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 ... 3 4 [5] 6 7 ... 50
English forum / GeeXLab 0.12.0 released
« on: May 31, 2016, 11:42:30 AM »
GeeXLab has been released for Windows 32 and 64-bit.

Complete story:


I should update a bit the support of Intel GPUs in GPU Caps Viewer one day...

Here are some specifications/speculations from wccftech about the real Pascal GPU for gamers:

- GeForce GTX 1080 Ti
- GPU: GP102
- cores: 3840 (60SMs)
- memory size: 16GB of GDDR5X
- TDP: 250W


English forum / Shadertoy demos in Vulkan
« on: May 25, 2016, 04:07:16 PM »
For testing how Vulkan plugin works (especially the GLSL to SPIRV converter tool) with more complex shaders, I ported this shadertoy demo to GeeXLab / Vulkan.

No particular surprise, everything worked as expected.

This demo will be available in the next update of the code sample pack.

According to this comparative table, the upcoming GTX 1070 with its 1920 CUDA cores is faster (ok by a very small margin) than the GTX Titan X that packs 3072 CUDA cores.


Rok Erjavec, Technical Director at Crytek, in an interview with DSOGaming:

Q: DSOGaming: Does CRYENGINE support Vulkan and have you experimented with it? If you had to choose one, would you go with DX12 or Vulkan?

A: Rok Erjavec: The current trajectory of Vulkan provides a path to reach PC users across all the widely used OS platforms, including legacy Windows versions, as well as mobile devices, so if I was shipping a title in 2017 or beyond, Vulkan looks like an appealing choice.

Full interview:

Linux 4.5 contains over 21 million lines of code. Around 4,000 developers and at least 440 different companies contribute to the kernel. There are over 10,800 lines of code added, 5,300 lines of code removed and over 1,875 lines of code modified. Every. Single. Day. That amounts to over 8 changes per second.


FireRays is ray intersection acceleration library provided by AMD which makes the most of AMD hardware and allows for efficient ray queries independently of the generation of underlying AMD hardware.

After the success of the first version, FireRays is moving to another major milestone. We are open sourcing the entire library which allows complete integration and contribution from the entire developer community to bring improvements in the rendering world. In fact FireRays 2.0 brings support for Windows, OSX, Linux, AMD, NV, Intel GPUs and CPUs with many back-ends.


3D-Tech News Around The Web / AMD Compressonator is now Open Source
« on: May 13, 2016, 09:43:31 AM »
Compressonator is a set of tools for compressing textures (ASTC, ATC, ATInN, BCn, ETCn, DXTn and swizzled DXTn formats) and creating mip-map levels.

Compressonator is a set of tools to allow artists and developers to more easily work with compressed 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 toolchain.

We are very pleased to be announcing that AMD is open-sourcing one of our most popular tools and SDKs.  Compressonator (previously released as AMD Compress on our Developer Central site ) is now available on GPUOpen.

You will get access to:

- GUI Tool source
- Command line Tool source
- Compress SDK source
- All Common Code used for the SDK, GUI and Command line.


English forum / How to Run a Vulkan Demo on a Particular GPU
« on: May 12, 2016, 01:19:44 PM »
Here is an article that explains how to run a Vulkan demo on particular GPU:

English forum / GeeXLab 0.11.0.x released
« on: May 12, 2016, 01:17:33 PM »

MUNICH, Germany. Officials at the Khronos Group consortium announced the availability of the OpenGL SC 2.0 specification for bringing programmable graphics to systems that require system safety certification such as commercial and military avionics systems as well as automotive safety systems. Khronos officials made the announcement at the Aviation Electronics Europe show held last week in Munich.

OpenGL SC 2.0 is a subset of OpenGL ES 2.0 that includes GLSL-based programmable shaders to enable enhanced graphics functionality, with increased performance and reduced power. It removes all debug functionality from OpenGL ES 2.0, but integrates the OpenGL robustness extension into the core specification for scheduling and memory access integrity. OpenGL SC 2.0 is designed to be both deterministic and testable while keeping compatibility with existing OpenGL ES 2.0-capable silicon, enabling the immediate deployment of high-volume desktop, mobile, and embedded silicon solutions.

