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 - JeGX

Pages: 1 [2] 3 4 ... 58
21
3D-Tech News Around The Web / AMD Ryzen 5 Processor Series Launched
« on: April 11, 2017, 04:08:22 PM »
AMD Ryzen5:  6-core (12-thread) and 4-core (8-thread) processors launched.

Post: http://www.geeks3d.com/20170411/amd-ryzen-5-processor-series-launched/


22
Quote
  * Bump API patch number and header version number to 47 for this update.

Github Issues:

  * Allow <<synchronization-pipeline-barriers-subpass-self-dependencies,
    self-dependencies>> (also described for slink:VkSubpassDependency) to
    have earlier stages depend on later stages if all stages are
    framebuffer-space (public issue 125).
  * Clarify when pipeline state structures are ignored in the
    slink:VkGraphicsPipelineCreateInfo strucure, when the tessellation
    structure must be valid, and remove 'if `NULL`' descriptions from the
    valid usage statements (public issue 445).
  * Remove the obsolete "validextensionstructs" attribute for
    flink:VkPresentRegionsKHR. This caused a pname:pNext valid usage
    statement to be generated which wasn't consistent with what is stated in
    the spec (public issue 481).

Internal Issues:

  * Clarify facingness of non-polygon fragments for slink:VkStencilOpState
    and in the code:FrontFacing <<interfaces-builtin-variables,built-in
    variable description>>. Define 'facingness' of a fragment as a distinct
    term from facingness of a polygon (internal issue 662).
  * Clarify that the texture compression features (e.g.
    pname:textureCompressionBC) means that all formats of that type
    (<<features-features-textureCompressionASTC_LDR,ASTC>>,
    <<features-features-textureCompressionETC2,ETC2>>,
    <<features-features-textureCompressionBC,BC>>) are supported, and that
    support for individual formats may: queried separately (internal issue
    663).
  * Clarify in the valid usage for slink:VkBindImageMemoryInfoKHX that each
    SFR rectangle must be a multiple of the sparse block size for each
    aspect, e.g. in a depth/stencil image using separate depth/stencil
    planes (internal issue 721).
  * Re-remove KHX variants of KHR structure types after promotion (internal
    issue 762).

source: https://github.com/KhronosGroup/Vulkan-Docs/commit/0285e9c97749eb0db63843b76c6202e14996b5aa

23
NVIDIA has announced the Tesla P100 (Pascal architecture), the first compute card with 12GB or 16GB of HBM2 stacked memory. AMD has been the first GPU maker to release a card with HBM memory (Radeon Fury) but NVIDIA is the first to release a card with HBM2.

Quote
Artificial intelligence for self-driving cars. Predicting our climate's future. A new drug to treat cancer. Some of the world's most important challenges need to be solved today, but require tremendous amounts of computing to become reality. Today's data centers rely on many interconnected commodity compute nodes, limiting the performance needed to drive important High Performance Computing (HPC) and hyperscale workloads.

NVIDIA® Tesla® P100 GPU accelerators are the most advanced ever built for the data center. They tap into the new NVIDIA Pascal™ GPU architecture to deliver the world's fastest compute node with higher performance than hundreds of slower commodity nodes. Higher performance with fewer, lightning-fast nodes enables data centers to dramatically increase throughput while also saving money.

Links:
- Tesla P100 @ NVIDIA
- techpowerup.com
- overclock3d.net






24
3D-Tech News Around The Web / Let's fix OpenGL
« on: April 10, 2017, 05:38:51 PM »
Quote
From windowing systems to virtual reality, real-time graphics code is ubiquitous. Programming
models for constructing graphics software, however, have largely escaped the attention of pro-
gramming languages researchers. This essay introduces the programming model of OpenGL, a
ubiquitous API for real-time graphics applications, for a language-oriented audience. It high-
lights six broad problems with the programming model and connects them to traditions in PL
research. The issues range from classic pitfalls, where established thinking can apply, to new
open problems, where novel research is needed.

Link: http://www.cs.cornell.edu/~asampson/media/papers/opengl-snapl2017-preprint.pdf

26
Geeks3D's GPU Tools / VK-Z 0.6.0 released
« on: April 04, 2017, 06:15:19 PM »
A new version of VK-Z is available. VK-Z is a simple command line utility, for Windows and Linux that displays Vulkan capabilities (hardware limits) and extensions of all capable Vulkan devices on the system.

