AMD is working with TSMC to optimize process technology, but expects overall temperatures to continue to rise in future generations of Ryzen processors as chip integration increases. AMD Ryzen 7000 CPU is currently the most efficient chip in client PCs. Due to the increase in transistor density and various architectural changes, the Zen4 core architecture they use has greatly improved both single-threaded and multi-threaded performance. However, the increase in transistor density within smaller sizes has also led to a sharp rise in CPU temperatures.
The latest Ryzen 7000 CPU, codenamed "Raphael," runs extremely hot. When running any CPU-intensive tasks, they will often reach a peak Tjmax of 95C, which requires very powerful cooling when overclocked. While AMD expects this trend to continue in future generations as chip density continues to increase, these CPUs can also maintain nearly similar performance at lower voltages while running cooler.
AMD Vice President David Mcafee said in an interview with QuasarZone that they are working hard to cooperate with TSMC to optimize the latest process technology to ensure the quality and stability of the chip. However, because the number of transistors in modern CPU architectures doubles or even more with each generation, and the chip size becomes smaller and smaller, the heat generation is either the same as now or continues to increase.
Q. One of the criticisms of AMD's desktop products is CPU temperature. The CPU power consumption is significantly lower than that of the competition, but the temperature is higher. Will these temperature issues be resolved in the future? Is it possible to install a fake chip next to the CCD chip to dissipate heat?
A. We are working closely with TSMC and investing a lot of energy in process technology. At the same time, we must be able to guarantee the quality and stability of semiconductors. With the adoption of more advanced processes in the future, we believe that the current high heat density phenomenon will be maintained or further intensified. Therefore, a way must be found in the future to effectively eliminate the high heat density generated by such high-density chips.
In addition, if you look at the TDP65W product, its overall performance is also very good. With these product examples, I think this is an important factor to consider when planning future roadmaps to ensure a good balance between TDP and heat generation.
There's also a heat density map below that shows the heat generated through a smaller chip surface area:
Intel previously believed that CPU heat generation would continue to rise. The company's latest 14th-generation CPU is one of the hottest chips currently available, with a maximum temperature of over 100C. This has led motherboard manufacturers to expand the thermal threshold beyond 115C, allowing chip operating temperatures to reach up to 121C.
But Intel engineers say this is to be expected and that modern chips are designed to maintain high temperatures while delivering optimal performance. AMD's CPUs show that despite reaching the thermal threshold, the CPU doesn't generate that much heat when it starts to downclock. In a way, it's like saying that if you want your chip to get the most out of it, then you have to hit the limits of the chip in terms of heat and power. The following interview with Intel engineers (via Der8auer) explains their perspective:
From this point of view, rising temperatures will become the development trend of next-generation CPUs from Intel and AMD. We hope the performance increase is enough to make up for the higher heat output. What is certain is that cooling equipment manufacturers will take a more innovative route with new designs of air and liquid cooling systems.