The quest to gain a greater insight into artificial Intelligence has been exciting and has also opened up a range of possibilities that have included “convolutional neural networks”, these are large visual networks of simple information-processing units which are loosely modelled on the anatomy of the human brain.
These networks are typically implemented using the more familiar graphics processing units (GPUs). A mobile GPU might have as many as 200 cores or processing units, this means that it is suited to “simulating a network of distributed processors”. Now, a further development in this area could lead to the potential for a specifically designed chip that has a sole purpose of implementing a neural network.
MIT researchers have presented the aforementioned chip at the “International Solid-State Circuits Conference in San Francisco”. The advantages of this chip include the notion that it is 10 times more efficient than an average mobile GPU, this could lead, in theory, to mobile devices being able to run powerful artificial intelligence algorithms locally, rather than relying on the cloud to process data.
The new chip, coined “Eyeriss” could, lead to the expansion of capabilities that includes the Internet of things, or put simply, where everything from a car to a cow, (yes apparently) would have sensors that are able to submit real-time data to networked servers. This would then open up horizons for artificial intelligence algorithms to make those important decisions.
Before I sign off I wanted to further delve into the workings of a neural network, the workings are that it is typically organised into layers, each of these layers contains a processing node. Data is then divided up among these nodes within the bottom layer, each node then manipulates the data it receives before passing it on to nodes within the next layer. This process is then repeated until “the output of the final layer yields the solution to a computational problem.” It is certainly fascinating and opens up a world of interesting avenues with which to explore, when you combine science and tech, the outcome is at the very least educational with the potential for it to be life changing. .
It’s been a busy time of year in the CPU market, with Haswell now on sale and making its stand as one mighty processor for its size, it’s time for AMD to release their update to the Trinity APU platform.
Whilst Intel’s new Z87 platform has seen a vast improvement in performance over Z77, it still has one major downside for some people and this relates to the cost. A new ground up platform means that users need to buy a new Z87 board in order to use the latest fourth generation processors and on the top end of the scale, this can equate to a large hole in the wallet. This is where AMD’s APU platform makes a strong stand against Intel. Whilst they have got their FX line of CPU’s that can perform virtually neck and neck with the 3rd Generation offerings from Intel, they do lack a built in GPU.
The APU or Accelerated Processing Unit is something that AMD have been proud of for a while now and the Trinity platform showed that with the inclusion of HD Radeon graphics into the same chip as a quad core CPU, it was able to give quite a substantial amount of power, especially for the price.
Richland is the next generation of APU’s to roll out of the AMD factories and even though AMD have made it clear that their HD Radeon 8xxx series of discrete GPUs will not be around until the early part of next year, back at the start of the year they did state that their 8000 series mobile graphics would be making appearance way before then within notebooks and within their APU’s
So what extra is there to be had over Trinity? Well over the last generation chips, AMD is promising a boost of 30-40% in performance and the biggest shouting point of all is the total cost of upgrading. Whereas Intel users need to buy both a chip and board in order to upgrade, the Richland APUs will all work on the current line of FM2 A85X motherboards with a simple BIOS update.
Other new features within Richland include the new HD Radeon 8000 series GPU cores, with up to 384 shaders, 8xAA and 16xAF support, DX11 support, DisplayPort 1.2 support and a clock speed of up to 844MHz. On top of this the A10 APUs will now also have native support DDR3-2133MHz memory speeds and the chips as a whole will offer more voltage and frequency levels for overclocking meaning that we should see some chips that are easier to work with when taking them to the next level in terms of speed.