Complete story and downloads: http://www.geeks3d.com/20170404/vk-z-0-6-0-vulkan-capabilities-and-extensions-viewer-released/



27
GeeXLab - english forum / VR programming with GeeXLab
« on: April 04, 2017, 06:10:48 PM »
Here is an introduction to VR (Virtual Reality) programming with GeeXLab:

http://www.geeks3d.com/hacklab/20170331/introduction-to-vr-programming-with-geexlab/


28
GeeXLab - english forum / GeeXLab 0.14.1 released
« on: April 04, 2017, 06:08:45 PM »

29
Quote
* Bump API patch number and header version number to 46 for this update.

Github Issues:

  * Add language to the <<fundamentals-validusage-enums, Valid Usage for
    Enumerated Types>> section allowing values to be returned from Vulkan
    that are not present in extensions explicitly enabled by the
    application, similar to existing language for bit flags in the
    <<fundamentals-validusage-flags, Valid Usage for Flags>> section (public
    issue 442).
  * *Important*: run `gem update --pre asciidoctor-pdf` before trying to
    build this version of the spec - 1.5.0.alpha15 is required for this
    change. Removes the monkey patch currently used to draw valid usage
    blocks across multiple pages which had numerous issues. A fixed version
    was incorporated into Asciidoctor-PDF for the latest release, so the
    monkey patch or any variant thereof is no longer required (public issue
    465).

Internal Issues:

  * Add ename:VK_DEBUG_REPORT_OBJECT_TYPE_DESCRIPTOR_UPDATE_TEMPLATE_KHR_EXT
    to `VK_EXT_debug_report` extension
  * Fix ptext:pNext member of
    slink:VkPhysicalDeviceDiscardRectanglePropertiesEXT to be a non-const
    pointer. Properties structures return values, so the chain should be
    non-const.
  * Explicitly remove gl_NumSamples from the `GL_KHR_vulkan_glsl` extension,
    against 1.0 (internal issue 612).
  * Add Valid Usage statements requiring that each structure type valid in a
    ptext:pNext chain must: not appear more than once in a chain (internal
    issue 752).
  * Use ename:VK_USE_PLATFORM_WIN32_KHX in the
    `VK_KHX_external_memory_win32` extension, rather than etext:_KHR
    (internal issue 754).

New Extensions:

  * `VK_KHR_incremental_present`

source: https://github.com/KhronosGroup/Vulkan-Docs/commit/f985a50c1f0b6f37c706db32d70da85bc1e2d78f

30
Geeks3D's GPU Tools / GPU Caps Viewer 1.34.1.x released
« on: April 01, 2017, 04:33:33 PM »
I think I fixed the bug with Radeon GPUs. Try the new version 1.34.1.0:

http://www.geeks3d.com/20170327/gpu-caps-viewer-1-34-0-released/#20170401

On my notebook (a Lenovo ThinkPad E460 with an Intel HD 520 + Radeom R7 M360), the R7 M360 is now properly detected by GPU Caps.

