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 ... 146
1
3D-Tech News Around The Web / NVIDIA GeForce GTX 980 Whitepaper
« on: September 20, 2014, 01:05:09 AM »
download from here

Quote
In our 21-year quest to bring the most realistic 3D graphics to gamers, NVIDIA has introduced a number
of innovations. With its hardware transform and lighting engine, NVIDIA’s GeForce 256 ushered in the
era of the GPU in 1999, bringing T&L support to consumer graphics for the first time. In late 2006 with
the G80 GPU and the GeForce 8800 GTX graphics board, CUDA and the formalization of GPGPU
computing as we know it today was first brought to the world. More recently, in 2010 we launched the
GeForce GTX 480—our first GPU built using the Fermi graphics architecture. GeForce GTX 480
incorporated fifteen parallel tessellation units, enabling dramatic speedups in geometry processing
compared to prior GPUs. With great strides in architectural efficiency, Kepler GPUs delivered
significantly improved performance and power efficiency when introduced in 2012. 
Earlier this year we launched the GeForce GTX 750 Ti, our first GPU from the Maxwell architectural
family. Maxwell GPUs were designed from the ground up for extreme power efficiency and exceptional
performance per watt consumed. In many DX11 applications, the GTX 750 Ti is capable of matching or
even beating the performance of our once flagship GeForce GTX 480, while consuming only a fourth the
power. Thanks to its remarkable power efficiency, our first generation Maxwell GPUs were ideal for use
in power-limited environments like notebooks and small form factor PCs, in addition to mainstream
desktops.
NVIDIA’s latest GPU, GM204, is the first to use the full realization of our 10th
 generation GPU
architecture, Maxwell. Our design goals for GM204 were to deliver:
  Extraordinary Gaming Performance for the Latest Displays
  Incredible Energy Efficiency
  Dramatic Leap Forward In Lighting With VXGI
Extraordinary Gaming Performance for the Latest Displays

2
3D-Tech News Around The Web / NVIDIA GeForce driver 344.11 WHQL
« on: September 19, 2014, 07:13:31 AM »
New in GeForce Game Ready Drivers

The new GeForce Game Ready driver, release 344.11 WHQL, allows GeForce owners to continue to have the ultimate gaming experience. This driver is aligned with today’s launch of the world’s most advanced GPUs—the GeForce GTX 980 and GTX 970. With support for NVIDIA G-SYNC Surround displays, gaming has never been more realistic and immersive. In addition, this Game Ready WHQL driver ensures you'll have the best possible gaming experience for the latest new blockbuster titles including Borderlands: The Pre-Sequel, The Evil Within, F1 2014, and Alien: Isolation.

Game Ready
Best gaming experience forBorderlands: The Pre-Sequel, The Evil Within, F1 2014, and Alien: Isolation

New GeForce GPUs
Supports the new GeForce GTX 980 and GTX 970 GPUs, based upon the second-generation Maxwell architecture

Gaming Technology
Supports G-SYNC technology and NVIDIA G-SYNC Surround configurations

Code: [Select]
DriverVer   = 09/13/2014, 9.18.13.4411
NVIDIA_DEV.13C0 = "NVIDIA GeForce GTX 980"
NVIDIA_DEV.13C2 = "NVIDIA GeForce GTX 970"

3
 
  • Added support for the following GPUs:
    • GeForce GTX 970
      GeForce GTX 980
  • Fixed a bug that prevented the "sync to vblank" setting from being honored for EGL applications.
  • Fixed a bug that could cause some OpenGL programs to encounter out of memory during a mode switch.
  • Fixed a bug that prevented the NVIDIA OpenGL driver from honoring the __GL_SHADER_DISK_CACHE_PATH environment variable.
  • Fixed a bug that caused disabled displays to be implicitly included in the target selection for some queries and assignments on the nvidia-settings command line interface, in the absence of any explicit target selection.
  • Added a new attribute to the NV-CONTROL API to query the current utilization of the video decode engine.
  • Fixed a bug where the Exchange Stereo Eyes setting in nvidia-settings didn't work in certain stereo configurations.
  • Worked around a Unigine Heaven 3.0 shader bug which could cause corruption when tessellation is enabled by implementing an application profile that uses the "GLIgnoreGLSLExtReqs" setting.  See the documentation for the __GL_IGNORE_GLSL_EXT_REQS environment variable for more details.
  • Fixed a memory leak when destroying EGL surfaces.
  • Added support for multiple simultaneous EGL displays.
  • Removed support for G8x, G9x, and GT2xx GPUs, and motherboard chipsets based on them. Ongoing support for new Linux kernels and X servers, as well as fixes for critical bugs, will be included in 340.* legacy releases through the end of 2019.
download from here


