Skip navigation

community

Up to Blog Posts in News

News

3 Posts tagged with the gpu tag
Cabe Atwell
2

Rosepoint3_p.jpg

Rosepoint chip (via Intel)

 

‘Fused’ chips are fast becoming the status quo in powering today’s mobile devices, particularly tablets and smartphones. For those of you who don’t know what fused chips are, they combine CPU’s and  For those of you who don’t know what fused chips are, they combine CPU’s and GPU’s on a single chip (or die) such as AMD’s Fusion. Intel has recently stepped up their game in this field with the introduction of their Sandy Bridge line of fused chips, but they have not stopped the integration there.

 

 

The company has recently stated that they have combined Wi-Fi with their line of Atom processors code named Rosepoint which will be unveiled at this year’s International Solid-State Circuits Conference in San Francisco. Not much is known about Rosepoint but a few ‘leaked’ images and a vague Intel press release. Details say that it features a 32nm SoC with a built-in Wi-Fi transceiver (running at a reported 2.4 GHz or 4G) with two Atom CPU’s all crammed onto the same die. Another goal is to reduce the chip-count. Although a wireless transmitter that close to other digital signals would cause interference, Intel has found some "hush-hush" way to shield the CPU from the WiFi onboard. The integration of wireless onto CPU cores means less power usage as well as costs. If all goes well, the technology could be found in mobile devices as early as 2013.

 

 

More information will be released at this year’s ISSCC so check back for an update! (ISSCC runs from February 19-23rd.)

 

Cabe

http://twitter.com/Cabe_e14

363 Views 2 Comments Permalink Tags: wireless, design, wifi, intel, communication, 4g, industry, innovation, prototyping, research, cpu, atom, gpu, cabeatwell
Cabe Atwell
3

Engineering On Friday CPU GPU a Toast to Us by Cabe Atwell b.jpg

If your reading this, chances are your using a laptop or PC to do so. Being that this article is on an engineering oriented website you most likely have some knowledge as to what components are housed in either one of them. I'm talking primarily about the CPU (Central Processing Unit) and the GPU (Graphical Processing Unit) that make up the computer's brain and muscle respectively.

 

The CPU processes complex code or software and executes the instructions based on what app/software is being used. However, it has a tough time when it comes to executing code that uses high-intensive 3D images or graphics. That is where the GPU comes in. It offloads most of the work needed for images from the host CPU and allows it to crunch 1's and 0's for other tasks. Companies like AMD and NVIDIA have even combined the two on a single die (or chip).  While they work in tandem with software, they do not really communicate with each other. Sad, I know.

 

All is not lost, as engineers from North Carolina State University have found a way to overcome that problem and even give the hybrid processor a 20% increase in performance. Dr. Huiyang Zhou, an associate professor of electrical and computer engineering, and his team accomplished this by having the GPU portion of the chip handle the computations while the CPU 'fetches' the data the GPU needs from system memory. Both grab data from system memory at relatively the same speed. However, the GPU can crunch the numbers faster when it comes to graphics, but the CPU is quicker when it comes to what information the GPU will need to accomplish its task. That makes the whole process more efficient according to Dr. Zhou. In recent tests, the team found that 'fused' chips increased their performance by 21.4%, which is no small feat, as any overclocker will tell you. Some tasks even rocketed over 114% faster.

 

fusion.jpgamd-fusion-desktop-roadmap.jpg

(Left) AMD Fusion APU (Right) Partial Roadmap (via AMD)

 

The research was partly funded by AMD, and the experiment was simulated on a future Accelerated Processing Unit (APU) where there is a shared L3 cache. The technique may be publicly available rather soon.

 

Cabe

http://twitter.com/Cabe_e14

 

See more Engineering On Friday comics in the Engineering Life group.

116 Views 3 Comments Permalink Tags: design, embedded, communication, industry, innovation, measurement, prototyping, university, research, hmi, cpu, amd, gpu, apu, on_campus, cabeatwell, engineering_on_friday
Cabe Atwell
2

graphics6.jpg

Imagination press image

 

Imagination dazels people when it is applied to technology (this is why there is a CES). A company of the same name, Imagination, has amazed many with their new PoweVR line of GPU’s. Currently, mobile devices' raw graphics power being dominated by giants such as AMD, Intel and NVIDIA. However, they may have to shave of a percent or two of dominance with Imagination’s G6200 and G6400 GPU IP cores. The company says that these cores are scalable for both the mobile market as well as high-end gaming machines. What sets these GPU’s apart from the others is use of what’s called ‘Compute Clusters’.


These are a cluster of programmable arrays that spread the ‘work load’ giving them efficiency in both power and bandwidth. Think of it like a LAN party on a chip, or like Stanford University’s Folding At Home project, where a ton of computers work on one project without any one PC handling the entire load. Imagination’s G6200 is equipped with two clusters while the G6400 takes advantage of four giving those 100GFLOPS and in some cases in the TFLOP range! All the series 6 in Imagination’s PowerVR line feature OpenGL 3.x/4.x, OpenCL and DirectX 10 (in some versions DirectX 11.1 is implemented). Imagination states that these new chips are 20 times more powerful than the current generation out today while being 5 times more efficient.


Cabe

http://twitter.com/Cabe_e14

187 Views 2 Comments Permalink Tags: design, embedded, communication, mobile, cell_phone, research, gaming, hmi, cpu, graphics, gpu, ces, cabeatwell, ces2012