31
Geeks3D's GPU Tools / Re: GPU Caps Viewer 1.34.x.x released
« on: April 01, 2017, 10:48:46 AM »
Update log file of AMD WX 7100
=================================================================================================
01:16:54 | 26 >> - DEV 0 - DeviceID: PCI\VEN_8086&DEV_193D&SUBSYS_07B11028&REV_09
01:16:54 | 27 >> - DEV 0 - DeviceName: \\.\DISPLAY1
01:16:54 | 28 >> - DEV 0 - DeviceString: Intel(R) Iris(TM) Pro Graphics P580
01:16:54 | 29 >> - DEV 0 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{674CA461-FAB1-44C5-BB14-5ED2DFBA2C37}\0000
01:16:54 | 30 >> - DEV 1 - DeviceID: PCI\VEN_8086&DEV_193D&SUBSYS_07B11028&REV_09
01:16:54 | 31 >> - DEV 1 - DeviceName: \\.\DISPLAY2
01:16:54 | 32 >> - DEV 1 - DeviceString: Intel(R) Iris(TM) Pro Graphics P580
01:16:54 | 33 >> - DEV 1 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{674CA461-FAB1-44C5-BB14-5ED2DFBA2C37}\0001
01:16:54 | 34 >> - DEV 2 - DeviceID: PCI\VEN_8086&DEV_193D&SUBSYS_07B11028&REV_09
01:16:54 | 35 >> - DEV 2 - DeviceName: \\.\DISPLAY3
01:16:54 | 36 >> - DEV 2 - DeviceString: Intel(R) Iris(TM) Pro Graphics P580
01:16:54 | 37 >> - DEV 2 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{674CA461-FAB1-44C5-BB14-5ED2DFBA2C37}\0002
01:16:54 | 38 >> - DEV 3 - DeviceID: PCI\VEN_8086&DEV_193D&SUBSYS_07B11028&REV_09
01:16:54 | 39 >> - DEV 3 - DeviceName: \\.\DISPLAY4
01:16:54 | 40 >> - DEV 3 - DeviceString: Intel(R) Iris(TM) Pro Graphics P580
01:16:54 | 41 >> - DEV 3 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{674CA461-FAB1-44C5-BB14-5ED2DFBA2C37}\0003
01:16:54 | 42 >> - DEV 4 - DeviceID: PCI\VEN_8086&DEV_193D&SUBSYS_07B11028&REV_09
01:16:54 | 43 >> - DEV 4 - DeviceName: \\.\DISPLAY5
01:16:54 | 44 >> - DEV 4 - DeviceString: Intel(R) Iris(TM) Pro Graphics P580
01:16:54 | 45 >> - DEV 4 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{674CA461-FAB1-44C5-BB14-5ED2DFBA2C37}\0004
01:16:54 | 46 >> - DEV 5 - DeviceID: PCI\VEN_8086&DEV_193D&SUBSYS_07B11028&REV_09
01:16:54 | 47 >> - DEV 5 - DeviceName: \\.\DISPLAY6
01:16:54 | 48 >> - DEV 5 - DeviceString: Intel(R) Iris(TM) Pro Graphics P580
01:16:54 | 49 >> - DEV 5 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{674CA461-FAB1-44C5-BB14-5ED2DFBA2C37}\0005
01:16:54 | 50 >> - DEV 6 - DeviceID: PCI\VEN_8086&DEV_193D&SUBSYS_07B11028&REV_09
01:16:54 | 51 >> - DEV 6 - DeviceName: \\.\DISPLAY7
01:16:54 | 52 >> - DEV 6 - DeviceString: Intel(R) Iris(TM) Pro Graphics P580
01:16:54 | 53 >> - DEV 6 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{674CA461-FAB1-44C5-BB14-5ED2DFBA2C37}\0006
01:16:54 | 54 >> - DEV 7 - DeviceID:
01:16:54 | 55 >> - DEV 7 - DeviceName: \\.\DISPLAYV1
01:16:54 | 56 >> - DEV 7 - DeviceString: RDPDD Chained DD
01:16:54 | 57 >> - DEV 7 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{DEB039CC-B704-4F53-B43E-9DD4432FA2E9}\0000
01:16:54 | 58 >> - DEV 8 - DeviceID:
01:16:54 | 59 >> - DEV 8 - DeviceName: \\.\DISPLAYV2
01:16:54 | 60 >> - DEV 8 - DeviceString: RDP Encoder Mirror Driver
01:16:54 | 61 >> - DEV 8 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{42cf9257-1d96-4c9d-87f3-0d8e74595f78}\0000
01:16:54 | 62 >> - DEV 9 - DeviceID:
01:16:54 | 63 >> - DEV 9 - DeviceName: \\.\DISPLAYV3
01:16:54 | 64 >> - DEV 9 - DeviceString: RDP Reflector Display Driver
01:16:54 | 65 >> - DEV 9 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{b043b95c-5670-4f10-b934-8ed0c8eb59a8}\0000

Thanks for the log. That's what I thought, the AMD GPU is not listed in the active Windows adapters. Actually I have the same problem with a notebook that has an Intel GPU + AMD GPU: GPU Caps Viewer shows the Intel GPU only. I'm investigating about this issue... I let you know asap.

