Recent Posts

Pages: 1 [2] 3 4 ... 10
1. Improved font readability with revised font
2. Fixed an UI related issue where parts of the software was not being displayed properly
3. Resolved the software conflict issue between GPU Tweak II and AURA software
4. Fixed a bug where running the stress test in FurMark ROG edition would have a pop-up indicating that the software has "stopped working"
5. Fixed a bug where using multiple cards in SLI or crossfire would cause the monitor tab to close or stop working if running any 3D application in full screen mode
6. Fixed a bug where four AMD cards in 4-way crossfire would prevent GPU Tweak II from starting up properly
7. Fixed a bug where reinstalling to a newer version of GPU Tweak II would cause the 0dB feature to not function properly
 In collaboration with Google, Codeplay is proud to announce the release of a new open-source tool allowing the compilation of OpenCL C language kernels to run on the Vulkan API.

Vulkan is a low-overhead, cross-platform 3D graphics and compute API, available on the Android, Linux, and Windows platforms. OpenCL C is the kernel language used in the OpenCL API allowing developers to program for execution on heterogeneous devices.
The tool, named 'clspv', allows a subset of the OpenCL C language to be targeted at the Vulkan API. This tool allows developers to port code containing more than a million lines of OpenCL C to run on the Vulkan API.
As active members of the Khronos Vulkan working group, Codeplay has been instrumental in pushing two extensions that the clspv codebase requires - VK_KHR_variable_pointers and VK_KHR_16bit_storage, which were announced yesterday by Khronos.
The tool source code is available at and is available under the Apache 2.0 open source license.
Some notes on the clspv:
  • [size=1.1rem]This is a prototype, with known issues.  However, it does compile significant portions of kernels[/size]
  • [/size][size=1.1rem]It uses close to top-of-tree versions of LLVM and Clang as third-party projects - this is not a fork[/size]
  • [/size][size=1.1rem]It is mainly a set of LLVM module passes. The flow starts with the SPIR code generated from stock Clang, then massages it into SPIR-V for Vulkan[/size]
  • [/size][size=1.1rem]It exercises SPV_KHR_variable_pointers (rev13)[/size]
3D-Tech News Around The Web / NVIDIA Vulkan beta driver 382.83
« Last post by Stefan on July 15, 2017, 08:12:40 AM »
 Windows 382.83
 Release Updates

July 14th, 2017 - Windows 382.83
  • Fixed issue with unused compute queue
  • Other minor bug fixes
Downloads added:

- VS2015 Redistributable Update 3 (32-bit):

- VS2015 Redistributable Update 3 (64-bit):

3D-Tech News Around The Web / Vulkan SDK 1.0.54 Released
« Last post by Stefan on July 14, 2017, 01:21:03 AM »
SDK Version now available

 Released: July 13, 2017

This new Vulkan SDK release includes the following updates:
  • Introduction of new groups of extension functionality: external fence extensions, external semaphore extensions, external memory extensions, storage memory extensions, and additional vendor and cross-vendor extensions
  • Renaming of formerly experimental extensions:  many of the external semaphore extensions and the external memory extensions were previously released as experimental Khronos extensions using the VK_KHX prefix.  These extensions are now official and stable and have been renamed.
  • Updates to loader, layers, and documents to reflect new extensions and changes defined in Vulkan version
  • Scores of bug fixes and validation coverage improvements.
Please read the Release Notes on LunarXchange in the Docs section for more details.
3D-Tech News Around The Web / Vulkan API specifications 1.0.54 released
« Last post by Stefan on July 13, 2017, 08:29:33 PM »
Change log for July 13, 2017 Vulkan 1.0.54 spec update:

  * Bump API patch number and header version number to 54 for this update.