 

4
3D-Tech News Around The Web / NVIDIA mobile GeForce driver 344.00 (ASUS OEM)
« on: September 18, 2014, 06:16:17 PM »
   This driver adds support for

 
NVIDIA_DEV.13D7 = "NVIDIA GeForce GTX 980M"
 NVIDIA_DEV.13D8 = "NVIDIA GeForce GTX 970M"

 
Works fine with Pieter's INF 5.012
 
15092935629_97b0c07b34_o.png
     

5
3D-Tech News Around The Web / Blender 2.72 test
« on: September 13, 2014, 04:48:53 PM »
  OFFICIAL TEST BUILD AVAILABLE NOW

https://download.blender.org/release/Blender2.72/
 
   Cycles Rendering

Volume rendering and Subsurface Scattering are available on GPU, Volume and Glossy sampling was improved, new Ashikhmin-Shirley distribution for Glossy and Anisotropic BSDFs, memory usage during rendering was lowered, OSL updated to version 1.5.
 

6
We are happy to announce the availability of our latest and most advanced SDK for OpenCL: Release 2 of Intel® SDK for OpenCL SDK 2014 is the industry’s first SDK to provide an OpenCL 2.0 development environment with the new Intel® Core™ M Processors.
This major advance in graphics programmability and accessibility will help you make greater use of the graphics engine to deliver new experiences on Intel-based platforms
New with SDK 2014 Release 2:
 
  • OpenCL™ 2.0: Support for latest OpenCL standard version 2.0 with shared virtual memory. Start now and write your first OpenCL 2.0 code on your current development platform, simply by using the new OpenCL 2.0 development environment that is installed with the SDK.
  • Intel® Core™ M Processors: Take advantage of OpenCL 2.0 with the latest Intel® Core™ M processors and future generations of Intel Core processors.
  • SPIR* 1.2: Improve portability and drive innovation with a standard, non-source intermediate representation (IR) for device programs. Supported on both CPU and Intel Graphics.
  • Greater development experience: The SDK provides everything you need to build, debug, and analyze OpenCL application. This release not only adds OpenCL 2.0 and SPIR 1.2 development support, but also adds new preview features for debugging and analyzing applications
Learn more

7
Fluid Simulation for Video Games (part 18)

This article showed how to extract surfaces from a function represented by a grid as a polygon mesh that you can then render. Combined with previous articles in this series, that lets you visualize surfaces of simulated fluids. This article also showed how to parallelize that isosurface extraction routine. Although the routine’s outputs for a given input cell do not appear to depend on those of other cells, there is a subtle dependency on where to write the output. To solve this question, I wrapped the output vertex buffer with a lightweight, thread-safe, linear pool allocator, thereby avoiding race conditions between threads, keeping the vertex buffer usage relatively balanced across threads and the synchronization overhead manageable. You can tune the tradeoff between balancing thread usage of the buffer and synchronization overhead by changing the block size.

8
3D-Tech News Around The Web / AMD GPUPerfAPI 2.14
« on: September 10, 2014, 06:36:24 PM »
Release Notes - Version 2.14 (9/5/14) 
  • Add support for additional GPUs and APUs.
  • DirectX 11: Fixed some memory leaks caused by incorrect reference counting of D3D11 devices and device contexts.
  • OpenGL: Improved memory consumption and performance during profiling.
  • OpenGL: Fixed crash in 32-bit Linux version.
   
  • OpenGL: Fixed CSMemUnitBusy, CSMemUnitStalled, and TexTriFilteringPct counters on Graphics IP v8 family of GPUs.
  • OpenGL: Fixed CSALUStalledByLDS  counter on Graphics IP v7 and v8 families of GPUs.
  • Removed Support for DirectX 10.

