Recent Posts

Pages: 1 ... 5 6 [7] 8 9 10
61
Quote
Welcome OpenCL™ developers and enthusiasts!  Here you will find a BETA version of our OpenCL 1.2 driver which includes HSA extensions that implement a subset of the functionality from the recently ratified OpenCL 2.0 specification.  This functionality is built on top of the HSA capabilities in AMD’s recently launched 4th generation APU desktop processors (formerly known as “Kaveri”) and include much of the Shared Virtual Memory and Platform Atomics functionality in the OpenCL 2.0 specification.

This BETA release is intended to enable developers who wish to prototype applications using this OpenCL 2.0 functionality now, rather than wait for the AMD driver compatible with OpenCL 2.0. OpenCL 2.0 will be our primary enablement path with a new driver available later this year to enable these features for developers, and so AMD does not plan to deliver a production version of OpenCL 1.2 with these extensions. OpenCL 2.0 will be a major milestone for our industry, and AMD expects it to become the developer tool of choice for heterogeneous computing enablement.

I posted OpenCL 2.0 functions list here

RTFM and edit path and registry

Recent regular Catalyst betas already contain HSA core, check with CLINFO.EXE
Quote
  Platform Version:             OpenCL 1.2 AMD-APP (1411.4)
  Platform Extensions:             cl_khr_icd cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_khr_dx9_media_sharing cl_amd_event_callback cl_amd_offline_devices cl_amd_hsa

62
Quote
NVIDIA, AMD, Intel Explain How OpenGL Can Unlock 15x Performance Gains
If you want to get a developer’s attention, all you need to do is start dropping whole numbers.
Offer something that’s not 1.2 times better — but two or three times better — you know you’ve got them.

With OpenGL, an open, vendor-neutral standard, developers can get significantly better performance – up to 1.3 times. But with a little tuning, they can get 7 to 15 times more performance.

63
3D-Tech News Around The Web / Civilization V Gets OpenCL Update for Mac Pro
« Last post by Stefan on March 21, 2014, 01:52:58 AM »
Quote
Aspyr has just released a new update for Civilization V that, among other improvements, takes advantage of the second video card in the Mac Pro.

Mac gamers that play on the Mac Pro will now be able to fully take advantage of the machine’s second video card, which means better visuals and improved performance for the critically acclaimed strategy game. In particular, Civilization V on the Mac Pro will now support 4k monitors while running into fewer crashes. Additionally, the game should run faster (meaning less waiting in between turns in late-game situations).

This update makes Civilization V the only Mac game that currently supports the second video card on the Mac Pro. It should automatically download and install for Steam users; Mac App users can head to the Mac App Store to get the update. For even more info on how to take advantage of the new update, check out the knowledge base article over at our Support site.
64
3D-Tech News Around The Web / AMD Catalyst 14.3 LINUX Beta V1.0 Driver
« Last post by Stefan on March 20, 2014, 07:26:20 PM »
Highlights of the AMD Catalyst™ 14.3 Beta V1.0 Driver for Linux:

Resolved Issues:

        [394848] - Xorg crashed  playing AVI video file in VLC player or Parole player
        [394705] - "Nexuiz - Demo3" Ubuntu performance lower than Redhat
        [394704] - "Nexuiz - Demo3" Linux performance  lower than Windows
65
3D-Tech News Around The Web / DirectX 12: A Major Stride for Gaming
« Last post by Stefan on March 20, 2014, 07:23:01 PM »
Quote
Microsoft today introduced DirectX 12 at the annual Game Developers Conference (GDC) in San Francisco. DX12 is Microsoft’s latest version of the graphics API that is the dominant standard in the growing, $25 billion PC gaming industry.

Developers have been asking for a thinner, more efficient API that allows them to control hardware resources more directly. Despite significant efficiency improvements delivered by continuous advancement of existing API implementations, next-generation applications want to extract all possible performance from multi-core systems. Developers also want to take direct advantage of advanced GPU hardware features, from which developers are currently insulated to provide fool-proof usage. DirectX 12 was designed from scratch to provide the infrastructure for these advanced applications.

read more at NVIDIA blog
66
3D-Tech News Around The Web / Blender 2.70 Final is Now Available
« Last post by Stefan on March 20, 2014, 05:55:10 PM »
Quote
After two release candidates, the Blender Foundation has now released Blender 2.70.

