Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Stefan

Pages: 1 [2] 3 4 ... 199
3D-Tech News Around The Web / HWiNFO32 & HWiNFO64 v5.54
« on: July 06, 2017, 05:14:18 PM »
Changes in HWiNFO32 & HWiNFO64 v5.54 - Released on:  Jul-6-2017: 
  • Extended reporting of NVMe drive temperatures and added thresholds to sensors.
  • Improved calculation of Corsair PSU efficiency.
  • Added monitoring of EVGA iCX additional GPU temperatures and fans.
  • Added support of Corsair Link Commander Pro (CLCP).
  • Added reporting of CPU process technology (nm).
  • Added monitoring of NVIDA GPU performance limiting reasons.
  • Enhanced sensor monitoring on ASUS ROG STRIX X299-E GAMING.
  • Enhanced sensor monitoring on ASUS PRIME X299-A, PRIME X299-DELUXE and TUF X299 series.
  • Enhanced sensor monitoring on MSI X299 series mainboards.
  • Improved measuring of BCLK on Skylake-X.
  • Fixed a bug causing Shift- and Alt- hot key combinations not to work properly.
  • Added support of RTSS OSD custom colors (requires RTSS v7.0.0 or later).
  • Added support of RTSS OSD graphs (requires RTSS v7.0.0 beta 24 or later).
  • Added AMD Radeon Vega Frontier Edition.
  • Added option to specify custom/fixed font size for tray icons.
  • Improved option for number alignment in RTSS OSD.
  • Enhanced sensor monitoring on ASUS ROG ZENITH EXTREME.
  • Enhanced sensor monitoring on GIGABYTE X299 series.
  • Enhanced support of AMD Radeon Vega.

3D-Tech News Around The Web / NVIDIA Profile Inspector
« on: July 05, 2017, 06:39:38 PM »
  fixes for binary values
 Download precompiled binary from Appveyor

When getting a new piece of hardware, the first step is to install the driver. You can see how to install them for the Radeon Vega Frontier card by following this link.
Then as a developer you need to access tools and libraries to use that new hardware.


3D-Tech News Around The Web / NVIDIA Vulkan beta driver 382.71
« on: July 01, 2017, 12:49:51 AM »
 Windows 382.71
  Linux 381.26.06
 Release Updates June 30th, 2017 - Windows 382.71, Linux 381.26.06
  • Various performance improvements and bug fixes

3D-Tech News Around The Web / Re: NVIDIA Game ready driver 384.76
« on: July 01, 2017, 12:42:56 AM »
 Some tidbits:

 the new branch is mandatory for VRWorks SDK V2.4  and Ansel SDK 1.4
 CUDA has been updated to v9.0.103
 As expected it supports compute capability 7.0 for Volta GPUs
 Surprisingly also compute capability 6.9 is exposed in nvcompiler*.dll
 Probably for the "Pascal refresh" GPUs speculated by the usual rumour sites?

Change log for June 24, 2017 Vulkan 1.0.53 spec update:

  * Bump API patch number and header version number to 53 for this update.

Github Issues:

Internal Issues:

  * Clarify mappings of coordinates for mutable, compatible image views in
    slink:VkImageViewCreateInfo (internal issue 815).
  * Make ename:VK_BIND_SFR_BIT require a logical device with multiple
    physical devices, so that standard sparse image block dimensions are
    only required on systems that support multi-GPU (internal issue 835).
  * Convert all files from use of // refBegin .. // refEnd comments to
    delimit ref pages, to use of open blocks, and update style guide
    accordingly (internal issue 839).
  * Add valid usage for slink:VkWriteDescriptorSet when performing updates
    to a ename:VK_STORAGE_IMAGE descriptor with layout
  * Add a hack to the validity generator script to support an odd
    interaction between flink:vkCmdFillBuffer and an extension (internal
    issue 853).
  * Remove redundant text describing slink:VkBufferCreateInfo::pname:usage,
    which was already covered by implicit valid usage (internal issue 854).
  * Update implicit validity generator script to properly handle the
    pname:sType and pname:pNext members of "returnedonly" structures
    (internal issue 874).
  * Note that slink:VkApplicationInfo::pname:pApplicationName &
    slink:VkApplicationInfo::pname:pEngineName are optional, and add missing
    implicit valid usage statements for flink:vkDestroyInstance.
  * Added missing valid usage for flink:vkCmdWriteTimestamp to require a
    timestamp query pool.
  * Simplify and/or split "`non-atomic`" valid usage statements.

