Shows 'No Signal' during RAM overclocking attempts
Shows 'No Signal' during RAM overclocking attempts
Your setup includes an R5 1600 processor, 3060Ti RAM, a 650W power supply, and two F4-3200C16S RAM sticks. When you tried to overclock the RAM from 2133MHz to 3200MHz using A-XMP, the system would randomly crash or show blank screens with messages like 'No signal' or BSOD. After running MemTest86+, no errors were detected in the RAM. You experimented with adjusting the voltage at the A-XMP profile—first increasing it to 1.38V and running a Prime95 stress test (covering 448k to 4096k, 70% of RAM size, without changes). The system crashed within three minutes. Then you raised the voltage to 1.42V, which allowed Prime95 to run smoothly for nearly two hours before closing. After these changes, crashes still occurred, and the CPU LED on the motherboard would light up frequently. This behavior suggests a possible incompatibility or instability issue between your RAM and CPU. You might want to verify the RAM specifications against your CPU model, check for firmware updates, or consider testing with different RAM modules. If problems persist, consulting a professional repair service could be helpful.
I don't recall this being a confirmed fact, though I've heard some people mention early Ryzen models struggled with memory performance. Perhaps consider options like the 2933 or similar.
This indicates your character is still quite problematic. A 1600 rating is an impressive feat, especially when it comes to boosting memory performance. If you're still unable to start up, try resetting your BIOS settings—this applies regardless of whether it's Intel or AMD hardware.
I tested all frequencies from "Memory Try it!", except 2133 mhz, and encountered issues. I updated the system, raised the voltage to 1.44V, and compressed the 25GB file into 110GB without errors. I’m hoping adjusting the voltage resolves the problem. I’ll run some benchmark tests with Prime95 to confirm. Is 1.44V safe? It may reduce lifespan and increase power use, but it shouldn’t damage the RAM.
I recall changing BIOS settings multiple times over the past few days. It's likely I've reset it several times recently. The goal is to fine-tune the system performance, as my RAM modules are designed for 3200Mhz and I don't want to run them at 2133Mhz. Ryzen's Infinity Fabric works best with higher RAM speeds, which is why I'm hesitant about switching to 5600x unless necessary. Right now, the memory sticks appear fine, but there seems to be an issue with R51600 compatibility. I wouldn't want to spend over $300 just for a modest gain in gaming performance at 1440p ultra settings. I'm trying to stay within budget while still getting decent results. For updates, I increased the memory voltage to 1.44v and noticed some improvement. Check my previous post for more details. Thanks for your time!
Memory controller problem detected. My 2600 runs at around 1600 refresh and stays stable up to about 2800MHz without overheating or frequent crashes. I’d aim for 2933 and fine-tune timing to reduce CAS latency. The Ryzen DRAM Calculator app can help pinpoint the right settings. Also, your voltage is near its daily operating limit.
Sure, I'm thinking about using 3600 or 5600x to fully utilize the available memory.
I bought the first AM4 CPU as the 3600XT from a 3770K chipset. It turned out to be an excellent processor. The one I actually received was the 5600X, which made me very pleased. It’s such a solid performance unit. Now that I have a new GPU, I just need to get a motherboard.
Identified the issue. The memory stick purchased three days ago on Amazon is defective. My previous RAM performed well on Hitman benchmarks (Miami/Mumbai), tested on OCCT and memtest64 for 15 minutes each with no errors or crashes. This new RAM failed during Hitman testing, crashed on OCCT within two minutes, and showed memory locking failure in the first memtest64 run. On a second attempt it crashed without any error display. The heatsink is also loose. I previously assumed this RAM was fine because it passed memtest86+, which ran for two hours. It remains under the 7-day replacement or return warranty, but I’m replacing it now.