- Conceptual Overview
- Read-Only Memory (ROM)
- Random Access Memory (RAM)
- Cycles and Frequencies
- Summary—Basic Memory
- Cache Memory
- Memory Pages
- Rambus Memory (RDRAM)
- Double Data Rate SDRAM (DDR SDRAM)
- Video RAM (VRAM)
- Supplemental Information
- Packaging Modules
- Memory Diagnostics—Parity
- Exam Prep Questions
- Need to Know More?
Exam Prep Questions
Part of a computer's RAM chip is dedicated to storing key system settings required for boot-up.
Answer B, false, is correct. Random access memory (RAM) is volatile and loses all its data without a source of power. RAM comes in modules and is almost never referred to as a chip. System boards commonly use nonvolatile CMOS to store system settings. CMOS memory uses very little current (a trickle charge) and continues to be powered for extended periods of inactivity by a small battery on the system board.
Over-clocking allows a microprocessor to run considerably faster than motherboard components. What type of memory structure was developed to minimize the delay of accessing RAM on the motherboard?
Processor resident pipeline
Answer B is correct. Intel 486 and Pentium processors have a small amount of memory integrated in the chip called an L-1 cache. However, as processor speeds increased, additional high-speed memory was needed. This second block of memory was called an L-2 cache and was located on a special high-speed bus. Later designs include the L-2 cache on the processor cartridge or the die itself. Duplex memory and a processor resident pipeline do not exist. Answers A and D are incorrect because they're imaginary terms. Answer C is incorrect because CMOS is used to store settings information and has no effect on processor speed.
DIMMs and SIMMs are interchangeable, provided speed and capacity requirements are observed.
Answer B, false, is correct. SIMMs and DIMMs look similar, and both use edge connectors. However, DIMMs use both sides of the connector to support a 64-bit or wider memory bus, and they have two separate connector pins, one on each side of the module board. SIMMs also have two separate connector pins, but they are wired together.
Parity chips on SIMMs no longer provide a useful purpose and have been largely removed.
Answer B, false, is correct. Parity chips allow memory to be tested during the POST, and they also monitor memory during computer operations. Some manufacturers have eliminated them or bypassed their function to cut costs. This allows less expensive SIMMs to be used, but at the expense of reliability.
Which of the following choices best describes what is meant by cache memory?
A place where instructions are stored about the operations of a device or application
Extended memory that can be made accessible with the SMARTDRV /ON command
Memory that holds applications and data that the CPU isn't running
Memory that holds data that the CPU will search first
Answer D is correct. The CPU looks in cache memory first. If it fails to find the necessary data, it looks in main memory. If it fails to find what it needs in main memory, the CPU looks on the disk. Answer A is incorrect because hardware instructions are handled at a bus level, not in system memory. Answer B is incorrect because cache memory is a small amount of specialized memory. Extended memory is part of main memory. Answer C is incorrect because a cache is used to speed up memory usage. If the CPU isn't running an application it has no need for speed.
Which of the following types of memory operate on a principle similar to serial transfers running through a cable? (Choose all that apply)
Answer C is correct. This is a trick question, not in the types of memory but in the misleading suggestion that there is more than one answer. Rambus memory (RDRAM) is different from other forms of memory in that not only are the chips on the RIMMs running in series, but the memory slots on the board also are in series, requiring a continuity module. Answer D is incorrect because DDR memory is particularly designed around a parallel transfer structure. Answers A and B are incorrect because SRAM and SDRAM move bytes across the memory bus in whatever width the system can support.