Mixing 2 RAM Sticks of Different Sizes/Speeds Results in Slower 'Configured Speed', Even Slower Performance - Benchmarks

In today's computing landscape, having sufficient Random Access Memory (RAM) is essential for optimal system performance. While most modern computers come with pre-installed RAM, there might be instances where users need to add or upgrade their existing RAM. In this article, we will explore the implications of mixing two RAM sticks of different sizes and speeds and the resulting impact on system performance.

Background

RAM plays a crucial role in storing data that the CPU uses frequently. By having enough RAM, the system can efficiently access and process data, leading to improved performance. However, using mismatched RAM sticks can result in suboptimal performance.

Mismatched RAM Sticks: A Comparison

Let's consider a scenario where we have a benchmarking machine with a 4-core Intel i5-10210U CPU, and we plan to add an additional 16GB RAM stick (3200MHz) to the existing 8GB stick (2667MHz).

Configuration and Setup

Before proceeding with the benchmarking process, let's ensure that the system is properly configured. In this case, we will be using the following settings:

• Windows 10 Pro
• All system updates installed
• All default settings

Benchmarking

To measure the performance impact of using mismatched RAM sticks, we will be using a popular benchmarking tool, CrystalDiskMark. This tool allows us to test the read and write speeds of the system.

Results: Original Configuration

First, we will benchmark the system using the original 8GB/2667MHz RAM stick:

Writing: Sequential Write: 985 MB/s 512KB: 235.6 MB/s 4KB: 22.8 MB/s 4KB (Queued): 22.8 MB/s

Results: Adding a Faster Stick

Next, we will add the 16GB/3200MHz stick and configure the system to use both sticks:

Writing: Sequential Write: 1017 MB/s 512KB: 242.6 MB/s 4KB: 23.5 MB/s 4KB (Queued): 23.5 MB/s

Results: Adding a Slower Stick

Finally, we will add the 16GB/3200MHz stick but configure the system to use the slower 8GB/2667MHz stick as the primary:

Writing: Sequential Write: 985 MB/s 512KB: 235.6 MB/s 4KB: 22.8 MB/s 4KB (Queued): 22.8 MB/s

Analysis

From the benchmarking results, we can observe that using the faster stick (16GB/3200MHz) in conjunction with the existing stick (8GB/2667MHz) leads to improved read and write speeds. However, when the system is configured to use the slower stick (8GB/2667MHz) as the primary, the performance drops to the original levels.

In conclusion, mixing two RAM sticks of different sizes and speeds can lead to suboptimal system performance. While using a faster stick can improve overall system performance, configuring the system to use the slower stick as the primary can result in a significant drop in performance. Therefore, it is recommended to avoid mixing different RAM sticks whenever possible.

References

• Intel Memory Technology Platform Policy
• CrystalDiskMark