AxioDL/metaforce
2
0
Fork 0
mirror of https://github.com/AxioDL/metaforce.git synced 2025-12-08 21:07:42 +00:00
Code Issues Projects Releases Wiki Activity

SIMD refactor

Browse Source
This commit is contained in:
Jack Andersen
2018-12-07 15:49:15 -10:00
parent e1fa938127
commit 7c7da6671a
201 changed files with 2475 additions and 1906 deletions
Download Patch File Download Diff File Expand all files Collapse all files

95
DataSpec/DNACommon/ANIM.cpp
View File

@@ -39,50 +39,53 @@ size_t ComputeBitstreamSize(size_t keyFrameCount, const std::vector<Channel>& ch
static inline QuantizedRot QuantizeRotation(const Value& quat, atUint32 div)
{
float q = M_PIF / 2.0f / float(div);
zeus::simd_floats f(quat.simd);
return
{
{
atInt32(std::asin(quat.v4.vec[1]) / q),
atInt32(std::asin(quat.v4.vec[2]) / q),
atInt32(std::asin(quat.v4.vec[3]) / q),
atInt32(std::asin(f[1]) / q),
atInt32(std::asin(f[2]) / q),
atInt32(std::asin(f[3]) / q),
},
(quat.v4.vec[0] < 0.f) ? true : false
(f[0] < 0.f)
};
}
static inline Value DequantizeRotation(const QuantizedRot& v, atUint32 div)
{
float q = M_PIF / 2.0f / float(div);
Value retval =
{
athena::simd_floats f = {
0.0f,
std::sin(v.v[0] * q),
std::sin(v.v[1] * q),
std::sin(v.v[2] * q),
};
retval.v4.vec[0] = std::sqrt(std::max((1.0f -
(retval.v4.vec[1] * retval.v4.vec[1] +
retval.v4.vec[2] * retval.v4.vec[2] +
retval.v4.vec[3] * retval.v4.vec[3])), 0.0f));
retval.v4.vec[0] = v.w ? -retval.v4.vec[0] : retval.v4.vec[0];
f[0] = std::sqrt(std::max((1.0f -
(f[1] * f[1] +
f[2] * f[2] +
f[3] * f[3])), 0.0f));
f[0] = v.w ? -f[0] : f[0];
Value retval;
retval.simd.copy_from(f);
return retval;
}
static inline Value DequantizeRotation_3(const QuantizedRot& v, atUint32 div)
{
float q = 1.0f / float(div);
Value retval =
{
athena::simd_floats f = {
0.0f,
v.v[0] * q,
v.v[1] * q,
v.v[2] * q,
};
retval.v4.vec[0] = std::sqrt(std::max((1.0f -
(retval.v4.vec[1] * retval.v4.vec[1] +
retval.v4.vec[2] * retval.v4.vec[2] +
retval.v4.vec[3] * retval.v4.vec[3])), 0.0f));
retval.v4.vec[0] = v.w ? -retval.v4.vec[0] : retval.v4.vec[0];
f[0] = std::sqrt(std::max((1.0f -
(f[1] * f[1] +
f[2] * f[2] +
f[3] * f[3])), 0.0f));
f[0] = v.w ? -f[0] : f[0];
Value retval;
retval.simd.copy_from(f);
return retval;
}
@@ -337,9 +340,10 @@ BitstreamWriter::write(const std::vector<std::vector<Value>>& chanKeys,
++it)
{
const Value* key = &*it;
maxTransVal = std::max(maxTransVal, std::fabs(key->v3.vec[0]));
maxTransVal = std::max(maxTransVal, std::fabs(key->v3.vec[1]));
maxTransVal = std::max(maxTransVal, std::fabs(key->v3.vec[2]));
zeus::simd_floats f(key->simd);
maxTransVal = std::max(maxTransVal, std::fabs(f[0]));
maxTransVal = std::max(maxTransVal, std::fabs(f[1]));
maxTransVal = std::max(maxTransVal, std::fabs(f[2]));
}
break;
}
@@ -350,9 +354,10 @@ BitstreamWriter::write(const std::vector<std::vector<Value>>& chanKeys,
++it)
{
const Value* key = &*it;
maxScaleVal = std::max(maxScaleVal, std::fabs(key->v3.vec[0]));
maxScaleVal = std::max(maxScaleVal, std::fabs(key->v3.vec[1]));
maxScaleVal = std::max(maxScaleVal, std::fabs(key->v3.vec[2]));
zeus::simd_floats f(key->simd);
maxScaleVal = std::max(maxScaleVal, std::fabs(f[0]));
maxScaleVal = std::max(maxScaleVal, std::fabs(f[1]));
