Popping the Hood

Removing heat spreaders is not recommended and should be avoided especially with BGA type memory. There are several advantages to removing the heat spreaders, foremost would be the ability to identify the IC's. Identifying the ICs isn't going to give a 100% accurate description of the memory's potential; it does help to know if there are particular attributes about the IC's. For example some may react better to overvolting then others, or run at tighter latencies. In most cases the manufactures' preprogrammed or SPD (Serial Presence Detect) settings work the best. These settings usually represent the most stable combination of speed, latency and operating voltage. In so far as the heat spreader itself, to be honest I've found very few designs (~30%) which actually help dissipate heat. In many cases the double-sided tape acts as an insulator rather then a conductor of heat. Unfortunately heat spreaders are used to conceal the ICs for any number of reasons.



As we discussed in the previous section OCZ chose "blank" ICs for their 800MHz DDR2 part. We'll not completely blank the original Fab stamped the packaging specs on the chips.



Taking a closer look.



The opposite side of the PCB reveals a part number.



Testing

Intel Test System
CPU	Pentium-D 930 Retail 3.0GHz Dual Core 2MB x2 L2 (1.285V ~ 1.388Vcore) Socket-775
Mainboards	Asus P5WD2-Premium (BIOS 0606)
Memory	Crucial Tracer Balliztix 5300 (2x512MB DC CL4-4-4-12) OCZ GOLD PC2-6400 (2x512MB)
Graphics	Gigabyte GV-NX68T256DH
Power Supply	PCPower&Cooling TurboCool 850 SSI
Cooling	Alphacool 12V, Black Ice Extreme Double radiator 2x Papst 4412 FG2/GL 120mm
Operating System	Windows XP

OCZ GOLD PC2-6400 is as we stated earlier designed for Gamers. Among OCZ's now extensive line of high end memory their Dual Channel GOLD may come as close to a budget part as you'll find without actually wearing the "budget" title. Because of this as I began testing the memory I didn't expect to break any overclocking records and in this respect our OCZ GOLD didn’t disappoint. Unfortunately overclocking beyond the speed seen on the thumbnails below wasn't possible. Any attempt to increase voltage albeit to increase speed or tighten latencies resulted in system failure and/or instability. This is not an uncommon occurrence with DDR2, since its release I've found many circumstances in which the memory reacted poorly to an increase in voltage. Tightening timings with or without voltage was unsuccessful there fore all tests were run under SPD (Serial Presence Detect).