gpu News - Page 21 of 31 - CNX Software

September 3, 2015September 3, 2015 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

Qualcomm Snapdragon 820 Promises Twice the Performance & Power Efficiency of Snapdragon 810

Qualcomm has provided more details about Snapdragon 820 processor featuring custom-designed 64-bit ARM Kryo cores, an Adreno 530 GPU and an Hexagan 680 DSP, with claims that it should be twice as fast and consume half the power of Snapdragon 810 processor with Cortex A57 and A53 cores. Based on the chart below, it will also be faster than Snapdragon 620 with Cortex A72 and A53 cores. Snapdragon 820 will feature four Kryo cores (succeeding Krait cores) clocked at up to 2.2 GHz and be manufactured using 14nm FinFET process technology. Adreno 530 GPU is expected to deliver over 40% performance improvement over Adreno 430 GPU while also keeping power consumption 40% lower, a new Spectra camera dual ISP will support 14-bit sensors, and Hexagon 680 DSP adds a separate low power DSP for always-on sensor processing, and support for HVX (Hexagon Vector eXtensions) to be used for advanced imaging […]

August 11, 2015August 11, 2015 by Jean-Luc Aufranc (CNXSoft) - 5 Comments

Freescale i.MX6 DualPlus and QuadPlus SoCs Gets Faster 2D and 3D Graphics, Higher Memory Bandwidth

Freescale unveiled three new models for its i.MX 6 family processor for consumer, industrial and automotive markets in May. Two models are an upgrade to existing i.MX6 Quad and i.MX6 Dual processors, as the new i.MX 6QuadPlus and i.MX6DualPlus processors features four and two Cortex A9 cores, together with improved 2D and 3D GPUs delivering around 50% faster performance, an “optimized” SDRAM controller, more SRAM, and a prefetch and resolve engine. The third model, i.MX 6UltraLite, features a single ARM Cortex A7 core and hardware security, and targets applications such as electronic Point Of Sales (ePOS). That means there are now 9 i.MX6 processors, and today, I’ll focus on the two new “Plus” versions. Since they are based on the original i.MX6 Quad and i.MX6 Dual processors, the best way to have a look at these is to compare them to their predecessors. Features i.MX 6QuadPlus i.MX 6Quad i.MX 6DualPlus […]

August 11, 2015August 11, 2015 by Jean-Luc Aufranc (CNXSoft) - No Comments

Google Plans Vulkan API Support for Android, Imagination Shows a Demo

Vulkan graphics API is the successors to OpenGL and OpenGL ES API, which will support multi-threaded rendering, move some of the complexity to the applications, and simplify graphics drivers, which may not be a bad things since these tend to be closed source, and bugs may be hard to get fixed. Google, which is now a subsidiary of a new company called Alphabet, has recently announced that Vulkan will be implemented in future versions of Android, although OpenGL ES will still be supported, so developers can select their preferred graphics API for their apps, as Vulkan will be more complex for application programmers than OpenGL ES. Separately, Imagination technologies showcased their Vulkan driver for PowerVR Rogue GPU on the Nexus Player powered by an Intel SoC including a PowerVR G6430 GPU, and compare the Vulkan demo to the same demo using OpenGL ES 3.0 drivers. The difference between Vulkan and OpenGL […]

July 14, 2015July 14, 2015 by Jean-Luc Aufranc (CNXSoft) - 26 Comments

ODROID-XU4 Development Board is a Smaller and Cheaper Version of ODROID-XU3 Board

Many people are expecting Hardkernel to launch an Exynos 7 board, but this won’t be with ODROID-XU4, maybe with ODROID-XU5, as the latest development board from Hardkernel is a modified version of ODROID-XU3 with the same Exynos 5422 octa-core processor with four Cortex-A15 cores @ 2GHz and four Cortex-A7 cores. ODROID-XU4 board specifications (differences with ODROID-XU3 highlighted and sometimes crossed out in bold): SoC – Samsung Exynos 5422 quad core ARM Cortex-A15 @ 2.0GHz quad core ARM Cortex-A7 @ 1.4GHz with Mali-T628 MP6 GPU supporting OpenGL ES 3.0 / 2.0 / 1.1 and OpenCL 1.1 Full profile System Memory – 2GB LPDDR3 RAM PoP (750 MHz, 12GB/s memory bandwidth, 2x32bit bus) Storage – Micro SD slot (up to 64GB) + eMMC 5.0 module socket (16, 32, and 64GB modules available) Video Output – micro HDMI (Up to 1080p) and DisplayPort (up to 2160p) Audio Output – HDMI, 3.5mm headphone jack, […]