maxScaleVal = std::max(maxScaleVal, std::fabs(f[2]));
}
break;
}
@@ -383,17 +388,19 @@ BitstreamWriter::write(const std::vector<std::vector<Value>>& chanKeys,
}
case Channel::Type::Translation:
{
chan.i = {atInt32((*kit)[0].v3.vec[0] / transMultOut),
atInt32((*kit)[0].v3.vec[1] / transMultOut),
atInt32((*kit)[0].v3.vec[2] / transMultOut)};
zeus::simd_floats f((*kit)[0].simd);
chan.i = {atInt32(f[0] / transMultOut),
atInt32(f[1] / transMultOut),
atInt32(f[2] / transMultOut)};
initVals.push_back(chan.i);
break;
}
case Channel::Type::Scale:
{
chan.i = {atInt32((*kit)[0].v3.vec[0] / scaleMultOut),
atInt32((*kit)[0].v3.vec[1] / scaleMultOut),
atInt32((*kit)[0].v3.vec[2] / scaleMultOut)};
zeus::simd_floats f((*kit)[0].simd);
chan.i = {atInt32(f[0] / scaleMultOut),
atInt32(f[1] / scaleMultOut),
atInt32(f[2] / scaleMultOut)};
initVals.push_back(chan.i);
break;
}
@@ -431,9 +438,10 @@ BitstreamWriter::write(const std::vector<std::vector<Value>>& chanKeys,
it != kit->end();
++it)
{
QuantizedValue cur = {atInt32(it->v3.vec[0] / transMultOut),
atInt32(it->v3.vec[1] / transMultOut),
atInt32(it->v3.vec[2] / transMultOut)};
zeus::simd_floats f(it->simd);
QuantizedValue cur = {atInt32(f[0] / transMultOut),
atInt32(f[1] / transMultOut),
atInt32(f[2] / transMultOut)};
chan.q[0] = std::max(chan.q[0], atUint8(cur.qFrom(last, 0)));
chan.q[1] = std::max(chan.q[1], atUint8(cur.qFrom(last, 1)));
chan.q[2] = std::max(chan.q[2], atUint8(cur.qFrom(last, 2)));
@@ -447,9 +455,10 @@ BitstreamWriter::write(const std::vector<std::vector<Value>>& chanKeys,
it != kit->end();
++it)
{
QuantizedValue cur = {atInt32(it->v3.vec[0] / scaleMultOut),
atInt32(it->v3.vec[1] / scaleMultOut),
atInt32(it->v3.vec[2] / scaleMultOut)};
zeus::simd_floats f(it->simd);
QuantizedValue cur = {atInt32(f[0] / scaleMultOut),
atInt32(f[1] / scaleMultOut),
atInt32(f[2] / scaleMultOut)};
chan.q[0] = std::max(chan.q[0], atUint8(cur.qFrom(last, 0)));
chan.q[1] = std::max(chan.q[1], atUint8(cur.qFrom(last, 1)));
chan.q[2] = std::max(chan.q[2], atUint8(cur.qFrom(last, 2)));
@@ -490,9 +499,10 @@ BitstreamWriter::write(const std::vector<std::vector<Value>>& chanKeys,
}
case Channel::Type::Translation:
{
QuantizedValue cur = {atInt32(val.v3.vec[0] / transMultOut),
atInt32(val.v3.vec[1] / transMultOut),
atInt32(val.v3.vec[2] / transMultOut)};
zeus::simd_floats f(val.simd);
QuantizedValue cur = {atInt32(f[0] / transMultOut),
atInt32(f[1] / transMultOut),
atInt32(f[2] / transMultOut)};
quantize(newData.get(), chan.q[0], cur[0] - last[0]);
quantize(newData.get(), chan.q[1], cur[1] - last[1]);
quantize(newData.get(), chan.q[2], cur[2] - last[2]);
@@ -501,9 +511,10 @@ BitstreamWriter::write(const std::vector<std::vector<Value>>& chanKeys,
}
case Channel::Type::Scale:
{
QuantizedValue cur = {atInt32(val.v3.vec[0] / scaleMultOut),
atInt32(val.v3.vec[1] / scaleMultOut),
atInt32(val.v3.vec[2] / scaleMultOut)};
zeus::simd_floats f(val.simd);
QuantizedValue cur = {atInt32(f[0] / scaleMultOut),
atInt32(f[1] / scaleMultOut),
atInt32(f[2] / scaleMultOut)};
quantize(newData.get(), chan.q[0], cur[0] - last[0]);
quantize(newData.get(), chan.q[1], cur[1] - last[1]);
quantize(newData.get(), chan.q[2], cur[2] - last[2]);