The official release still don't support CUDA sm5.0 kernels.
Blender Artists to the rescue:

Quote from: Rolf
Bug is fixed and 2.70 is out now, so here some updates
Blender 2.70 64-bit Windows with sm_50 kernel Hash: cc2cdfb
http://goo.gl/3FKOHI

Blender 2.70 32-bit Windows with sm_50 kernel Hash 19fcb4d
http://goo.gl/ZfD2rY

and for testing a special sm_50 kernel, this kernel enabled subsurface scattering and volume scatter/absorption on 750/ti, only for 64bit.
Vram usage is much higher with this kernel.
Put this file to ...Blender\2.70\scripts\addons\cycles\lib\
http://goo.gl/l0ZIWC

Quote from: JuicyFruit
I have also compiled sm_50 kernels for release (both 32 bit and 64 bit)

windows http://martijnberger.nl/file/2.70_sm50.zip

linux http://martijnberger.nl/file/2.70_sm50.tar.gz

extract into 2.70\scripts\addons\cycles\lib\ windows

67
Quote
PowerVR Wizard brings ray tracing to real-time graphics

...
Today at GDC, the firm announced that its PowerVR Wizard GR6500 graphics processor IP is available for licensing, and it also revealed a major software alliance to enable game developers to take advantage of Wizard's custom ray-tracing hardware.

The Wizard GPU is based on the PowerVR Rogue graphics processor that we're pretty darned certain resides in Apple's latest phones and tablets. More specifically, the PowerVR GR6500 is a ray-tracing enabled version of the PowerVR Series6XT Rogue core...
68
3D-Tech News Around The Web / Crytek Adds NVIDIA GameWorks to “Warface”
« Last post by Stefan on March 19, 2014, 06:06:52 PM »
Quote
Crytek Adds NVIDIA GameWorks to “Warface”

We’re announcing today at the Game Developers Conference, in San Francisco, that we’ve teamed up with Crytek to add support for several NVIDIA GameWorks technologies to its hit free-to-play online first-person shooter “Warface,” powered by CRYENGINE, their award-winning game development solution.

CRYENGINE has been praised for over a decade for its advanced technology. It’s the framework behind Crytek’s own hit franchises, including “Crysis” and “Ryse: Son of Rome.” It’s also licensed to top developers, such as NCsoft, XL Games, Tencent and 2K Games.

“Warface” uses CRYENGINE’s advanced graphics and game technology to create a fast-paced, online first-person shooter, featuring intuitive controls and slick gameplay. The addition of GameWorks technologies like NVIDIA PhysX and VisualFX will take the gameplay in “Warface” to a higher plane of realism and engagement.
69
So OpenCL 1.2 has 98 functions, OpenCL 2.0 has 111 functions.
I'm too lazy to isolate the new ones though  :P

OpenCL 1.2 function list
Code: [Select]
clBuildProgram
clCompileProgram
clCreateBuffer
clCreateCommandQueue
clCreateContext
clCreateContextFromType
clCreateFromGLBuffer
clCreateFromGLRenderbuffer
clCreateFromGLTexture2D
clCreateFromGLTexture3D
clCreateFromGLTexture
clCreateImage2D
clCreateImage3D
clCreateImage
clCreateKernel
clCreateKernelsInProgram
clCreateProgramWithBinary
clCreateProgramWithBuiltInKernels
clCreateProgramWithSource
clCreateSampler
clCreateSubBuffer
clCreateSubDevices
clCreateUserEvent
clEnqueueAcquireGLObjects
clEnqueueBarrier
clEnqueueBarrierWithWaitList
clEnqueueCopyBuffer
clEnqueueCopyBufferRect
clEnqueueCopyBufferToImage
clEnqueueCopyImage
clEnqueueCopyImageToBuffer
clEnqueueFillBuffer
clEnqueueFillImage
clEnqueueMapBuffer
clEnqueueMapImage
clEnqueueMarker
clEnqueueMarkerWithWaitList
clEnqueueMigrateMemObjects
clEnqueueNDRangeKernel
clEnqueueNativeKernel
clEnqueueReadBuffer
clEnqueueReadBufferRect
clEnqueueReadImage
clEnqueueReleaseGLObjects
clEnqueueTask
clEnqueueUnmapMemObject
clEnqueueWaitForEvents
clEnqueueWriteBuffer
clEnqueueWriteBufferRect
clEnqueueWriteImage
clFinish
clFlush
clGetCommandQueueInfo
clGetContextInfo
clGetDeviceIDs
clGetDeviceInfo
clGetEventInfo
clGetEventProfilingInfo
clGetExtensionFunctionAddress
clGetExtensionFunctionAddressForPlatform
clGetGLObjectInfo
clGetGLTextureInfo
clGetImageInfo
clGetKernelArgInfo
clGetKernelInfo
clGetKernelWorkGroupInfo
clGetMemObjectInfo
clGetPlatformIDs
clGetPlatformInfo
clGetProgramBuildInfo
clGetProgramInfo
clGetSamplerInfo
clGetSupportedImageFormats
clLinkProgram
clReleaseCommandQueue
clReleaseContext
clReleaseDevice
clReleaseEvent
clReleaseKernel
clReleaseMemObject
clReleaseProgram
clReleaseSampler
clRetainCommandQueue
clRetainContext
clRetainDevice
clRetainEvent
clRetainKernel
clRetainMemObject
clRetainProgram
clRetainSampler
clSetCommandQueueProperty
clSetEventCallback
clSetKernelArg
clSetMemObjectDestructorCallback
clSetUserEventStatus
clUnloadCompiler
clUnloadPlatformCompiler
clWaitForEvents