Github Issues:

Internal Issues:

  * Fix tessellation domain to have an upper-left origin in the
    <<img-tessellation-topology-ul, tessellation toplogy image>> and related
    language. CTS and all implementations were already doing this, it was
    just a documentation bug that it was flipped to lower-left (internal
    issue 603).
  * Add a section to the style guide describing how VUID tags are changed
    and removed when the corresponding Valid Usage statements are modified
    (internal issue 829).
  * Add explicit Valid Usage statement to
    slink:VkPipelineDynamicStateCreateInfo to require that members of
    pname:pDynamicStates must be unique (internal issue 851).

New Extensions:

  * `VK_KHR_16bit_storage`
  * `VK_KHR_dedicated_allocation`
  * `VK_KHR_external_fence`
  * `VK_KHR_external_fence_capabilities`
  * `VK_KHR_external_fence_fd`
  * `VK_KHR_external_fence_win32`
  * `VK_KHR_get_memory_requirements2`
  * `VK_KHR_storage_buffer_storage_class`
  * `VK_KHR_variable_pointers`

Extensions Promoted From KHX To KHR Status:

  * `VK_KHR_external_memory`
  * `VK_KHR_external_memory_capabilities`
  * `VK_KHR_external_memory_fd`
  * `VK_KHR_external_memory_win32`
  * `VK_KHR_external_semaphore`
  * `VK_KHR_external_semaphore_capabilities`
  * `VK_KHR_external_semaphore_fd`
  * `VK_KHR_external_semaphore_win32`
  * `VK_KHR_win32_keyed_mutex`

3D-Tech News Around The Web / Unity 2017.1 has been released
« Last post by Stefan on July 13, 2017, 07:12:26 PM »
We’re excited to announce that Unity 2017.1 has been released and is now available for download. We want to thank the Unity community for their valuable contributions during the beta phase.

Read more

I'll make some tests over the weekend...
3D-Tech News Around The Web / AMD Crimson ReLive Edition 17.7.1
« Last post by Stefan on July 11, 2017, 12:46:19 AM »
Radeon Software Crimson ReLive Edition 17.7.1 Highlights

Support For
    • Radeon RX 460/470 Series Graphics-based mining cards
    • Radeon RX 550/560 Series Graphics
Fixed Issues
    • Fixed a bug causing crashes on Tekken 7 on Radeon RX 380 Series graphics products.
    • Fixed a bug causing crashes on FFXIV and Little Nightmares on Radeon RX 300 Series graphics products.
    • Fixed an issue causing crashes with Adobe Lightroom CC 2015.10
3D-Tech News Around The Web / AMD - What’s new in HIP and HCC for ROCm 1.6
« Last post by Stefan on July 06, 2017, 08:37:27 PM »
HIP has a new home HIP now easier-to-use
  • We’ve made it even easier to port CUDA code to HIP:
    • The hipLaunchParm extra kernel argument is no longer necessary, and HIP does not modify the function signature for __global__ functions.  This removes one of last manual steps in the porting process.  Just use “hipLaunchKernelGGL” instead of hipLaunchKernel
    • HIP_KERNEL_NAME wrapper is only needed if the kernel name include commas. Typically, this only occurs in a template with multiple parameters.
  • HIP now provides two flavors of the hipify translation tools:
    • hipify-perl :  Does a simple text-based conversion for the CUDA code into HIP.  Easy to use and useful for quick scans and conversions, but can be fooled by macros or complex code paths.
    • hipify-clang: Parses the CUDA code using clang and performs a translation of the symbols.   Requires some setup to configure the correct source paths and defines, but provides most robust translation.  Useful if you are familiar with the source code and make infrastructure.
Faster performance
  • The ROCm 1.6 HCC compiles HIP code up to 2X faster than the version included in ROCM1.4.
  • HCC code generation includes latest version of the LLVM Direct-To-GCN compiler, which includes optimizations for scheduling and register allocation.
  • HIP/HCC now optimize GPU cache flushing so it is performed only as-needed, rather than on each kernel command.
Pages: 1 [2] 3 4 ... 10