9
3D-Tech News Around The Web / AMD GPU PerfStudio V3.0
« on: September 10, 2014, 06:35:13 PM »
GPU PerfStudio V3.0

 New Features
  • Adds Linux Server support
  • Adds OpenGL 4.3 Support
  • Calls between glBegin & glEnd are now collapsed to a single entry.
  • Adds support for separate shader objects in GL
Bug Fixes
  • Many stability enhancements for titles running on DirectX 11 and OpenGL
  • Fix for missing call data in FrameDebugger
  • Fix for missing HUD data in GL applications

10
3D-Tech News Around The Web / AMD APP SDK v2.9.1 Now Available
« on: September 04, 2014, 07:19:26 PM »
AMD APP SDK v2.9.1 Now Available!



The AMD APP SDK v2.9.1 includes samples for OpenCL™ as well as accelerated libraries such as the Open Source  C++ template library called “Bolt” and the latest OpenCL™ accelerated OpenCV (Open Computer Vision) library. APP SDK 2.9.1 now enables Linux users who need to install with non-root permission allowing users of IT managed and server systems to develop for the powerful compute capabilities of AMD APUs and GPU.  This release supports Catalyst 14.4 driver.  Read the Getting Started Guide for more details.  Please give it a try and provide us with feedback.

11
3D-Tech News Around The Web / AMD Radeon SDK samples August 2014
« on: September 04, 2014, 07:17:49 PM »
What’s new (August 2014) 
  • New SDK sample: GPUParticles
  • New SDK sample: SilhouetteTessellation
  • TressFX version 2.2 sample with new features including simulation performance improvements
   
  • ForwardPlus version 1.1: fixed bug causing visual differences when culling was enabled with a high number of lights
  • DepthBoundsTest version 1.1: fixed bug with enabling MSAA causing a black screen
   
  • All DirectX 11 SDK samples were updated to reference the Windows 8 or 8.1. software development kits
  • Legacy SDK samples moved to archive page
   


12
3D-Tech News Around The Web / HWiNFO32 & HWiNFO64 v4.44 released
« on: September 04, 2014, 06:26:34 PM »
Changes in HWiNFO32 & HWiNFO64 v4.44 - Released on:  Sep-4-2014: 
 
  • Enhanced sensor monitoring on additional MSI 9-series mainboards.
  • Fixed restoring of sensors window position on some systems with multiple monitors.
  • Fixed reporting of invalid values in graphs, tray, LCD, OSD.
  • Added support of Lewisburg PCH.
  • Added preliminary support of NVIDIA GeForce GTX 870 and 880 (GM204).
  • Sensor minimize button will minimize graphs too.
  • Added ability to rename sensor units.
  • Added ability to modify sensor values (constant multiply/add).
  • Added support of On Semi ASP1106D.
  • Fixed reporting of memory clock on Haswell-E/EN/EP/EX.
  • Improved DDR4 support.
  • Enhanced sensor monitoring on ASUS X99 series.
  • Enhanced sensor monitoring on MSI X99 series.
  • Patch for occasional invalid BCLK measurement.
  • Enhanced sensor monitoring on ASUS RAMPAGE V EXTREME.
  • Enhanced sensor monitoring on ASUS CROSSBLADE RANGER.
  • Added AMD Radeon R9 285.
  • Enhanced sensor monitoring on ASRock X99 series.

13
3D-Tech News Around The Web / OpenGL Extensions Viewer 4.27 for Windows
« on: August 27, 2014, 06:43:56 PM »
Download from here

No changelog available, but adds OpenGL 4.5 support




14
Today we’re excited to announce the release of the CUDA Toolkit version 6.5.

CUDA 6.5 adds a number of features and improvements to the CUDA platform, including support for CUDA Fortran in developer tools, user-defined callback functions in cuFFT, new occupancy calculator APIs, and more.

15
3D-Tech News Around The Web / TechPowerUp GPU-Z v0.7.9
« on: August 11, 2014, 06:37:41 PM »
GPU-Z is a lightweight utility designed to give you all information about your video card and GPU. Version History0.7.9
  • Added preliminary support for NVIDIA GM204
  • Added preliminary support for AMD Tonga
  • Added support for AMD Radeon R9 M275X, FirePro W5100, W9100
  • Added support NVIDIA GeForce GTX 780 6 GB, GTX 860M, GTX 780M, GT 830M, GT 740, GT 730, GT 720, Quadro NVS 510, FX 380M, GRID K520, Tesla K40c
  • Added release date for R9 290
  • More robust PhysX detection
  • Fixed fan speed monitoring on some recent AMD cards
  • Fix for sensor graph over/underflow
  • Performance improvements to sensor graph drawing
  • Fix for French translation

16
3D-Tech News Around The Web / (Demoscene) Evoke 2014, Assembly 2014
« on: August 10, 2014, 07:40:25 AM »
download demos via pouet:

Evoke 2014

Assembly 2014

17
3D-Tech News Around The Web / OpenGL Extensions Viewer 1.2.0 for Android
« on: August 10, 2014, 07:34:52 AM »
In this update

• Updated for OpenGL ES 3.1

• Added all the 'GL_MAX' values of the renderer for OpenGL ES 3.0 and 3.1

• Bug fixes and improvements.

18
3D-Tech News Around The Web / OpenGL Extensions Viewer 2.0.1 for iOS
« on: August 10, 2014, 07:31:17 AM »
  What's New in Version 2.0.1

- Fixed benchmark not working after several tries
- OpenGL ES 3.0 benchmark rewritten and optimized for OES 3.0
- OpenGL ES 3.0 capabilities reported :
        GL_MAX_3D_TEXTURE_SIZE
        GL_MAX_TEXTURE_LOD_BIAS
        GL_MAX_FRAGMENT_UNIFORM_COMPONENTS
        GL_MAX_VERTEX_UNIFORM_COMPONENTS
        GL_MAX_ARRAY_TEXTURE_LAYERS
        GL_MIN_PROGRAM_TEXEL_OFFSET
        GL_MAX_PROGRAM_TEXEL_OFFSET
        GL_MAX_VARYING_COMPONENTS
        GL_MAX_TRANSFORM_FEEDBACK_INTERLEAVED_COMPONENTS
        GL_MAX_TRANSFORM_FEEDBACK_SEPARATE_ATTRIBS
        GL_MAX_COLOR_ATTACHMENTS
        GL_MAX_SAMPLES
        GL_MAX_VERTEX_UNIFORM_BLOCKS
        GL_MAX_FRAGMENT_UNIFORM_BLOCKS
        GL_MAX_COMBINED_UNIFORM_BLOCKS
        GL_MAX_UNIFORM_BUFFER_BINDINGS
        GL_MAX_UNIFORM_BLOCK_SIZE
        GL_MAX_COMBINED_VERTEX_UNIFORM_COMPONENTS
        GL_MAX_COMBINED_FRAGMENT_UNIFORM_COMPONENTS     
        GL_MAX_VERTEX_OUTPUT_COMPONENTS
        GL_MAX_FRAGMENT_INPUT_COMPONENTS
        GL_NUM_PROGRAM_BINARY_FORMATS
        GL_MAX_ELEMENT_INDEX
        GL_NUM_SAMPLE_COUNTS
- UI during rendering test.
- Display a more complete device name.
 

19
3D-Tech News Around The Web / OpenGL Extensions Viewer 4.26 for Windows
« on: August 10, 2014, 07:29:39 AM »
New version of OpenGL Extensions Viewer now available !
 What's New
 Version 4.2.6:
 OpenGL 4.0 rendering tests now using indirect rendering
 Faster frame rate with now text font code
 Added atomic counters reports
 More variables now reported
 Updated database

20
3D-Tech News Around The Web / Qualcomm Adreno GPU driver August 2014
« on: August 08, 2014, 06:44:59 PM »
 This release contains a developer preview of the user-mode driver binaries for Qualcomm Adreno 3xx GPU on Nexus 4, Nexus 5, and Nexus 7 devices running Google Android 4.4.4 KitKat. This release includes OpenCL in addition to OpenGL ES. It has been tested with the KTU84P factory image. This release is intended only for developers, and does not meet our customer release quality. Refer to the included release notes for a list of fixes and enhancements.