OpenCL 2.0 functions list
Code: [Select]
clBuildProgram
clCompileProgram
clCreateBuffer
clCreateCommandQueue
clCreateCommandQueueWithProperties
clCreateContext
clCreateContextFromType
clCreateFromGLBuffer
clCreateFromGLRenderbuffer
clCreateFromGLTexture2D
clCreateFromGLTexture3D
clCreateFromGLTexture
clCreateImage2D
clCreateImage3D
clCreateImage
clCreateKernel
clCreateKernelsInProgram
clCreatePipe
clCreateProgramWithBinary
clCreateProgramWithBuiltInKernels
clCreateProgramWithSource
clCreateSampler
clCreateSamplerWithProperties
clCreateSubBuffer
clCreateSubDevices
clCreateUserEvent
clEnqueueAcquireGLObjects
clEnqueueBarrier
clEnqueueBarrierWithWaitList
clEnqueueCopyBuffer
clEnqueueCopyBufferRect
clEnqueueCopyBufferToImage
clEnqueueCopyImage
clEnqueueCopyImageToBuffer
clEnqueueFillBuffer
clEnqueueFillImage
clEnqueueMapBuffer
clEnqueueMapImage
clEnqueueMarker
clEnqueueMarkerWithWaitList
clEnqueueMigrateMemObjects
clEnqueueNDRangeKernel
clEnqueueNativeKernel
clEnqueueReadBuffer
clEnqueueReadBufferRect
clEnqueueReadImage
clEnqueueReleaseGLObjects
clEnqueueSVMFree
clEnqueueSVMMap
clEnqueueSVMMemFill
clEnqueueSVMMemcpy
clEnqueueSVMUnmap
clEnqueueTask
clEnqueueUnmapMemObject
clEnqueueWaitForEvents
clEnqueueWriteBuffer
clEnqueueWriteBufferRect
clEnqueueWriteImage
clFinish
clFlush
clGetCommandQueueInfo
clGetContextInfo
clGetDeviceIDs
clGetDeviceInfo
clGetEventInfo
clGetEventProfilingInfo
clGetExtensionFunctionAddress
clGetExtensionFunctionAddressForPlatform
clGetGLObjectInfo
clGetGLTextureInfo
clGetImageInfo
clGetKernelArgInfo
clGetKernelInfo
clGetKernelWorkGroupInfo
clGetMemObjectInfo
clGetPipeInfo
clGetPlatformIDs
clGetPlatformInfo
clGetProgramBuildInfo
clGetProgramInfo
clGetSamplerInfo
clGetSupportedImageFormats
clLinkProgram
clReleaseCommandQueue
clReleaseContext
clReleaseDevice
clReleaseEvent
clReleaseKernel
clReleaseMemObject
clReleaseProgram
clReleaseSampler
clRetainCommandQueue
clRetainContext
clRetainDevice
clRetainEvent
clRetainKernel
clRetainMemObject
clRetainProgram
clRetainSampler
clSVMAlloc
clSVMFree
clSetCommandQueueProperty
clSetEventCallback
clSetKernelArg
clSetKernelArgSVMPointer
clSetKernelExecInfo
clSetMemObjectDestructorCallback
clSetUserEventStatus
clUnloadCompiler
clUnloadPlatformCompiler
clWaitForEvents
70
3D-Tech News Around The Web / Khronos Releases Wave of New Standards and Initiatives
« Last post by Stefan on March 19, 2014, 05:23:01 PM »
Quote
The Khronos Group today announced a number of new and significant updates to its portfolio of open, royalty free industry standards that enable the authoring and acceleration of parallel computing, graphics, vision, sensor processing and dynamic media on a wide variety of platforms and devices:

    OpenGL® ES 3.1 Specification Released – The next leap in mobile & embedded graphics
    WebCL™ 1.0 Specification Released – Browser-based heterogeneous compute acceleration
    SYCL™ 1.2 Provisional Specification Released – Enabling high-level C++ tools for OpenCL
    OpenCL™ 2.0 Adopters Program Launched – Driving OpenCL 2.0 conformance
    EGL™ 1.5 Specification Released – The essential glue for secure, integrated API rendering

AMD is probably already working on OpenCL 2.0 - at least the recent Catalysts come with Khronos OpenCL ICD 2.0 (OPENCL.DLL)
Pages: 1 ... 5 6 [7] 8 9 10