New Extensions:

  * `VK_AMD_gpu_shader_int16`
  * `VK_EXT_blend_operation_advanced`
  * `VK_EXT_sampler_filter_minmax`
  * `VK_NV_framebuffer_mixed_samples`

Note: the 1.0.52 spec wasn't published on github, so the 1.0.53 release
combines both change sets.

Change log for June 13, 2017 Vulkan 1.0.52 spec update:

  * Bump API patch number and header version number to 52 for this update.

Github Issues:

Internal Issues:

  * Clarify behavior when non-coherent memory has
    <<memory-device-unmap-does-not-flush, not been flushed before being
    unmapped>> (internal issue 819).
  * Fix description of code:WorkgroupSize builtin to note it decorates an
    object, not a variable (internal issue 836).
  * Fix asciidoc attributes so that trailing '{plus}' symbols in [eq] style
    equations are rendered properly (internal issue 845).
  * Add language to the "`Extension Handles, Objects, Enums, and Typedefs`"
    section of the Procedures and Conventions document stating that any new
    handle type requires a corresponding entry in the elink:VkObjectType
    enumerated type (internal issue 856).
  * Update style guide to use slink macro for Vulkan handle type names, and
    define narrow conditions under which to use the *name and *text macros
    instead of *link (internal issue 886).
  * Add a dependency of the <<VK_KHX_device_group,VK_KHX_device_group>>
    extension on VK_KHX_device_group_creation to +vk.xml+ and the extension
  * Change the copyright on Vulkan specification asciidoc *source* files to
    CC-BY 4.0, and update the proprietary Khronos copyright applied to the
    generated *output* formats (internal issue 327). This enables broader
    re-use and modification of the Vulkan specification sources, while not
    affecting the Reciprocal IP License between Vulkan Adopters and Working
    Group Members.

New Extensions:

  * `VK_NV_fill_rectangle`
  * `VK_NV_fragment_coverage_to_color`


This driver offers miscellaneous code optimizations that improves 3rd Generation Graphics.

NOTE: Microsoft Windows® 10 drivers for 3rd Generation Intel® Core™ Processors with Intel® HD Graphics 4000 and 3rd Generation Intel® Core™ Processors with Intel® HD Graphics 2500 will include support for Windows Display Driver Model (WDDM) 1.3. For reference, WDDM 1.3 features are available on Microsoft’s Web site. This driver will not be digitally signed for Windows® 10, which means that it will not have been certified by Microsoft Windows Hardware Quality Labs (WHQL). Note that Windows® 10 may alert you with a message if a driver is not digitally signed, but you can continue the installation. The driver is being provided as-is with no Intel support for any graphics issues found. Intel advises upgrading to Intel approved Windows 10 products for the best user experience and support.

The graphics driver has been posted to Intel Download Center at the following direct link to the driver:

3D-Tech News Around The Web / NVIDIA Vulkan beta driver 382.68
« on: June 28, 2017, 06:55:07 PM »
Developer Beta Driver Downloads Windows driver version 382.68 and Linux driver version 381.10.10 provide new features for Vulkan developers to test their upcoming Vulkan applications.
Windows 382.68
  Linux 381.10.10
 Release Updates June 27th, 2017 - Windows 382.68, Linux 381.10.10


