 |
OCZ EL DDR PC-3200 Titanium |
|
Test Setup
| Test System | |
|---|---|
| CPU: | AMD Athlon64 3000+ Venice |
| Motherboard: | DFI LanParty NF4 |
| Memory: | 2x 512 MB OCZ EL DDR PC-3200 Titanium |
| Video Card: | ATI Radeon X850 XT PCI-E |
| Harddisk: | Maxtor Diamondmax 160GB |
| Power Supply: | HEC PurePower 475 |
| Software: | Windows XP SP2, Catalyst 5.3 |
Performance
The first test we did, was test how the memory performs at a stock frequency of 200 FSB with the default timings recommended by OCZ.Next, we increased the memory clock step by step to find out what the maximum clocks are when running at the default voltage of 2.6V. We reached 216 MHz, which is a rather dissapointing result.
Once we increased voltage to 2.8V, things were starting to become strange. The maximum overclock dropped a few MHz, to 213 MHz. With a really extreme voltage of 3.3V we could get an extra 2 MHz over the default 2.6V. So basically increasing voltage does not do anything at all.
The last test "JEDEC DDR400A" is for comparison with a generic DDR module running at JEDEC standard timings.
| OCZ EL DDR PC-3200 Titanium | ||||||||
|---|---|---|---|---|---|---|---|---|
| CPU Clock & Memory Ratio | Memory Speed | Memory Timings | Everest Read | Everest Write | Everest Latency | Quake 3 Timedemo | 3DMark 2001SE | SuperPi Mod 1M |
| 8 x 200 1:1 | 200 MHz | 2-3-2-5 2.6V | 4095 MB/s | 2144 MB/s | 51.3 ns | 218.1 fps | 19154 | 50.80 s |
| 8 x 216 1:1 | 216 MHz | 2-3-2-5 2.6V | 4397 MB/s | 2030 MB/s | 47.4 ns | 234.9 fps | 20174 | 47.28 s |
| 8 x 213 1:1 | 213 MHz | 2-3-2-5 2.8V | 4334 MB/s | 1959 MB/s | 48.1 ns | 231.5 fps | 19889 | 47.79 s |
| 8 x 215 1:1 | 215 MHz | 2-3-2-5 3.1V | 4364 MB/s | 1976 MB/s | 47.5 ns | 233.1 fps | 20178 | 47.59 s |
| 8 x 219 1:1 | 218 MHz | 2-3-2-5 3.3V | 4428 MB/s | 2006 MB/s | 47.0 ns | 236.6 fps | 20250 | 46.86 s |
| JEDEC DDR400A | 200 MHz | 2.5-3-3-8 2.6V | 3952 MB/s | 1939 MB/s | 53.5 ns | 213.3 fps | 18548 | 51.97 s |
We also tried relaxed timings of 2.5-3-3-x and 3-3-3-x, but without much success, a few MHz could be gained, but this is not worth it if you take the performance loss into consideration.
Base performance is good. When overclocking, performance scales accordingly, except for the Sandra Write benchmark which shows best performance when the module is running at 200 MHz stock, everything above that results in less write bandwith, even though the other benchmarks show that total performance does improve.
For easier comparison with other modules, we set a maximum voltage of 3.1V und tested until we found the highest clock frequency and fastest timings for this memory. The benchmarks Everest Read, Everest Write and Quake 3 were run. We then calculated the performance increase in percent compared to some standard DDR400 memory running at JEDEC DDR400A (2.5-3-3-8). The average percentage of the three benchmarks is listed in following table:
