/s | 78.6 ns | 374.2 fps | 24638 | 28.43 s |
| 7 x 320 4:5 | 400 MHz | 3-3-3-10 2.25V | 7328 MB/s | 5793 MB/s | 68.3 ns | 427.7 fps | 26681 | 24.68 s |
| 7 x 336 4:5 | 421 MHz | 3-3-3-10 2.35V | 7623 MB/s | 6108 MB/s | 64.8 ns | 439.8 fps | 27064 | 23.50 s |


As you can see in the Voltage graph, it becomes apparent, that the maximum memory is rises steadily with the increase of voltage. Then when the maximum is reached, simply change the CL setting and continue at lowest voltage from the same point. This makes this one of the most versatile memory we have tested so far. Remember, that every kit performs differently, so your milage may vary!