Historically, improvements in GPU-based high performance computing have been tightly coupled to transistor scaling. As Moore's law slows down, and the number of transistors per die no longer grows at historical rates, the performance curve of single monolithic GPUs will ultimately plateau. However, the need for higher performing GPUs continues to exist in many domains. To address this need, in this paper we demonstrate that package-level integration of multiple GPU modules to build larger logical GPUs can enable continuous performance scaling beyond Moore's law. Specifically, we propose partitioning GPUs into easily manufacturable basic GPU Modules (GPMs), and integrating them on package using high bandwidth and power efficient signaling technologies. We lay out the details and evaluate the feasibility of a basic Multi-Chip-Module GPU (MCM-GPU) design. We then propose three architectural optimizations that significantly improve GPM data locality and minimize the sensitivity on inter-GPM bandwidth. Our evaluation shows that the optimized MCM-GPU achieves 22.8% speedup and 5x inter-GPM bandwidth reduction when compared to the basic MCM-GPU architecture. Most importantly, the optimized MCM-GPU design is 45.5% faster than the largest implementable monolithic GPU, and performs within 10% of a hypothetical (and unbuildable) monolithic GPU. Lastly we show that our optimized MCM-GPU is 26.8% faster than an equally equipped Multi-GPU system with the same total number of SMs and DRAM bandwidth.


3D-Tech News Around The Web / MSI Kombustor v3.5.2.1
« on: June 26, 2017, 07:29:45 PM »
This is standalone benchmark/stability test tool based on software from the maker of Furmark, this is MSI software that can be combined with MSI AfterBurner.

For the MSI AfterBurner we are creating a new function, a stress test which will be called Kombustor. This is standalone benchmark/stability test tool based on FurMark, but now with big and furry MSI logo. You can either run in separately, or link with Afterburner to read more detailed information during stress test.
After installing Kombustor, you can simply press the "K" button in Afterburner and instantly launch 3D stability test. Just that simple! You can also click the Kombustor icon on the desktop to launch GUI to customize your testing preference.

Features of MSI Kombustor
  • Based on FurMark technology
  • Latest OpenGL 3.2 rendering path
  • Benchmark and stability test mode
  • Link with Afterburner 1.5.0+
  • GPU temperature monitoring
  • Fullscreen or windowed for each type of run mode
  • Window size selection (standard or custom)
  • MSAA samples selection
  • Benchmarking parameters: time based or frames based
  • Xtreme Burning Mode for Xtreme tester

3D-Tech News Around The Web / AMD Vulkan Memory Allocator
« on: June 21, 2017, 11:09:59 PM »
Vulkan Memory Allocator library and accompanying sample application.

Problem Statement

Memory allocation and resource (buffer and image) creation in Vulkan is difficult (comparing to older graphics API-s, like D3D11 or OpenGL) for several reasons:
  • It requires a lot of boilerplate code, just like everything else in Vulkan, because it is a low-level and high-performance API.
  • There is additional level of indirection: VkDeviceMemory is allocated separately from creating VkBuffer/VkImage and they must be bound together. The binding cannot be changed later - resource must be recreated.
  • Driver must be queried for supported memory heaps and memory types. Different IHV-s provide different types of it.
  • Resources that don't fit in VRAM are not automatically evicted to RAM. Developer must handle out-of-memory errors on his own.
  • It is recommended practice to allocate bigger chunks of memory and assign parts of them to particular resources.

This library is helps game developers to manage memory allocations and resource creation by offering some higher-level functions. Features of the library could be divided into several layers, low level to high level:
  • Functions that help to choose correct and optimal memory type based on intended usage of the memory.
    • Required or preferred traits of the memory are expressed using higher-level description comparing to Vulkan flags.
  • Functions that allocate memory blocks, reserve and return parts of them (VkDeviceMemory + offset + size) to the user.
    • Library keeps track of allocated memory blocks, used and unused ranges inside them, finds best matching unused ranges for new allocations, takes all the rules of alignment into consideration.
  • Functions that can create an image/buffer, allocate memory for it and bind them together - all in one call.

A hardware upgrade in the final quarter of 2016 saw “Piz Daint”, Europe’s most powerful supercomputer, more than triple its computing performance.
ETH Zurich invested around CHF 40 million in the upgrade, so that simulations, data analysis and visualisation can be performed more efficiently than ever before.


