Compare the Difference Between Similar Terms

Difference Between Apple A5X and Nvidia Tegra 3 Processors

Apple A5X vs Nvidia Tegra 3 Processors
 

This article compares two recent System-on-Chips (SoC), Apple A5X and NVIDIA Tegra 3, designed for consumer electronics by Apple and NVIDIA respectively. In a Layperson’s term, a SoC is a computer on a single IC (Integrated Circuit, aka chip). Technically, a SoC is an IC that integrates typical components on a computer (such as microprocessor, memory, input/output) and other systems that cater electronic and radio functionalities. Both Apple A5X and NVIDIA Tegra3 are Multiprocessor System-on-Chip (MPSoC), where the design uses multiprocessor architecture for exploiting the computing power available. While NVIDIA released Tegra 3 in November 2011, Apple will release A5X with its iPad 3 this week (March 2012).

Typically, the major components of a SoC are its CPU (Central Processing Unit) and GPU (Graphics Processing Unit). The CPUs in both Apple A5X and Tegra 3 are based on ARM’s (Advanced RICS – Reduced Instruction Set Computer – Machine, developed by ARM Holdings) v7 ISA (Instruction Set Architecture, the one that is used as the starting place of designing a processor).

NVIDIA Tegra 3 (Series)

NVIDIA, originally a GPU (Graphics Processing Unit) manufacturing company [claimed to have invented GPUs in the late nineties] have recently moved into the mobile computing market, where NVIDIA’s System on Chips (SoC) are deployed in phones, tablets and other handheld devices. Tegra is a SoC series developed by NVIDIA targeting deployment in the mobile market. The first MPSoC in Tegra 3 series was released in early November 2011 and was first deployed in ASUS Transformer Prime.

NVIDIA claims that Tegra 3 is the first mobile super processor, for the first time putting together quad core ARM Cotex-A9 architecture. Although Tegra3 has four (and therefore quad) ARM Cotex-A9 cores as its main CPU, it has an auxiliary ARM Cotex-A9 core (named the companion core) which is identical in architecture to the others, but is etched on a low power fabric and is clocked at a very low frequency. While the main cores can be clocked at 1.3GHz (when all four cores are active) to 1.4GHz (when only one of the four cores is active), the auxiliary core is clocked at 500MHz. The target of the auxiliary core is to run background processes when the device is in standby mode and therefore saving power. The GPU used in Tegra3 is NVIDIA’s GeForce which has 12 cores packed into it. Tegra 3 allows packing of up to 2GB DDR2 RAM.

Apple A5X

The new iPad (aka iPad 3 or iPad HD), the first consumer electronic device that will be equipped with A5X MPSoC will be released in the mid of March 2012 (during the course of this week). During the new iPad launch event on 7th March 2012, Apple revealed that they will be using Apple A5X processor to drive the device. Apple A5X has a dual core CPU like A5 and therefore not going to perform very different compared to its previous A5 MPSoC. It is worth to note that, this is opposed to the previous belief that Apple will use a quad core processor, the trend of 2012 MPSoCs (such as Tegra 3), for its new iPad. Based on the information leaked up to now, Apple will clock its A5X CPUs at 1.2 GHz as opposed to 1GHz in its predecessor A5. Apple claims that their A5X will have 4x better performance in graphics compared to devices equipped with NVIDIA Tegra3.

Although A5X has a dual core CPU, the GPU used (that is responsible for the graphics performance) is a quad core PowerVR SGX543MP4. Therefore, the graphics performance of A5X is going to be theoretically doubled compared to Apple’s A5 processor. In fact, the “X” in A5X stands for graphics. Therefore, A5X is a high end graphics processor that is expected to support the new iPad HD graphics (the retina display that Apple is introducing in the new iPad, the first in tablet PCs). It is worth noting that for some benchmark applications Apple A5 performed 2x better in graphics compared to Tegra3 and therefore Apple’s claim of 4x better graphics performance compared to Tegra3 is theoretically possible. A5X is expected to be shipped with a 32KB L1 private cache memory per core (for data and instruction separately) and a 1MB shared L2 cache. It will also be expected to be packaged with a 512MB memory.

A comparison between Apple A5X and NVIDIA Tegra3 is tabulated below.

 

Apple A5X

Tegra 3 Series

Release Date

March 2012

November 2011

Type

MPSoC

MPSoC

First Device

The new iPad (iPad 3 or iPad HD)

ASUS Transformer Prime

ISA

ARM v7 (32 bits)

ARM v7 (32bit)

CPU

ARM Cortex-A9 (dual core)

ARM Cortex-A9 (Quad Core)

CPU’s Clock Speed

1.2GHz

Single Core – up to 1.4 GHz

Four Cores – up to 1.3 GHz

Companion Core – 500 MHz

GPU

PowerVR SGX543MP4 (quad core)

NVIDIA GeForce (12 cores)

GPU’s Clock Speed

Not Available

Not Available

CPU/GPU Technology

TSMC’s 45nm

TSMC’s 40nm

L1 Cache

32kB instruction, 32kB data

(per CPU core)

32kB instruction, 32kB data

(per CPU core)

L2 Cache

1MB

(shared among all CPU cores)

1MB

 (shared among all CPU cores)

Memory

512MB DDR2, 533MHz

Up to 2GB DDR2

 

Summary

In summary, Apple A5X has higher potential and given that it is going to be used by one of the best technology integrator will make A5X utilized better. As the “X” in name A5X suggests, A5X will play a serious role in bringing high definition video and graphics to mobile devices such as tablet PCs. In fact, it is a need for Apple to have the best performing graphics processor to drive their retina display with the highest resolution available for tablet PCs. On the other hand, how good the dual core CPU will cope with the computation demand while Tegra 3 is out there with a quad core CPU is to be seen after the launch in the near future (when some benchmark tests can be run).