Jon Peddie Blogs

Qualcomm’s powerful new HPU the S4

Posted by Jon Peddie on October 7th 2011 | Permalink
Categories Blogs, Engineering and Development
Tags: gpu qualcomm krait simd hexagon dsp adreno snapdragon s4 processor

Initially leveraging a 28nm process from TSMC, Qualcomm has announced its Snapdragon S4 class of processors, of which the first member is the MSM8960 with an Adreno 225 GPU. The new 1.5 GHz processor (S4 will scale up to 2.5GHz) has Qualcomm’s micro-architectural design with four independent, proprietary ARM Cortex A15-class CPU cores, plus a 32-core GPU, plus 128 bit SIMD engine, plus three DSPs, plus a handful of hardwired engines for codecs and other special-purpose functions—basically a five processor+ heterogeneous processor that has an open programming environment, and a fast memory interface and manager. The new CPU core is compatible…

AMD, ARM, and all that jazz — reading between the lines will make you cross-eyed

Posted by Jon Peddie on August 3rd 2011 | Permalink
Categories Blogs, Engineering and Development
Tags: amd, gpu, fusion, x86, fsail, arm,

There have been persistent rumors, opinions, and speculation since AMD’s Fusion Developer’s Summit (AFDS) that because ARM was one of the keynote speakers a grand collusion was in the works—the ARMing of AMD.The concept gets fuel from the abrupt discharge of Dirk Meyer, the company’s former president for allegedly not having a mobile strategy.AMD’s announcement of a Fusion System Architecture—Intermediate Layer, into an open platform was one of the main messages AMD wanted to get out at AFDS. They are calling it FSAIL. AMD has said they will publish the Fusion System Architecture (FSA) virtual Instruction Set Architecture (ISA) (FSAIL), FSA…

Getting vertical – what Nvidia’s acquisition of Icera means

Posted by Jon Peddie on May 9th 2011 | Permalink
Categories Blogs, Engineering and Development
Tags: nvidia, gpu, icrea, 3g modem, 4g modem, tsmc, tegra

Nvidia announced the acquisition of UK-based maker of baseband chips for 3G and 4G handsets, Icrea for $367 million in cash. This is a really big deal – not for the purchase price but for the impact it’s going to have on the industry. Now the playing field in mobile devices is shaping up to a big three (or four) players, a normal consolidation in a broadly fragmented market. With Nvidia’s acquisition of Icera, Nvidia now has a total processor stack like Qualcomm, and Intel. Qualcomm and Nvidia are going to be head-to-head competitors. Icrea products already compete directly and quite…

AMD’s CPUs disappear from the workstation market

Posted by Alex Herrera on November 25th 2010 | Permalink
Categories Blogs, Engineering and Development
Tags:

Losing its final major workstation OEM customer, AMD seems to have essentially withdrawn from the workstation market for CPUs ... and it doesn't seem to care. It's still selling GPUs but not CPUs. The strategy might have me scratching my head, but when it comes to marketing its wares to the workstation market, that's precisely where AMD's head is at. Yes, AMD maintains a presence as a supplier to the workstation industry, plying its professional-brand FirePro GPUs, but as far as we can tell the company has thrown in the towel when it comes to selling its CPUs into the same…

Market Expansion? ISV’s hold the reins

Posted by Jon Peddie on October 31st 2010 | Permalink
Categories Blogs, Engineering and Development
Tags: market software isv semiconductor

The major volume and/or super expensive software suppliers have had a free lunch for the past fifteen or more years. Basically making bug fixes and a few feature improvements they got performance boasts from CPU clock and pipeline improvements, plus I/O improvements, and dropping DRAM and disc drive costs with increased capacity – the ISVs have been the biggest benefactors of Moore’s law.Now they have to do something. Now they have to embrace Heterogeneous Processors (HPUs), GPU-compute, multi-threading, and parallel-processing.Having spent the last three decades trying to obtain silo dominance with proprietary and incomprehensible file formats, obnoxious, infuriating and incompatible (within…

x86 isn’t do-or-die for Nvidia

Posted by Alex Herrera on October 29th 2010 | Permalink
Categories Blogs, Engineering and Development
Tags:

...but pass on x86, and the company better execute flawlessly on its GPU strategies and technologies Intel's Sandy Bridge and AMD's Fusion are right around the corner, and the die-integrated combinations of x86 CPU and GPU have heightened the interest about how one very specific company will navigate this new landscape of SoCs. Nvidia, the dominant provider of discrete GPUs over the past decade, cannot make the same leap in silicon integration that its two key rivals will imminently launch. The simple reason: it has no x86 IP in its technology arsenal. The easy prediction made by many already is that…

Nvidia, CUDA, and x86

Posted by Alex Herrera on September 23rd 2010 | Permalink
Categories Blogs, Engineering and Development
Tags:

It's not what you're thinking. Announced last week at the company's Graphics Technology Conference (GTC), "CUDA on x86" has nothing to do with Nvidia getting its hands on x86 IP, a notion that's been the fodder for constant speculation over the past few years. Rather, what CUDA-on-x86  accomplishes is the ability to run CUDA-written applications on x86 hardware. The cross-compiler was the result of a collaboration between Nvidia and PGI (Portland Graphics). If one's knee were jerking, one might assume such a thing would run counter to Nvidia's long-term strategy. The company's in the midst of trying to get applications off…

Quadrillions and Quadrillions of Cycles

Posted by Jon Peddie on July 24th 2010 | Permalink
Categories Blogs, Engineering and Development
Tags:

The number of processors, both x86 and GPU, available for rendering has been increasing exponentially. Rendering is one of the applications that can soak up all the cycles that are available to it, which is an example of Peddie’s first law – In computer graphics, too much is not enough. We looked at the installed base of x86 and GPU processors, applied a factor for the average number of cores and developed the following chart. Cores alone don’t tell the whole story, the real measure is how millions of operations per second can the processor execute. A general figure of merit…

Nintendo goes with DMP for S3D graphics engine

Posted by Jon Peddie on June 20th 2010 | Permalink
Categories Blogs, Engineering and Development
Tags:

Today, DMP announced in Japan that Nintendo has adopted DMP OpenGL ES 1.1 compliant PICA200 for the 3DS. The new Nintendo 3DS is an amazing little device. The DS has already been a beloved machine attracting over 100 million users since 2004. Not many products (that I can think of, at least) can match that volume of enthusiasm or the customer base. And it’s self perpetuating because the installed base attracts developers which create new games which attracts new consumers – it is a perfect ecosystem. Nintendo has experimented with S3D for years, starting with the Nintendo Virtual Boy monochrome system, and…

Intel Will Never Buy Nvidia

Posted by Jon Peddie on December 9th 2009 | Permalink
Categories Blogs, Engineering and Development
Tags:

Someone just sent me an email and asked if I thought Intel might buy Nvidia now that Larrabee is dead. I would have just answered it and then disregarded it if I hadn't gotten a phone call asking the same dumb question. Intel won't buy Nvidia for the following reasons: Larrabee isn't dead - there will be a Larrabee graphics chip, based on x86 architecture. There will be a whole family of Larrabee chips. Wishful thinking won't make Intel or its ambitions go away. The company has, and continues to make, huge investments in the graphics technology and space. Intel believes…