Code: [Select]
Known Issues
  * Applications may hang when repeatedly disconnecting and reconnecting Adreno Profiler while plotting GPU Stall Metrics in Grapher.
  * Capturing OpenCL metrics with Adreno Profiler Scrubber may crash target application.
Fixes and Improvements
  * Fix a failure to release performance counter resources in some cases when using Adreno Profiler with certain NDK apps and OpenCL.
  * Fix a crash when connecting Adreno Profiler to an app that uses the disjoint timer query extension.
  * Add support to Adreno Profiler for OpenCL kernel type qualifier arguments.
  * Add support to Adreno Profiler for clCreateImage() 1D image, 1D image array, 2D image array.
  * Add support to Adreno Profiler for displaying multiple OpenCL queue/device/context combinations.
  * Fix a potential target application crash when Adreno Profiler metrics are enabled while the GPU is not idle.
  * Fix for incorrect scissor results if an app calls scissor and glBufferSubdata() between two draw calls.
  * Fix an issue with binning configuration calculation when MSAA is enabled.
  * Add support for ASTC texture formats in Adreno Profiler.
  * Fix potential null-pointer dereference in OpenCL.
  * Fix rotation issues with glBlitFramebuffer() and glClipPlane() when device is not in native orientation.
  * Fix for possible crash when using occlusion query while changing render targets.
  * Fix incorrect Adreno Profiler capture for applications using 2D texture arrays.
  * Fix potential crash during VAO validation in buffer update calls.
  * Add Adreno Profiler support for buffer transmission in clFillBuffer() and clFillImage().
  * Improve performance when invalidating texture attachments.
The following fixes were included in the June 2014 Adreno driver release for Nexus 4, Nexus 5, and Nexus 7 and are also included in this release:
  * Return CL_INVALID_MEM_OBJECT in clReleaseMemObj() as per spec.
  * Improve performance for {WebXPRT} benchmark.
  * Eliminate unnecessary resolves when invalidating FBO attachments to improve performance.
The following fixes were included in the May 2014 Adreno driver release for Nexus 4, Nexus 5, and Nexus 7 and are also included in this release:
  * Add OpenCL libraries.
  * Fix an issue with return statements in the middle of GLSL shaders causing a GPU hang.
  * Fix a performance regression in VBO validation.
  * Improve accuracy of Renderscript kernel execution duration measurements.
  * Fix crashes in some dEQP test cases.
  * Improve support for GL_RGB10_A2UI textures in Adreno Profiler.
  * Fix intermittent pink color tinting with certain WebGL content.
The following fixes were included in the April 2014 Adreno driver release for Nexus 4, Nexus 5, and Nexus 7 and are also included in this release:
  * Fix an intermittent crash that may happen when running dEQP-GLES2.functional.light_amount.1dir.
  * Fix rendering corruption in GFXBench Manhattan caused by out-of-sync VBO cache.
  * Fix small memory leak when waiting for thread termination.
  * Fix dEQP clipping test failures for point and line primitive types.
  * Add support for GL_HALF_FLOAT_OES rendering.
  * Improve RenderScript performance.
  * Prevent GPU page fault in certain SurfaceFlinger error conditions.
  * Allow 0 as an argument to the GLSL #line directive, as per spec.
  * Remove erroneous report of paletted texture formats from the GL_COMPRESSED_TEXTURE_FORMATS query when GL_OES_compressed_paletted_texture is not reported in the extension string list.
  * Fix a possible segmentation fault in glTexImage2D().
  * Fix mishandling of trailing whitespace after a GLSL #if directive.
  * Fix intermittent crashes from GLSL symbol table corruption.
  * Fix a scenario where performing a glQueryCounter() followed by glGet(GL_TIMESTAMP_EXT) and taking the difference between the two could result in a negative value, wrongly indicating a negative latency between the CPU and GPU.
  * Fix possible page fault when a shared EGL surface has a depth buffer.
  * Fix intermittent crash in Chrome browser that sometimes occurred while rotating and locking the device at the same time.
  * Add thread-safety for EGL images.
  * Fix Adreno Profiler shader override issues with shaders that use uniforms.
  * Reset flags for previously bound Transform Feedback buffer after draw call in case it is bound to more than one attribute.
