ARM Community: Multimedia - ARM Community

Moore’s law is not broken: shock, horror, screaming headline! It will last for a while longer yet - new generations of silicon process will continue to give us more transistors on chips, but that won’t, on its own, give us the increases in performance and decreases in power consumption we have become used to. However, we can utilise those extra transistors, to build multicore processors and more of them, and through heterogeneous computing and appropriate use of domain-specific processors. This will give us the increased performance and improved energy-efficiency we need. These are critical areas for us to concentrate on for the future if we are to continue to lead in energy-efficiency.

At ARM, we focus on energy-efficiency: not just making IP blocks that make the best possible use of energy themselves but also that lead to partner SoCs using less energy overall (e.g. by reducing external memory bandwidth). One of today’s exciting technical challenges is heterogeneous computing, and we invest a lot of time working in this area to enable more energy-efficient consumer electronics devices. Consequently, when AMD kindly invited me to give a keynote speech at their ...

So how did we do it? How did we create the coolest new GPU and architecture? Well, a few people have asked me this recently, so I thought I’d pull together all my stories, videos and blogs into one place (or at least the ones I can mention publicly).

Recently we had The Travelling Salesman stop by on his epic journey around the Nordic start-up scene and I gave him a short interview on our story from founding a tech company right up to being acquired by ARM. You can find the video from his visit below and his blog entry. And no I don’t usually wear sunglasses around the office...

...

How can someone from the company with the most CPUs in use in the world say something like this? Well it wouldn’t be the first time I’ve been told I was wrong and I confidently predict it won't be the last.

GPU Computing
After a recent lecture, I started thinking again about some of the basic assumptions and language in use when this subject gets discussed: "What CPU tasks are suitable to run on a GPU?", for example. On reflection, I feel the language is wrong and I take issue with the question. For example, the CPU has been doing graphics for years and rather badly, at that - just look at the quality and richness of modern consumer electronics devices which use GPUs to see the difference. I believe we should be talking about moving computation tasks to their rightful homes - where they can be done most efficiently in terms of performance and in terms of energy consumption. Does anyone else think I'm on the right lines here? It's not a war between CPU designers and GPU (or video engine) designers about who is right. Surely it's about appropriate engineering design?

Thought-provoking Lecture at University (You Don't Understand Something Until You Teach ...

Recently, a lot of airline miles have been flown between my team and ARM partners to engage in conversations about the latest and future uses of GPUs. Mostly, I have been discussing our graphics roadmap which I still can’t tell you about publicly, but I also got a chance to see some of the latest cool demos on our current GPUs. I also got a chance to hear from our partners – about the slew of consumer products that are coming out soon based on our technology, and about the things they are doing with our GPUs. These partners, a fantastic collection of the most important companies in our industry, really are doing it all: printers, PNDs, digital still cameras, digital TV, set-top-box, automotive, mobile computing, and ...

What is the killer app for OpenCL? It's a question we get asked all the time. Recently, I figured out the answer. The killer app for OpenCL, and for GPU computing in general, is… are you ready? are you sitting down? It’s graphics.

Open C-What?

Let’s back up a bit. OpenCL, if you hadn’t heard, is the latest API for general-purpose computing on GPUs. That is, it lets you use the graphics chip in your laptop / tablet / cell phone to run all kinds of computations that used to run a CPU – and in many cases, to run them a hundred times faster. OpenCL was created by the Khronos Group from a draft specification contributed by Apple. It ships standard with the latest version of Mac OS X, and implementations are available for Windows and Linux as well. (If it’s new to you, check out our previous posts on the topic.) It’s very cool technology. The question is, why do I, my kids, my mom, and (by the way) about a billion other consumers want this technology? Why is it going to be a mus...

Audio is increasingly becoming a mandatory feature in every range of consumer appliances seen of late. Audio centric features like voice instructions, compressed music playback provide for an enhanced user interaction. Such features are now expected in every application segment, even in the low cost applications based on microcontrollers without DSPs. The solution for low cost, efficient audio processing lies in the upcoming microcontrollers based on ARM Cortex-M4.

With the arrival of digital audio and the success of compressed music with innovative ways of enabling these functionalities, speech and audio processing has dramatically changed in the last decade. However, in the embedded world, most of such intensive audio processing is typically done on DSP kind of devices, as they can provide both the capability and the level of power required. Those days are now gone, with future microcontrollers being as capable of processing audio as their DSP counterparts.

“With the launch of ARM Cortex-M4 processor, intensive audio processing can be efficiently...