On Windows/MSVC this gives a minor (~20%) speedup presumably due to a faster float/int formatter. On macOS (Xcode13), this gives a massive speedup, since snprintf that is in system libraries ends up spending almost all the time inside some locale-related mutex lock.
The actual exporter code becomes quite a bit smaller too, since it does not have to do any juggling to support std::string arguments, and the buffer handling code is smaller as well.
Windows (VS2022 release build, Ryzen 5950X 32 threads) timings:
- Blender 3.0 splash scene (2.4GB obj): 4.57s -> 3.86s
- Monkey subdivided level 6 (330MB obj): 1.10s -> 0.99s
macOS (Xcode 13 release build, Apple M1Max) timings:
- Blender 3.0 splash scene (2.4GB obj): 21.03s -> 5.52s
- Monkey subdivided level 6 (330MB obj): 3.28s -> 1.20s
Linux (ThreadRipper 3960X 48 threads) timings:
- Blender 3.0 splash scene (2.4GB obj): 10.10s -> 4.40s
- Monkey subdivided level 6 (330MB obj): 2.16s -> 1.37s
The produced obj/mtl files are identical to before.
Reviewed By: Howard Trickey, Dalai Felinto
Differential Revision: https://developer.blender.org/D13998
This is patch D14349 from Aras Pranckevicius.
The logic in the code was _completely different_ from the documentation
and what the python exporter in 3.0 did. The new code assumed that
"export material groups" meant "append material name to the object name",
and was only ever kicking in when the "export object groups" option was
also checked. But the proper behavior (as in 3.0 exporter & the online docs),
is to emit g objectname_materialname before each usemtl line. Which is something entirely else.
This is patch D14347 from Aras Pranckevicius.
Instead of scaling "the scene" (i.e. transform vertices by object matrix,
then multiply by scale factor), it was instead first applying the scale
factor in local space, and then transforming by the object matrix.
This applies patch D14343 from Aras Pranckevicius, with a description:
The new 3.1+ OBJ exporter did not have correct logic when faced with
non-uniform & mirrored (negative on odd number of axes) object scale:
- Normals were not transformed correctly (should use inverse transpose of the matrix),
and were not normalized,
- Face order was not "flipped" when transform has negative scale on odd number of axes
(visible when using "face orientation" viewport overlay).
Use a shorter/simpler license convention, stops the header taking so
much space.
Follow the SPDX license specification: https://spdx.org/licenses
- C/C++/objc/objc++
- Python
- Shell Scripts
- CMake, GNUmakefile
While most of the source tree has been included
- `./extern/` was left out.
- `./intern/cycles` & `./intern/atomic` are also excluded because they
use different header conventions.
doc/license/SPDX-license-identifiers.txt has been added to list SPDX all
used identifiers.
See P2788 for the script that automated these edits.
Reviewed By: brecht, mont29, sergey
Ref D14069
Fixes T95384. New exporter was missing a fix for T94516 that recently got applied to the python exporter.
Also changed the obj export tests code so that when save_failing_test_output is requested and MTL result is different from the golden expectation, it is saved as well, similar to how it's done for the OBJ file result.
Also fixed conflicts due to the change in file writing in the new obj exporter
in master, and fixed one of the tests that was added in master but not 3.1.
The new wavefront .obj exporter in 3.1 was producing slightly invalid parm line syntax (missing u), and was not setting first/last N params to zeroes and ones for curves with "endpoint" flag properly.
This is a patch from Aras Pranckevicius, D13927. See that patch for full
details. On Windows, the many small fprintfs were taking up a large amount
of time and significant speedup comes from using snprintf into chained buffers,
and writing them all out later.
On both Windows and Linux, parallelizing the processing by Object can also lead
to a significant increase in speed.
The 3.0 splash screen scene exports 8 times faster than the current C++ exporter
on a Windows machine with 32 threads, and 5.8 times faster on a Linux machine
with 48 threads.