  * Fix rendering corruption in Basemark Hover.
  * Fix a crash in the Tumblr app.
  * Fix intermittent crash in glBufferData() during VBO validation.
  * Fix rotation-related UI rendering issues in Gunman Clive.
  * Fix rendering corruption issue with H.265 video.
  * Fix YUV video rendering issues in Group Play app.
  * Fix Adreno Profiler conflict with occlusion query.
  * Fix screen tearing when using virtual displays via the developer options with HW overlays disabled.
  * Fix small memory leak on thread destruction.
  * Fix a crash-on-launch with Redline Rush.
  * Fix performance issues in Arma Tactics.
  * Allow non-constant indices on attribute vectors, as per GLSL spec.
  * Fix rendering corruption in GLBenchmark.
  * Fix eglClientWaitSyncKHR() to correctly use a 0 ms timeout when requested instead of a 1 ms timeout.
  * Improve glReadPixel() performance in landscape mode.
  * Fix page fault in dEQP glBlitFrameBuffer() test.
  * Fix crash-on-launch in Royal Revolt app.
  * Fix incorrect behavior of glClear() on MRTs.
  * Fix a crash when EOF immediately follows an #endif directive in GLSL.
  * Improve performance of aliased VBO updates.
  * Fix crash in GLES conformance test.
  * Fix rendering corruption in Unity’s “The Chase” demo.
  * Fix rendering corruption that occurs when the device is rotated and a visible surface is resized.
  * Fix for dEQP GLES3 polygon offset tests.
  * Enforce mutual exclusion of RGB and L for RGB buffers and luminance buffers.
  * Fix race condition in EGL triggered by certain Android mediaserver scenarios.
  * Create separate scope for if/if else/else statements in GLSL as per spec.
  * Fix incorrect conversion of negative float values to FALSE in glGetBoolean().
  * Reject invalid GLSL version numbers in #version directive.
  * Allow all integer constant expressions in GLSL #line directives.
  * Add driver support for EGL_KHR_create_context extension.
  * Fix rendering corruption with TGCOM24 app.
  * Fix the total count of uniform samplers when uniform samplers are used both in the vertex and fragment shader.
  * Fix erroneous compile errors for certain YUV GLSL shaders.
  * Fix a crash when indexing a sampler by a non-constant index in GLSL.
  * Fix issues with dEQP transform feedback tests.
  * Fix miplevel corruptions in conformance tests.
The following fixes were included in the March 2014 Adreno driver release for Nexus 4, Nexus 5, and Nexus 7 and are also included in this release:
  * Increase the maximum available number of uniform vectors from 251 to 256 for vertex shaders and from 221 to 224 for fragment shaders.
  * Fix possible lost nodes in GLSL preprocessor symbol table. Under certain conditions, using the #undef directive in a shader could cause other, unrelated symbols to be lost. This fixes a shader issue in The Amazing Spider-Man game from Gameloft.
  * Fix RBOs with no allocated storage to correctly report a zero sample count.
  * Fix potential crash when multiple GL contexts are rendering to the same EGL surface and the EGL surface is unmapped. This fixes a possible crash while running Google’s Compatibility Test Suite.
  * Fix crash in Epic’s Soul demo when adding grapher metrics in Adreno Profiler.
  * Add processor selection in Renderscript, to measure GPU and CPU execution times and choose the fastest processor for a given kernel to execute.
  * If a texture is changed in a shared context and the texture is bound or rebound in the current context, then update texture samplers to ensure the changes are visible in the current context, in accordance with Rule 4, section D.3.3 of the OpenGL® ES 3.0 {spec}.
  * Fast clear is not compatible with stencil mask, so disable it when stencil mask is enabled.
  * Allow glBlitFramebuffer() to blit within the default framebuffer if color buffers are for different surfaces.
  * Do not kill the application in case of GPU reset.
  * Fix problems with some uniforms in override programs not being set correctly by Adreno Profiler.
  * Fix handling of GL_RED, GL_RED_INTEGER, GL_RG, and GL_RG_INTEGER texture formats in Adreno Profiler.
  * Add GL3 constant names to token descriptions in Adreno Profiler.
  * Fix an Adreno Profiler crash exposed by a change to the way text rendering is implemented in the Android UI framework.
