juli/OrcaSlicer
1
1
Fork 0
mirror of https://github.com/OrcaSlicer/OrcaSlicer.git synced 2026-10-17 04:41:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

Rotate instead of shear for slicing stage (#30)

Browse Source 
* initial commit

* fix upper bounds for assemblies

* significantly less Z shift issues, still not quite tamped down yet though

* add instrumentation to logs

* finally found the issue

* update printer defaults
This commit is contained in:
Joseph Robertson
2026-05-22 15:21:33 -05:00
committed by GitHub
parent 218881c6f6
commit a9bae54f20
16 changed files with 513 additions and 96 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/libslic3r/BeltGCodeWriter.cpp
+24 -2
View File
@@ -57,17 +57,39 @@ Vec3d BeltGCodeWriter::to_cartesian(const Vec3d &pos) const
Vec3d BeltGCodeWriter::to_machine_coords(const Vec3d &pos) const
{
// Step 1+2: To Cartesian (back_transform + axis_remap).
Vec3d result = to_cartesian(pos);
Vec3d after_back = m_belt_back_transform.apply(pos);
Vec3d result = apply_axis_remap(after_back);
Vec3d after_remap = result;
// Step 3: Per-axis origin snap (computed in the Cartesian frame).
for (int i = 0; i < 3; ++i)
if (m_origin_snap[i])
result[i] -= (m_origin_bbox_min[i] - m_origin_offset[i]);
Vec3d after_snap = result;
// Step 4: Machine-frame transform (gcode_shear / gcode_scale / post_gcode_remap)
// applied LAST so it acts as a global linear transform on the placed coords.
// Order matters: putting it before origin_snap would feed sheared bbox corners
// into the snap's per-object min calculation, mis-normalizing non-cubic geometries
// (the corners of the original bbox aren't extreme points of the sheared shape).
return m_machine_frame_transform.apply(result);
Vec3d final = m_machine_frame_transform.apply(result);
// [BELT-DEBUG] One-shot log per layer transition (i.e. when the input Z
// crosses an integer mm boundary) to keep the log volume manageable while
// still capturing one sample per ~5 layers. Shows the full pipeline so
// Case A vs Case B can be compared step-by-step.
static thread_local int s_last_logged_z = std::numeric_limits<int>::min();
int z_bucket = static_cast<int>(std::floor(pos.z() * 5.0)); // every 0.2mm
if (z_bucket != s_last_logged_z) {
s_last_logged_z = z_bucket;
BOOST_LOG_TRIVIAL(warning) << "[BELT-DEBUG] to_machine_coords"
<< " slicer_in=(" << pos.x() << "," << pos.y() << "," << pos.z() << ")"
<< " after_back=(" << after_back.x() << "," << after_back.y() << "," << after_back.z() << ")"
<< " after_remap=(" << after_remap.x() << "," << after_remap.y() << "," << after_remap.z() << ")"
<< " after_snap=(" << after_snap.x() << "," << after_snap.y() << "," << after_snap.z() << ")"
<< " final=(" << final.x() << "," << final.y() << "," << final.z() << ")"
<< " mft_active=" << m_machine_frame_transform.is_active()
<< " back_active=" << m_belt_back_transform.is_active();
}
return final;
}
// ---- Overridden movement methods ------------------------------------------