Here are the notes from today's meeting in irc.freenode.net #blendercoders.
Ton writes:
Hi all,
Here are the notes from today's 14 UTC meeting in irc.freenode.net #blendercoders.
1) Release targets for Blender 2.76
- Planning
- According to schedule, we move to BCon3 today. That means that the release targets are defined, only what's agreed on and listed will go to git master.
- Target 'PTex' is going to be postponed again... we might need another developer to assist. Also to check on UDIM texture support.
- OpenSubdiv is in progress. Sergey suggests to make it a build option, because of OS X not supporting a mix of OpenGL 4 calls and older versions. Discussion on this is open...
will be done here and/or in IRC.- AMD news: OS X driver team managed to get the old Cycles ("Mega Kernel") compile and work, the next update of the beta "El Capitan" OS X should (might?) have this driver,
we'll keep you posted. The OS X team is working with us on getting split kernel to work too.- Interesting article, AMD writing about their Blender commitment.
- Antonis Ryakiotakis: speed optimizations for viewport are ready for master. Doc coming.
2) Other projects
- Joshua Leung works on new Grease Pencil 'sculpt' feature for 2.77:
- Joerg Mueller: the new Audaspace library will go to the system libraries and not into external, so that we can check how building with that works out.
- Campbell worked on script setup to do more exact OpenGL performance benchmarking. (Volunteers needed to add content and create a great Blender bench for gfx cards).
- Gooseberry final dev report is still in progress...
-Ton-
10 Comments
We finally have a text effect in the video editor:-) (in the partial release notes, before someone asks) That is going to save making so many images! Could do with more options though, particularly regarding size and font. Also, given the subtitle output ability, a 'hide all text strips' operator would prove handy (and that is before we even consider a multilingual work flow). Regardless, I am glad to see a first version:-)
Same here, I'm really glad to see that feature. Still waiting for a better multi core/GPU support for video editing, but that text feature is a big step forward. :)
Where do you see the part that tell about text effect strip ?
It's in the release notes for 2.76 http://wiki.blender.org/index.php/Dev:Ref/Release_Notes/2.76/Sequencer
I'd like to know why I can run a pre-compiled archive of Blender 2.75a with an old Nvidia 8600GTS GPU but cannot compile without CUDA from Trunk due to the following error:
/home/mdriftmeyer/Projects/Blender/amd-kernel-split/blender/intern/cycles/device/device_opencl.cpp: In member function ‘std::__cxx11::string ccl::OpenCLDeviceBase::kernel_build_options(const string*)’:
/home/mdriftmeyer/Projects/Blender/amd-kernel-split/blender/intern/cycles/device/device_opencl.cpp:1254:30: error: ‘CL_DEVICE_COMPUTE_CAPABILITY_MAJOR_NV’ was not declared in this scope
clGetDeviceInfo(cdDevice, CL_DEVICE_COMPUTE_CAPABILITY_MAJOR_NV,
^
/home/mdriftmeyer/Projects/Blender/amd-kernel-split/blender/intern/cycles/device/device_opencl.cpp:1256:30: error: ‘CL_DEVICE_COMPUTE_CAPABILITY_MINOR_NV’ was not declared in this scope
clGetDeviceInfo(cdDevice, CL_DEVICE_COMPUTE_CAPABILITY_MINOR_NV,
^
[ 9%] Building CXX object extern/libmv/third_party/ceres/CMakeFiles/extern_ceres.dir/internal/ceres/conjugate_gradients_solver.cc.o
intern/cycles/device/CMakeFiles/cycles_device.dir/build.make:146: recipe for target 'intern/cycles/device/CMakeFiles/cycles_device.dir/device_opencl.cpp.o' failed
make[2]: *** [intern/cycles/device/CMakeFiles/cycles_device.dir/device_opencl.cpp.o] Error 1
CMakeFiles/Makefile2:1061: recipe for target 'intern/cycles/device/CMakeFiles/cycles_device.dir/all' failed
make[1]: *** [intern/cycles/device/CMakeFiles/cycles_device.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
I'm planning on upgrading to the Radeon Fury Nano but that's dependent upon the release.