There is admittedly more memory usage for this, but it is still using 25 times
less memory than the old python exporter on the 3.0 splash screen scene, so
this seems an acceptable tradeoff. If use cases come up for exporting obj files
that exceed the memory size of users, a flag could be added to not parallelize
and write the buffers out every so often.
Previously, the new obj exporter was only exporting per-vertex normals for faces
marked as "smooth". But a face can have custom normals, as soon as the normals
data layer exists. This change makes it follow the behavior of USD & Collada
exporters and the old Python one, which also export per-vertex normals as soon
as the layer is there. (From Patch D13957.)
The new OBJ exporter did not handle object instances.
The fix is to use a dependency graph iterator, asking for instances.
Unfortunately that iterator makes a temporary copy of instance objects
that does not persist past the iteration, but we need to save all the
objects and meshes to write later, so the Object has to be copied now.
This changed some unit tests. Even though the tests don't have instancing,
the iterator also picks up some Text objects as Mesh ones (which is a good
thing), resulting in two more objects in the all_objects.obj file output.
Some new obj exporter tests were disabled because the normals were different
in the last decimal place on different platforms.
The old python exporter deduped normals with their coordinates rounded to
four decimal places. This change does the same in the new exporter.
On one test, this produced a file 25% smaller and even ran 10% faster.
The switch to how normals are kept has led to tiny differences in
the normal output values on different platforms. Disabling the failing
tests while working on a solution to this problem.
This patch implements the vector types (i.e:`float2`) by making heavy
usage of templating. All vector functions are now outside of the vector
classes (inside the `blender::math` namespace) and are not vector size
dependent for the most part.
In the ongoing effort to make shaders less GL centric, we are aiming
to share more code between GLSL and C++ to avoid code duplication.
####Motivations:
- We are aiming to share UBO and SSBO structures between GLSL and C++.
This means we will use many of the existing vector types and others
we currently don't have (uintX, intX). All these variations were
asking for many more code duplication.
- Deduplicate existing code which is duplicated for each vector size.
- We also want to share small functions. Which means that vector
functions should be static and not in the class namespace.
- Reduce friction to use these types in new projects due to their
incompleteness.
- The current state of the `BLI_(float|double|mpq)(2|3|4).hh` is a
bit of a let down. Most clases are incomplete, out of sync with each
others with different codestyles, and some functions that should be
static are not (i.e: `float3::reflect()`).
####Upsides:
- Still support `.x, .y, .z, .w` for readability.
- Compact, readable and easilly extendable.
- All of the vector functions are available for all the vectors types
and can be restricted to certain types. Also template specialization
let us define exception for special class (like mpq).
- With optimization ON, the compiler unroll the loops and performance
is the same.
####Downsides:
- Might impact debugability. Though I would arge that the bugs are
rarelly caused by the vector class itself (since the operations are
quite trivial) but by the type conversions.
- Might impact compile time. I did not saw a significant impact since
the usage is not really widespread.
- Functions needs to be rewritten to support arbitrary vector length.
For instance, one can't call `len_squared_v3v3` in
`math::length_squared()` and call it a day.
- Type cast does not work with the template version of the `math::`
vector functions. Meaning you need to manually cast `float *` and
`(float *)[3]` to `float3` for the function calls.
i.e: `math::distance_squared(float3(nearest.co), positions[i]);`
- Some parts might loose in readability:
`float3::dot(v1.normalized(), v2.normalized())`
becoming
`math::dot(math::normalize(v1), math::normalize(v2))`
But I propose, when appropriate, to use
`using namespace blender::math;` on function local or file scope to
increase readability.
`dot(normalize(v1), normalize(v2))`
####Consideration:
- Include back `.length()` method. It is quite handy and is more C++
oriented.
- I considered the GLM library as a candidate for replacement. It felt
like too much for what we need and would be difficult to extend / modify
to our needs.
