The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Benchmark tests conducted in 1998 showed most applications run slower with RDRAM. Although RDRAM was shown to be slightly faster than SDRAM alternatives in UMA solution, Intel 820 was not a low-end product, and no low-end products using RIMM has ever been made, so the advantage is useless for end users.[3]
In 1999, benchmark for Intel 840, Intel 820, Intel 440BX showed the performance gain (if any) from using Rambus chipsets do not justify its premium price over 440BX chipsets with PC-133 SDRAM except for workstation use.[4]
Later in 2002, it was shown that single channel DDR400 SDRAM modules, coupled with SiS648, can closely match against dual channel 1066MHz RDRAM setup with Intel 850E in real-life applications[5]. Furthermore, there were upcoming chipsets that can use dual channel DDR400 SDRAM modules.
...and this is probably why most PCs have DIMM slots and use DDR2
RDRAM has HORRIBLE latency issues, even though it has a high clock speed. It was the same as the Pentium 4 type processors... great for high-speed, sequential throughput, and that's about it. The DDR architecture as a whole (all iterations) is much better at random access of locations in memory, and works better for what 99% of people use a computer for. The only reason you want a P4/RDRAM is if you ONLY do video encoding or something similar that streams a ton of data. Even 3D rendering is bad on those types of systems.
Problems? See this thread at archive.org.