Complete story:

* Bump API patch number and header version number to 12 for this

Github Issues:

  * Change valid usage statements intended to be "sub-points" to
    be actual sub-points (public issue 66).
  * Replace double negation in description of
    slink:VkRenderPassBeginInfo::pname:pClearValues (based on public
    merge 142).
  * Cleanup minor typos in spec, ref pages and XML, including those
    proposed in public pull requests 144, 150, 151, 167, 168, 181, and
  * Use *strict subset* in describing the partial order of memory
    property types for slink:VkMemoryType, and update the style guide
    accordingly (public issue 190).
  * Fix various "a image" -> "an image" typos (public issue 191).
  * Note in the <<fundamentals-validusage,Valid Usage>> and
    <<extensions-interactions,Extension Interactions>> sections that
    structures defined by extensions which may be passed in structure
    chains using the ptext:pNext member must: include initial
    ptext:sType and ptext:pNext members (public issue 192).

Internal Issues:

  * Remove duplicate language from the description of the pname:fence
    parameter to flink:vkQueueSubmit and improve validity language
    (internal issue 91).
  * Added documentation for "optional" attribute to XML readme.tex/pdf
    (internal issue 149).
  * Clarify the host-side data validity rules and behavior of
    flink:vkFlushMappedMemoryRanges and
    flink:vkInvalidateMappedMemoryRanges (internal issue 266).

Other Commits:

  * Added clarification to flink:vkCmdFillBuffer regarding the use of
  * Fixed and documented implementation of "validextensionstructs"
    attribute. in XML processing scripts and readme.tex/pdf.
  * Add missing validity statements to flink:vkResetEvent and
  * Fix validity for the
    Correct all the draw/dispatch commands to mention optimally tiled
    images as well as linear tiled images, and say image VIEWS instead
    of images. Add validity statement to flink:vkCmdBlitImage
  * Replace the {apiname} macro with hardcoded "Vulkan", now that we've
    committed to that name.
  * Add the VK_AMD_rasterization_order extension to vk.xml.


3D-Tech News Around The Web / NVIDIA R364.19 for Linux
« on: April 22, 2016, 05:27:33 PM »
Release highlights since 364.15:
Added support for the following GPU:
Tesla M40 24GB

Fixed some locking issues in
Fixed several interaction problems with buffer management Wayland protocols.

- Linux 64-bit
- Linux 32-bit
- Linux ARM
- Solaris
- FreeBSD 32-bit
- FreeBSD 64-bit

English forum / GeeXLab released
« on: April 21, 2016, 04:39:50 PM »
GeeXLab is available for Windows 64-bit only. This is a bugfix release. More information can be found here:

The changelog is available here:

English forum / Re: ANY HLSL webcam effects that are MIT licensed?
« on: April 21, 2016, 03:28:33 PM »
If the shader posted on geeks3d (shaderlib or in the posts) is a simple copy with minor modifications of another shader, you have to check the license of the original shader (there is always a link to the original shader). If the shader author is Geeks3D (there is no link to a source, except if I forgot it...), yes you can use it as you want, I don't care.  If you have a doubt, do not hesitate to use the forum.

English forum / Re: GeeXLab command line parameters?
« on: April 19, 2016, 07:46:23 AM »
I forgot to check command line parameters in GeeXLab but this is done for v0.10.1.1.
The /benchmark_duration_ms parameter is fixed (as well as the display of the score box).

English forum / Re: Fastest way to upload vertices?
« on: April 14, 2016, 05:27:32 PM »
A code sample that shows how to initialize a vertex pool with a Python array is available in the new code sample pack:


I tested with 1'000'000 particles and initialization from Python arrays is really fast!

English forum / GeeXLab released
« on: April 14, 2016, 05:20:49 PM »

Pages: 1 ... 3 4 [5] 6 7 ... 50