|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Best viewed @ 1024 x 768 and higher
Closer Look: Memory timings are set to 4-4-4 on the PC2-6400 compared to the PC2-9200 that have 5-5-5 memory timings. Voltages can be applied from a stock 2.0v up to a 2.4v warranty maximum which still covers you if something were to go wrong. I haven't had any problems with any OCZ to date. So, I will put these through a kind of 'tweaked hell' and not worry about problems occurring. The only thing anyone has to worry about is not dropping the RAM inside your system during installation. These are heavy enough to cause some damage if they slip from your fingers. The two biggest differences between the PC2-9200 and PC2-6400 Flex kits are the level of overclocking of the 9200 kit and the smaller price tag of the 6400 kit. The modules in the 9200 kit are the coveted Micron D modules, while the 6400 kit uses ProMOS chips. We know that Microns can hit top speeds of 1066MHz at 5-5-5-15 with or close to the maximum voltage being offered. These ProMOS chips aren't the same exact ones found in some of the CAS3 memory modules, but I am hoping that they can at least reach 1000MHz overclocked. Of course this is just a reasonable guess based on other modules tested and the fact that this is an extreme dual channel memory kit. It's a pretty cool thing when looking at two FlexXLC sticks on a motherboard. It's even better looking when you have four. They add a pretty unique look to your system as well as some nice system performance. One thing that I originally didn't think about was how difficult fitting each kit with tubing would be inside the system. Two modules alone mean an extra 4 sets of 1/4 tubes providing the water cooling. Add another two modules and you now have to manage 8 pieces of 1/4 tubing which quickly clutter up the inside of your system. It will take a little bit of planning to cut and make the cooling system look good. Then again, if the memory works fine without the water cooling, just work with the stock modules. It looks perfectly fine. Let's get on with the benchmarks and testing. Installation and Testing: After installing the memory, I immediately went in to the BIOS to see what timings were automatically detected. While the BIOS memory timings set to AUTO, the system booted up using 5-5-5 timings rather than 4-4-4. CPUZ also revealed different memory timings and stock speeds. Manually setting the memory timings to 4-4-4 in the BIOS while leaving the Command Rate to AUTO revealed another oddity. The memory and system would operate at 4-4-4 2T and at 1T Command Rate after restarting and depending on how much voltage was applied. At default DIMM voltage, the Command Rate would default to 2T. Changing the DIMM voltage to 2.2v would automatically set to 1T. I was unable to narrow down the exact reason
why, but I suspect that both the RAM and motherboard are equally the culprits
depending on how much voltage is being applied to the memory via the BIOS. This
is an unexpected plus out of the test system. If the ram is perfectly stable at
4-4-4 1T Command Rate, then it's not a bad thing at all. Most the quality
modules I have worked with that run CAS4 at 800MHz have also operated fine at
3-3-3 1T Command Rate and work very well. The 1T Command Rate can equate in to
better performance just like the DDR day when 1T modules were desired over 2T
modules due to the gains.
<< Intro & Specs | Testing & Conclusion >>
Got News? Send 'em in!
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
| |
|