Litigation is a machine which you go into as a pig and come out as a sausage - Ambrose Bierce, allegedly
SEMICONDUCTOR OVERLORDS Intel and AMD didn't wait around and kicked off 2013 with the announcement of some major new chip architecture at CES in Las Vegas just days into the new year.
Back in 2011, Intel showed up at CES with its Sandy Bridge chip architecture and blew AMD out of the water. However, two years on, it didn't quite deliver the same impact as Sandy Bridge, but it did bring its Ivy Bridge processor architecture below the 10W TDP threshold, a move that would worry ARM vendors.
AMD, on the other hand, announced next-generation accelerated processing units (APUs), claiming it had signed big brands to use its chips. This was something it desperately needed to do to win the shelf space battle against Intel.
AMD's CES announcements were slightly out of the ordinary because the firm didn't just reveal products that it was shipping into the channel, but next-generation architectures such as Kaven.
The firm also announced that Hondo, which it launched back in October 2012, would be surpassed within six months by Temash, a quad-core x86 APU that would ship in the first half of 2013 and be pitched at tablets and laptops that have swivel screens.
While Intel and AMD battled out their ongoing war, Nvidia took the stage to announce its latest Tegra 4 system on a chip (SoC), a quad-core chip with a significant graphics boost. The firm did its best to play down the fact that its Tegra 4 has the same CPU core count as its previous-generation Tegra 3, and instead it focused on GPU performance, an area where the Tegra 3 was starting to look dated against newer chips from rivals such as Samsung.
After CES, semiconductor news refused to fall quiet. Qualcomm, for instance, kicked off February by officially announcing its high-end flagship quad-core Snapdragon 600 and Snapdragon 800 chips, which it had showed off the month earlier at CES. The firm also took that time to unveil its quad-core Snapdragon 200 and Snapdragon 400 processors for entry-level and mid-range smartphones.
The fact that Qualcomm opted for quad-core ARM Cortex A5 CPU cores even on its low-end Snapdragon 200 hinted that four processing cores would become the baseline for almost every smartphone in 2013.
Curiously, Qualcomm offered dual-core and quad-core Snapdragon 400 chips, with the dual-core option being based on its Krait chip from 2012, clocked at 1.7GHz. The firm's quad-core variant was based on four ARM Cortex A7 cores running at 1.4GHz, but unlike the Snapdragon 200, the firm boosted graphics performance by offering the Adreno 305 GPU.
It had taken Intel until 2012 to finally dip its toes into the Android smartphone market, though a mix of largely forgettable devices saw the firm make very little headway. However, later in February Intel went out of its way to piss off Qualcomm, updating its mobile chip line-up and announcing three Atom dual-core Clover Trail+ processors for Android smartphones, citing significant improvement in graphics performance. The dual-core Clover Trail+ processors sported clock speeds of up to 2GHz and Hyper Threading, allowing four threads to run simultaneously.
February's chip announcements ended with Texas Instruments unveiling two software packages for its Keystone SoC that help the chip power small-cell base stations.
The firm said its software PHY package gave base station designers a quick and easy way to create small cells that support 4G LTE and WCDMA protocols. The firm's transport package was intended for wireless and broadband network applications such as routing and switching.
The next big announcement came in April and was delivered by British chip designer ARM, which announced the availability of POP IP for its upcoming ARMv8-based Cortex A53 and Cortex A58 chips using TSMC's 28nm process node.
ARM's POP IP products were designed to help smaller ARM vendors design and manufacture chips based on its various architectures using specific process nodes that ARM and the foundry, in this case TSMC, had worked on developing.
ARM stressed the ability for its Cortex A53 and Cortex A58 chips to be used independently or in its Big Little architecture. The firm said its POP IP products would enable dual-core and quad-core implementations of both chips.
A month later, ARM then announced a dual-core hard macro for its Cortex A15 chip that used TSMC's 28nm high-performance mobile (HPM) process node.
ARM's Cortex A15 chip is perhaps best known as the processor in Google's first Nexus 10 tablet and is based on the ARMv7 instruction set. ARM's low-power version of its Cortex A15 chip aimed to help smaller licensees speed up the end-to-end process of development to final products. In this case, the firm offered an implementation that integrates its Neon single instruction, multiple data (SIMD) engines and 1MB of Level 2 cache and is tested on TSMC's 28nm HPM process.
The month of May ended with Nvidia announcing an all-Kepler Geforce GTX 700M series in preparation for Intel's Haswell chips.
Nvidia's previous generation Geforce 600M series was a mish-mash of parts, with some being based on the firm's well-received Kepler architecture, while others were still using Fermi. However, the firm has announced that all parts in its Geforce GTX 700M series would be based on its Kepler GPU architecture.
The next big event for chip news arrived in the form of Computex, a computer trade show that took place in June in Taiwan, giving chip makers a chance to flex their semiconductor muscles.
Sign up for INQbot – a weekly roundup of the best from the INQ