NewsHome > News >
Ryzen, AMD an 8-core, 16-thread chip

AMD has been making processors for a long time and Zen is the name of the new 'core architecture' around which a whole family of products will be based. One of these is the newly announced Ryzen processor. This is not a single CPU, but rather a range (just like Athlon back in the day). Ryzen CPUs will be available for desktop PCs, laptops and even servers.

The x86 Zen architecture is built on a 14nm FinFET manufacturing process. Essentially, this means that Ryzen processors are able to do a lot more work per clock cycle - 40 percent, in fact. This is the key, or one of the keys, to competing with Intel processors. Previously AMD could only compete on performance at a much higher power consumption, because it needed to use a higher clock speed to do the same amount of work as the equivalent Intel Core CPU.

With Ryzen, AMD is claiming that an 8-core, 16-thread chip is 10 percent faster than an Intel Core i7-6900K in various benchmarks, such as Blender and Handbrake. Importantly, these tests were run with the Intel chip using its Turbo Boost speeds, while the Ryzen chip had its boost disabled. So there's more performance on tap, and that is exciting and not just for AMD fans.

Intel is about to launch the next-generation chip, the seventh-generation Core processors (codenamed Kaby Lake). This shouldn't worry AMD, though, since early indications are that the Core i7-7700K is no more efficient than the 6900K in terms of performance per clock cycle: it is simply more power efficient. An incremental improvement, at best.

Ryzen, meanwhile, is 40 percent more efficient than the Excavator chips it replaces. To be specific, it is able to process 40 percent more instructions per clock cycle - this is the '40% More IPC' in the slide below.

One of the way it does this is by using a smaller manufacturing process: 14nm. This is nothing new - Intel has been using this process for a while now. On top of this change is what AMD is calling SenseMI. 

SenseMI has five components:

·         Pure Power – more than 100 embedded sensors with accuracy to the millivolt, milliwatt, and single degree level of temperature enable optimal voltage, clock frequency, and operating mode with minimal energy consumption;

·         Precision Boost – smart logic that monitors integrated sensors and optimizes clock speeds, in increments as small as 25MHz, at up to a thousand times a second;

·         Extended Frequency Range (XFR) – when the system senses added cooling capability, XFR raises the Precision Boost frequency to enhance performance;

·         Neural Net Prediction – an artificial intelligence neural network that learns to predict what future pathway an application will take based on past runs;

·         Smart Prefetch – sophisticated learning algorithms that track software behaviour to anticipate the needs of an application and prepare the data in advance.

 

Arguably the most interesting part of the Zen architecture is its ability to support 'Simultaneous Multi-Threading', a technology that's been used in Intel's CPUs for years under the name Hyper-Threading. This allows a single core to have multiple threads, such as an Intel Core i7 having four cores and eight threads.
Solutions | Products | Download | Careers | Contact Us | Copyright (C)2001-2018 www.PesComm.com