Time to switch back to AMD
from 07/16/2024, by uni — 4m read
Let me preface by saying that I've switched from Intel to AMD before and am a non-partisan consumer. This blog post is being written on an Intel 12700K, and luckily not the 13th and 14th generation chips that are currently being plagued by issues.
It doesn't take a semiconductor expert to understand that constantly pushing 1.5 volts or 500 watts directly to the CPU isn't a good thing. It's become evident that intel has hit a developmental brick wall years ago. They have their eyes set for the top spot on the charts but can only accomplish it by brute force. Unfortunately for the consumers, this leads to the detriment of everything else.
Several independant and reputable sources have released information about Intel's Raptor Lake disaster. Even NVIDIA themselves have blamed Intel for GPU VRAM errors. Most suspect that chip degradation is caused by Intel's out of control power settings. The trend of increasing power limits each generation would eventually be an infeasible plan right? Some 14th generation users have said that their once stable underclocked chips would require more voltage every week until the chip would essentially need to be overclocked to even boot. Others reported that lowering the chips frequency mitigated the problem for a while, but would eventually crash at a later point.
The server-based motherboards used by these 13900K and 14900K servers are focused entirely on stability and running the chips within specifications, with no way to overclock these chips. - Tom's Hardware
What's weird is that these issues also seem to plague enterprise hardware as well. This leads me to suspect that the ring bus is being deteriorated as it runs at the same voltage as the cores. That would explain why non-overclocked chips are suffering the same issues as voltage and wattage heavy i9 chips. At this point Intel has mastered squeezing the most out of their process nodes as AMD continues to shrink their own.
Over the last 3–4 months, we have observed that CPUs initially working well deteriorate over time, eventually failing. The failure rate we have observed from our own testing is nearly 100%, indicating it's only a matter of time before affected CPUs fail. - Alderon Games
Even more confusing to me is how Intel could have missed this issue througout their own testing. It doesn't matter if this issue is caused by voltage, wattage, or any other power metric because surely their engineers would have discovered issues... right? In my opinion there exists no situation where Intel didn't come across these issues at any step of their testing process. Could they have knowingly sent out defective chips? It may be possible, but more realistically their engineering samples worked just as intended and had no degradation issues. All of this points to Intel's own fabs potentially having manufacturing defects.
The truth is that to the average consumer, users that seldom enter the BIOS and don't know anything about power settings, will lose their trust in Intel after encountering these issues. For me personally, currently on the Alder Lake platform and have not encountered these issues, I have also lost trust in Intel. Seeing the current line-up of Ryzen 3D V-Cache chips honestly makes me jealous, and makes me even more excited for the upcoming Ryzen 9000 series chips. I plan on switching to the Ryzen 9800X3D chip after seeing independant reviews. AMD's PBO seems like so much less of a headache than Intel's BIOS power options by a long shot.
My current system is a SFFPC containing a 12700K and 240mm AIO. I've seen other FormD T1 systems with the Ryzen 7800X3D running on air that perform faster and cooler than my i7. Whether Intel is finally facing the consequences of pushing their chips to the limit, or this is all due to their fabs having a manufacturing defect; there isn't really any reason to buy an Intel CPU in 2024.