  * Change Adreno Profiler metric timestamps from milliseconds to microseconds to provide better data for per-rendercall metrics.
  * Fix a crash when Adreno Profiler is collecting scrubber metrics from a GLES3 app with occlusion querying.
  * Fix intermittent Adreno Profiler connection timeouts.
  * Fix intermittent Adreno Profiler disconnects while capturing.
  * Fix beginning-of-race crash in Asphalt 8.
The following fixes were included in the February 2014 Adreno driver release for Nexus 5 and Nexus 7 and are also included in this release for Nexus 4:
  * Add Adreno Profiler support for the following pixel data formats: GL_RED,GL_RED_INTEGER, GL_RG, GL_RG_INTEGER, GL_RGB_INTEGER, GL_RGBA_INTEGER.
  * Add Adreno Profiler support for reporting bin configuration.
  * Add Adreno Profiler support for uniform buffers.
  * Improve performance of uniform block look-ups when using Adreno Profiler program overrides.
  * Add Adreno Profiler support for transform feedback objects, including program overrides.
  * Add more Adreno Profiler structure version checks to better warn of data structure size differences between Adreno Profiler and driver.
  * Add Adreno Profiler support for GL_RGB9_E5 texture format.
  * Add Adreno Profiler support for GL_SRGBA8 texture format.
  * Add Adreno Profiler support for 2D texture arrays.
  * Add the ability to identify which GL thread is most active, to automatically select the best thread in the Adreno Profiler.
  * Fix Adreno Profiler reporting of uninitialized data.
  * Fix occasional crash of BombSquad (and other multithreaded apps) when connecting Adreno Profiler.
  * Add Adreno Profiler option to maintain overrides when capturing a new frame.
  * Fix Adreno Profiler failure with Epic Citadel.
  * Fix WebGL 1.0.2 conformance test textures/texture-size-limit.html case.
  * Fix handling of \r line endings in shader source.
The following fixes were included in the January 2014 Adreno driver release for Nexus 5 and are also included in this release for Nexus 7 and Nexus 4:
  * Add support for GL_EXT_multisampled_render_to_texture {extension}.
  * Fix incorrect output with glReadPixels() on an FBO with a texture attachment which is a cubemap face.
  * Fix incorrect lighting in the UBOTest example from the Adreno SDK. This fix addresses problems with scalar operations on UBOs.
  * Performance enhancements for rasterizer discard.
  * Performance enhancements for transform feedback.
  * Fix issue with graphics driver being unable to find and load Adreno Profiler libraries.
  * Cleaner handling for internal compiler error cases for UBOs.
  * Faster rsGetElementAt() and rsSample() runtime performance for programs that access many Allocation objects.
  * Improved stability when compiling complex RenderScript code.
  * Allow a count of zero in calls to glDrawBuffers() as per {spec}.
  * Fix errors in calculation of 2D texture array slice sizes for NPOT textures.
  * Generate GL_INVALID_VALUE from glTexStorage2D() and glTexStorage3D() when width, height, or depth exceed supported size.
  * Fix behavior of glGetFramebufferAttachmentParameteriv() to properly return GL_FRAMEBUFFER_DEFAULT for the default framebuffer instead of returning GL_NONE. (See OpenGL® ES Version 3.0 {spec} §6.1.13.)
  * Change glCopyTexImage2D() to prohibit SNORM as a destination format, as per Table 3.15 of the OpenGL® ES Version 3.0 {spec}.
  * Relaxed restrictions on the width and height parameters of glCompressedTexSubImage2D() to allow updates of 1x1 and 2x2 miplevels for ETC2/EAC formats. The incoming data buffer must still have its width and height padded to four, as required by ETC2/EAC, but the width and height parameters to the function call may reflect the actual texture dimensions.
  * Fix a potential double-free of memory inside glLinkProgram().
  * Fix reported value of gl_MaxVertexTextureImageUnits for GLES 3.0 hardware.
  * Add support for the following built-in constants in GLSL 3.0:
    * gl_MaxVertexOutputVectors
    * gl_MaxFragmentInputVectors
    * gl_MinProgramTexelOffset
    * gl_MaxProgramTexelOffset
  * Fix transposition of u and v coordinates in GLSL textureOffset() and texelFetch() functions.
  * Fix loss of w coordinate in GLSL textureOffset() and texelFetch() functions.
  * Fix GLSL matrix constructor to accept an array of matrices.



 

Pages: [1] 2 3 ... 146