3D-Tech News Around The Web / Blender 2.79 release in progress
« on: June 19, 2017, 05:34:50 PM »
Blender 2.79 release in progress

Release is expected in July.
Get the test build!

Important forward-compatibility warning: Any .blend file using new ID Pointer custom properties saved in Blender 2.79 and beyond will crash when opened with Blender 2.78c and any previous releases (follow link above for more details). This does not affect any existing custom property type, only a new type introduced in 2.79.

 Testbuild available!

3D-Tech News Around The Web / Vulkan SDK 1.0.51 Released
« on: June 17, 2017, 06:04:10 AM »
SDK Version now available

 Release date: June 15, 2017

This new Vulkan SDK release includes the following updates:
  • Improvements to the Loader and documentation to support Vulkan version
  • Inclusion of Visual Studio 2017 templates
  • Introduction of a new extension: VK_AMD_texture_gather_bias_lod (added to the Vulkan 1.0.50 specification)
  • Vulkan Layers changes:
    • The remaining functionality in the VK_LAYER_LUNARG_swapchain layer has been migrated into the VK_LAYER_LUNARG_core_validation layer
    • The VK_LAYER_LUNARG_api_dump utility layer has been improved with the addition of an HTML back-end for a better api_dump user experience
  • VkTrace/VkReplay additions:
    • vktraceviewer is included in both the Windows and Linux SDKs as an alpha release
    • vktrace trim feature is now available and has been improved to allow specification of a frame interval to capture - all other frames are removed from the trace file, resulting in a smaller trace file
    • vktrace has been improved to increase portability of generated trace files across display driver versions
  • Valid usage identification (ID) number changes to accomodate the official valid usage ID management process.  If you are currently using these IDs to filter messages or for debugging purposes, they will need to be remapped
  • Scores of bugfixes, validation coverage and accuracy improvements, and many feature additions
Read the Release notes for more detail.

3D-Tech News Around The Web / NVIDIA Aftermath 1.2
« on: June 15, 2017, 10:37:16 PM »
Aftermath is a compact, C++ library aimed at Microsoft Windows based developers, enabling post-mortem GPU crash analysis on NVIDIA GeForce based GPUs.

Code: [Select]
    // Aftermath not enabled in the driver.  This is
    //  generally dealt with via application profile,
    //  but if installed driver version is greater than
    //  382.53, this issue can be resolved by renaming
    //  the application to: "NvAftermath-Enable.exe".

3D-Tech News Around The Web / AMD Crimson ReLive Edition 17.6.2
« on: June 15, 2017, 03:53:56 PM »
Radeon Software Crimson ReLive Edition 17.6.2 Highlights

Fixed Issues
    • Performance in DiRT4™ when using 8xMSAA is lower than expected with the latest game build.
Known Issues
    • Graphical corruption may be experienced in Tom Clancy's™: Rainbow Six Siege when MSAA is enabled.
    • Adobe™ Lightroom may experience an application crash with GPU acceleration enabled on Windows® 7 system configurations using Radeon RX 390 Series graphics products.
    • A small amount of apps may experience issues with Borderless Fullscreen mode and AMD FreeSync™ technology if other applications or game launchers are running on the primary screen in the background.
    • Counter-Strike™: Global Offensive and World of Warcraft™ may experience flickering or performance issues the first time the game is launched on a system boot with AMD FreeSync™ technology enabled. Workarounds include exiting and restarting the application or task switching (alt+tab) in and out of the game to fix the issue.

3D-Tech News Around The Web / NVIDIA Flow 1.0.1
« on: June 14, 2017, 06:26:49 PM »
Flow is a sparse grid-based fluid simulation library for real-time applications.


- Moving grid support.
- Single pass advection.
- Vorticity confinement masks.


- D3D12 support.
- MultiGPU, allowing DX12 Flow simulation and inter-GPU tranfer, in both DX11 and DX12 applications.
- Multiple material support, allowing multiple active layers in the same grid.
- Self shadowing.
- Custom emitter APIs.
- Grid data access APIs.


- Initial Release

Pages: 1 [2] 3 4 ... 199