32
Geeks3D's GPU Tools / Re: GPU Caps Viewer 1.34.x.x released
« on: March 30, 2017, 03:59:00 PM »
Actually what I need is the log file of GPU Caps Viewer (_log.txt in GPU Caps folder). I need to know if the  WX 7100 is listed or not in Windows devices. On my system for example, the GTX 1060 is listed like that:
Quote
15:52:49 | 28 >> - DEV 0 - DeviceID: PCI\VEN_10DE&DEV_1C03&SUBSYS_61633842&REV_A1
15:52:49 | 29 >> - DEV 0 - DeviceName: \\.\DISPLAY1
15:52:49 | 30 >> - DEV 0 - DeviceString: NVIDIA GeForce GTX 1060 6GB
15:52:49 | 31 >> - DEV 0 - DeviceKey: \Registry\Machine\System\CurrentControlSet\Control\Video\{3FDB2FF5-A64B-4CA6-9A2B-0D133FF04B78}\0000

33
Just added to the code sample pack this cool multipass demo. I will try to upload the new code sample pack today. The demo can be found in the gl-21/shadertoy-multipass/gl21-alien-corridor/ folder.




34
DepthOfFieldFX is a DirectX 11 library that provides a GCN-optimized Compute Shader implementation of Depth of Field.

Quote
The DepthOfFieldFX library provides a GCN-optimized Compute Shader implementation of Depth of Field using the Fast Filter Spreading approach put forth by Todd Jerome Kosloff, Justin Hensley, and Brian A. Barsky in their 2009 paper – Fast Filter Spreading and its Applications.

The library implements depth of field using a Bartlett and Box filter. There is an option to run the Bartlett at quarter resolution for improved performance in some cases. The advantage of this approach is a fixed performance cost that is not dependent on kernel size. The library takes the scene Color Buffer and Circle of Confusion buffer as input and an Unordered Access View for the final results.

Links:
- https://github.com/GPUOpen-Effects/DepthOfFieldFX



35
You might have read that Microsoft will stop supporting Windows Vista on April 11. Futuremark will stop supporting two Windows Vista benchmarks, 3DMark Vantage and PCMark Vantage, on the same day.

Quote

3DMark Vantage and PCMark Vantage support ends April 11
Our Vantage benchmarks were designed for Windows Vista, an operating system that is over 10 years old and which is now used by only 1.12% of Windows users according to StatCounter.
 
From April 11, 3DMark Vantage and PCMark Vantage will:
- No longer be sold on our website.
- No longer receive updates, (other than SystemInfo).
- No longer be guaranteed to work with our online services.
- No longer be eligible for customer support.

Alternatives to 3DMark Vantage
3DMark Vantage is a benchmark for DirectX 10, an API which is now two generations old. For benchmarking modern gaming systems, we recommend using the latest 3DMark benchmarks:
- 3DMark Time Spy for DirectX 12 systems.
- 3DMark Fire Strike for high performance DirectX 11 PCs.
- 3DMark Sky Diver for DirectX 11 laptops and mid-range PCs.

Alternatives to PCMark Vantage
PCMark Vantage is a system benchmark for Windows Vista. If your PC is less than seven years old, you will get more relevant results by using a more recent version of PCMark. We recommend:
- PCMark 7 for Windows 7
- PCMark 8 for Windows 8
- PCMark 10, coming soon, for Windows 10.

Legacy benchmarks are free to download
All of our legacy benchmarks going back to 3DMark99 are available to download for free from our website. Each benchmark comes with a key to unlock the Advanced Edition features. 3DMark Vantage and PCMark Vantage will be available for free from April 11.

https://www.futuremark.com/benchmarks/legacy


37
Geeks3D's GPU Tools / GPU Shark 0.9.11.1 released
« on: March 27, 2017, 01:41:52 PM »

38
Geeks3D's GPU Tools / GPU Caps Viewer 1.34.x.x released
« on: March 27, 2017, 01:40:55 PM »
The new version of GPU Caps Viewer is available.

More information and dowmload links:
http://www.geeks3d.com/20170327/gpu-caps-viewer-1-34-0-released/

39
Download zone updated: HWMonitor 1.31

40
3D-Tech News Around The Web / AMD Compressonator v2.5 released
« on: March 24, 2017, 06:14:29 PM »
Compressonator is a set of tools to allow artists and developers to more easily create compressed texture image assets and easily visualize the quality impact of various compression technologies. It consists of a GUI application, a command line application and an SDK for easy integration into a developer tool chain.

The v2.5.3297 beta version has been released few days ago.

- Compressonator @ github
- changelog

Pages: 1 [2] 3 4 ... 58