June 18, 2015 by Jean-Luc Aufranc (CNXSoft) - 9 Comments

Open Source Linux Drivers for PowerVR GPUs Might Be in the Works

When I write about a new processor with a PowerVR GPU, I can be sure there will be one or two comments saying something like “It has a PowerVR GPU, it’s a non-starter”, mostly because of the status of graphics drivers in Linux. Luc Verhaegen (libv) also made a presentation at FOSDEM 2013 listing various open source projects for GPUs found in ARM processors and PowerVR was the only one without any community project. Bear in mind that apart from Nvidia, and to some extend Broadcom for the VideoCore IV GPU found in BCM2835 / BCM2836 processors used in Raspberry Pi boards, no other companies have released user space drivers for their GPUs, and all work is done by volunteers for other open source GPU drivers, and I’m not sure any of them work 100% yet. Imagination has already provided working GPU drivers for their MIPS Creator-CI20 board, these are […]

March 11, 2015March 11, 2015 by Jean-Luc Aufranc (CNXSoft) - No Comments

Embedded Systems Conference 2015 Schedule – May 6-7, 2015

The Embedded Systems Conference took the name “Design West” for a couple of years, but this year, there’s no mention of Design West, and the Embedded System Conference 2015 will take place in Boston, MA, US on May 6-7, 2015. The 2-day event will have a demo hall, and well as sessions divided into 8 tracks: Connected Devices and the IoT Embedded Software Design Hardware: Design, I/O and Interfacing Prototyping Embedded Systems Design Software: Design, Languages, & Quality Fantastical Theater Teardowns The full schedule has now been posted, and I’ll build a virtual schedule with some of the sessions provided. Wednesday May 6, 2015 8:00 – 8:45 – Understanding Google/Nest Thread by Michael Anderson, Chief Scientist, The PTR Group, Inc. The IoT will live or die based on its connectivity. In examining existing wireless protocols, Google/Nest found most of them lacking. In order to address the needs for low-power wireless […]

March 4, 2015 by Jean-Luc Aufranc (CNXSoft) - 7 Comments

Meet Vulkan, The Successor of OpenGL and OpenGL ES 3D Graphics APIs

So far embedded systems are typically using SoCs with GPU supporting OpenGL ES, a subset of the full fledge OpenGL API used in desktop computers and workstations. These royalty-free standards are defined by Khronos Group non-profit organization, and as features in the embedded space and traditional computers merge, the group has now revealed the next-generation OpenGL specs will be called Vulkan. The new API will run on GPUs supporting OpenGL ES 3.1 or greater, take less CPU resources than its predecessors, and support multiple command buffers that can be created in parallel. More work will be required at the application level, but direct GPU control by the drivers will apparently result in less memory copies improving performance, or at least off-loading the CPU. It might be transparent to developers using game engines. GPU drivers will also be less complex. Vulkan will also use the new SPIR-V language shared with OpenCL […]

February 27, 2015 by Jean-Luc Aufranc (CNXSoft) - 3 Comments

Imagination PowerVR GT7900 GPU is Made for Game Consoles and Notebooks

Imaginatino Technologies has just introduced the most powerful GPU of their new PowerVR Series7 GPU family with GT7900 featuring 16 unified shading clusters and 512 ALU cores for game consoles, micro-servers, and notebooks. PowerVR GT7900 is still an embedded GPU with support for OpenGL ES 3.1 + AEP (Android Extension Pack) compatibility, full hardware support for tessellation and ASTC LDR and HDR texture compression standards. The new GPU delivers up to 800 GFLOPS (FP32) or 1.6 TFLOPS (FP16) graphics and compute performance at 800 MHz (using 16 nm FinFET+ manufacturing process), which compares to about 550 GFLOPS (FP32) for an Nvidia GeForce GT730M at comparable frequency. This is even more powerful than the Maxwell GPU found in Tegra X1 which delivers 1+ TFLOPS (FP16), although the Nvidia processor may have a lower power envelope. Combined with PowerVR Series5 decode and encode video processors and PowerVR V2500 camera processor, GT7900 can […]