Explain the discrepancy.
For reference:
CPU: 8 Core FX-8350
RAM: 32GB DDR3 1866
Would really like to know if the OpenCL split kernel recognizes the FX processor as clinfo does and leverages it properly:
driftmeyer@horus:~/Projects/Blender/amd-kernel-split/gcc-blender$ clinfo
Number of platforms: 1
Platform Profile: FULL_PROFILE
Platform Version: OpenCL 2.0 AMD-APP (1642.5)
Platform Name: AMD Accelerated Parallel Processing
Platform Vendor: Advanced Micro Devices, Inc.
Platform Extensions: cl_khr_icd cl_amd_event_callback cl_amd_offline_devices
Platform Name: AMD Accelerated Parallel Processing
Number of devices: 1
Device Type: CL_DEVICE_TYPE_CPU
Vendor ID: 1002h
Board name:
Max compute units: 8
Max work items dimensions: 3
Max work items[0]: 1024
Max work items[1]: 1024
Max work items[2]: 1024
Max work group size: 1024
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 8
Preferred vector width double: 4
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 8
Native vector width double: 4
Max clock frequency: 1400Mhz
Address bits: 64
Max memory allocation: 8427588608
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 64
Max image 2D width: 8192
Max image 2D height: 8192
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 4096
Alignment (bits) of base address: 1024
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: Yes
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: Read/Write
Cache line size: 64
Cache size: 16384
Global memory size: 33710354432
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Global
Local memory size: 32768
Max pipe arguments: 16
Max pipe active reservations: 16
Max pipe packet size: 4132621312
Max global variable size: 1879048192
Max global variable preferred total size: 1879048192
Max read/write image args: 64
Max on device events: 0
Queue on device max size: 0
Max on device queues: 0
Queue on device preferred size: 0
SVM capabilities:
Coarse grain buffer: Yes
Fine grain buffer: Yes
Fine grain system: Yes
Atomics: Yes
Preferred platform atomic alignment: 0
Preferred global atomic alignment: 0
Preferred local atomic alignment: 0
Kernel Preferred work group size multiple: 1
Error correction support: 0
Unified memory for Host and Device: 1
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: Yes
Queue on Host properties:
Out-of-Order: No
Profiling : Yes
Queue on Device properties:
Out-of-Order: No
Profiling : No
Platform ID: 0x7f6cef05a830
Name: AMD FX(tm)-8350 Eight-Core Processor
Vendor: AuthenticAMD
Device OpenCL C version: OpenCL C 1.2
Driver version: 1642.5 (sse2,avx,fma4)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1642.5)
Extensions: cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_device_fission cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_spir cl_khr_gl_event
I see no reference to OpenCL AMD being able to recognize the FX directly.
I would think that should be there as a way to disable any hang on the old NVidia card.
Hmmm, I have a suggestion for PTex and OpenSubdiv - please just remove it temporarily from the release plans. It looks quite bad and unprofessional when it appears in plans for months now, while it seems no real chances to release anytime soon. I don't want to say that no great features are added every new release, but keeping PTex and OpenSubdiv in all an every release plans month by month feels just... bad. It'd be better to simply remove it and return when there are a REAL chances to implement in the particular release.
It sounds like OpenSubDiv is going in, just not available on every platform. Which is fine, not every feature needs to be on every platform, and the more users testing the implementation faster, the better the final implementation will be.
Wonderful, as ever, to see the pace of Blender development never slows down! Video editing captions is a very good addition; let's see that grow. Amusing to see that Grease Pencil is being expanded - it's something I have never ever needed to use, yet it appears like a plague all over the UI ;-)
I really like how Grease Pencil is progressing, waiting to see great works made with it :)