Intel recently announced that one of its next-generation processors will default to 6GHz, sparking a new arms race in CPU design. What challenges have CPUs faced over the past decade, what has Intel announced and how could this fuel a new race?
What challenges have CPUs faced over the past decade?
Since the first microprocessors The computing power of the devices has grown exponentially from year to year, and devices that used to take up an entire floor of a building now fit comfortably on a tiny silicon chip the size of a grain of rice. Despite numerous advances in mobile and network technology, computers have stagnated over the last decade and operating systems have taken as long to boot up as they did 10 years ago (if not longer in some cases). In addition, long-term stability still seems to be problematic (a Windows 10 system will show a significant drop in performance after a year or two of operation).
Now there is no doubt that CPU performance has increased over time, but the performance of CPUs can be measured in numerous ways. For example, efficiency performance can indicate how energy efficient a processor is, while multitasking performance shows how well a CPU can perform multiple tasks simultaneously. In the case of modern computers, the performance of single-threaded tasks hasn’t increased much in the last decade, and it’s this lack of improvement that causes operating systems to be slow to boot and responsive.
Basically, Operating systems and user applications rely on a single process using sequential instructions, and it is not possible to execute future statements before the previous statements have been executed. Therefore, it doesn’t matter how many processors a CPU has or how much RAM is available. A single task runs just as fast on a 10-core CPU as it does on a single-core CPU.
It is possible to split an operating system’s tasks into different processes, each of which can run on its own specific core (thus allowing parallel execution), which can contribute significantly to improved performance. However, many operating system services rely on the kernel for operation, making it difficult to separate these additional services from the main operating system thread.
For the past decade, increasing the core frequency has been the simplest solution to improving single-threaded performance, but considering that exceeding 5GHz results in massive heat dissipation (unavailable to the average consumer), we had to Processors rely on pipelining and superscalar designs to try and improve performance.
Intel announces native 6 GHz processor
Lately, Intel has announced that it will release a new CPU, the Raptor Lake, which provides users with a CPU that can run at 6 GHz natively. Although pricing and availability have not been released, it is expected that the new processor could be a version of the upcoming KS-branded 13900K chip. At the same time, Intel’s announcement hits hard at AMD, which recently announced that its next-gen CPU will run at 5.7GHz.
While a 300MHz difference may seem marginal, it can offer massive improvements in single-threaded performance. In addition, Intel announced that the 6 GHz speed will be available on all cores at the same time and not just for individual cores running in a performance mode. According to Intel, the new processor speeds will see a 15% increase in single-threaded performance, while multi-threaded performance will increase by 41%.
However, Intel went on to announce that the new architecture will support overclocking to 8 GHz, which is twice the speed of typical computer cores. Considering that such speeds have historically only been possible with liquid nitrogen and other exotic cooling methods, this announcement suggests that a new CPU race may be on the horizon.
The second CPU race?
for the longest time Computer manufacturers used MHz as the defining characteristic of their machines. A computer operating at 8MHz would be considered inferior to those operating at 10MHz, which existed well into the early 2000s. However, it became clear that frequency had little impact on multi-core designs, so the battle shifted from GHz to core count, FLOPS, and memory bandwidth.
However, the announcement by AMD and Intel of their next generation of CPUs could ignite a new race for faster single-threaded machines. While a 15% increase may not seem large, it is undoubtedly a step in the right direction to improve application performance through improved responsiveness and faster launch times. Aside from that, The increased performance in multi-core designs also benefits multi-threaded applications since each thread sees accelerated execution.
But during faster CPUs can help improve overall computing power, it is certainly not the best option for numerous modern applications, especially in the mobile space. By far the most important feature in low-end computing tasks and those that depend on portable power supplies is power efficiency; A mobile device that runs out of power in 20 minutes due to an 8GHz core CPU is not a practical device. With the vast majority of tech users now moving to tablets and phones with cloud-based applications, high-performance CPUs may be more of a niche product.
Overall, the increase in CPU speeds from Intel and AMD is certainly welcome, as you can only get so far with many cores, but while it might be great for tech-savvy people and supercomputers, it might not be the best option for large data centers or mobile users.