- I used Macros to reduce code in operators declaration and potential
copy paste bugs. This could reduce debugability and could be reverted.
- This touches `delaunay_2d.cc` and the intersection code. I would like
to know @howardt opinion on the matter.
- The `noexcept` on the copy constructor of `mpq(2|3)` is being removed.
But according to @JacquesLucke it is not a real problem for now.
I would like to give a huge thanks to @JacquesLucke who helped during this
and pushed me to reduce the duplication further.
Reviewed By: brecht, sergey, JacquesLucke
Differential Revision: https://developer.blender.org/D13791
This patch implements the vector types (i.e:float2) by making heavy
usage of templating. All vector functions are now outside of the vector
classes (inside the blender::math namespace) and are not vector size
dependent for the most part.
In the ongoing effort to make shaders less GL centric, we are aiming
to share more code between GLSL and C++ to avoid code duplication.
Motivations:
- We are aiming to share UBO and SSBO structures between GLSL and C++.
This means we will use many of the existing vector types and others we
currently don't have (uintX, intX). All these variations were asking
for many more code duplication.
- Deduplicate existing code which is duplicated for each vector size.
- We also want to share small functions. Which means that vector functions
should be static and not in the class namespace.
- Reduce friction to use these types in new projects due to their
incompleteness.
- The current state of the BLI_(float|double|mpq)(2|3|4).hh is a bit of a
let down. Most clases are incomplete, out of sync with each others with
different codestyles, and some functions that should be static are not
(i.e: float3::reflect()).
Upsides:
- Still support .x, .y, .z, .w for readability.
- Compact, readable and easilly extendable.
- All of the vector functions are available for all the vectors types and
can be restricted to certain types. Also template specialization let us
define exception for special class (like mpq).
- With optimization ON, the compiler unroll the loops and performance is
the same.
Downsides:
- Might impact debugability. Though I would arge that the bugs are rarelly
caused by the vector class itself (since the operations are quite trivial)
but by the type conversions.
- Might impact compile time. I did not saw a significant impact since the
usage is not really widespread.
- Functions needs to be rewritten to support arbitrary vector length. For
instance, one can't call len_squared_v3v3 in math::length_squared() and
call it a day.
- Type cast does not work with the template version of the math:: vector
functions. Meaning you need to manually cast float * and (float *)[3] to
float3 for the function calls.
i.e: math::distance_squared(float3(nearest.co), positions[i]);
- Some parts might loose in readability:
float3::dot(v1.normalized(), v2.normalized())
becoming
math::dot(math::normalize(v1), math::normalize(v2))
But I propose, when appropriate, to use
using namespace blender::math; on function local or file scope to
increase readability. dot(normalize(v1), normalize(v2))
Consideration:
- Include back .length() method. It is quite handy and is more C++
oriented.
- I considered the GLM library as a candidate for replacement.
It felt like too much for what we need and would be difficult to
extend / modify to our needs.
- I used Macros to reduce code in operators declaration and potential
copy paste bugs. This could reduce debugability and could be reverted.
- This touches delaunay_2d.cc and the intersection code. I would like to
know @Howard Trickey (howardt) opinion on the matter.
- The noexcept on the copy constructor of mpq(2|3) is being removed.
But according to @Jacques Lucke (JacquesLucke) it is not a real problem
for now.
I would like to give a huge thanks to @Jacques Lucke (JacquesLucke) who
helped during this and pushed me to reduce the duplication further.
Reviewed By: brecht, sergey, JacquesLucke
Differential Revision: http://developer.blender.org/D13791
This was originally written by Ankit Meel as a GSoC 2020 project.
Howard Trickey added some tests and made some corrections/modifications.
See D13046 for more details.
This commit inserts a new menu item into the export menu called
"Wavefront OBJ (.obj) - New".
For now the old Python exporter remains in the menu, along with
the Python importer, but we plan to remove it soon (leaving the
old addon bundled with Blender but not enabled by default).
