Comparing ECC and non-ECC gaming performance highlights differences in stability and responsiveness.
Comparing ECC and non-ECC gaming performance highlights differences in stability and responsiveness.
I found two sets of 8GB 2666MT CL22 Kingston DDR4 ECC RAMs, so I checked how much slower my current build would be compared to the one I had when I first started. When setting up my system, I chose not to use ECC for speed reasons. After testing, here’s what happened: - Processor: AMD Ryzen 5800X3D - Graphics: Nvidia/Gigabyte RTX2070 Super - Motherboard: Gigabyte Aorus X570 Pro I wanted to share this if anyone is considering a purchase. My decision is settled—since the CPU isn’t sensitive to RAM speed, ECC might not be worth it here. I mainly use my machine for games, streaming, office tasks, and occasional 3D work. Feel free to let me know if you run any other tests!
Boost the ECC RAM first, then check the ICS (black squares) for confirmation. If they look different, the burner is likely the right choice.
I’d opt for the 32GB with 3600MT/s RAM. ECC isn’t necessary for a setup like yours; it’s more suited for servers, NAS units, or high-performance workstations where reliability matters. For gaming or productivity tasks, you’ll benefit more from the speed and capacity of your 32GB configuration.
I see the testing done so far primarily as a comparison of 2666 vs 3600. Because it is consumer Ryzen, there is no official support for ECC, so you don't know it is working. There is a big possibility it is running as normal ram. Also the CPU is the X3D, the large cache of which will reduce the influence of the ram anyway. IMO if you really want ECC, you have to use a platform that fully and officially supports it end to end. Consumer AM4 does not qualify. And finally, I'd also question the usefulness of 3DMark's benches as a test. Different games behave differently, so testing ideally would be expanded to a variety of actual games.
Despite your sincere effort, the results don't reflect meaningful improvements. This performance remains limited by its large L3 cache and 3D architecture, even though the graphics benchmarks look impressive. For a clearer view of DRAM speed variations, consider using Linpack or SuperPi instead.
Use 7-Zip to bundle more than 10GB of random data and then test the speed at which it decompresses. Ideally, work with the folder on an NVMe storage device or set up a RAM drive so that access remains fast during testing.
There was debate about whether enabling ECC reporting guarantees its operation. You’d likely have to perform error injection and verify fixes.