aboutsummaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
* Matrix4f Perf: Enhance invert(), Drop (test) load on Matrix4f.mul(Matrix4f) ↵Sven Gothel2023-04-055-187/+409
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | for fair and realistic numbers - Both mul() ops faster than FloatUtil Enhanced invert() of Matrix4f* and FloatUtil: Use 1f/det factor for burst scale. Enhanced Matrix4f.invert(..): Use factored-out mulScale() to deliver the scale, giving a good 10% advantage on aarch64 and amd64. Brings Matrix4f.invert(..) on par w/ FloatUtil, on aarch64 even a 14% advantage. +++ TestMatrix4f02MulNOUI added an additional Matrix4f.load() to the mul(Matrix4f) loop test, which surely is an extra burden and not realistic as the mul(Matrix4f, Matrix4f) and FloatUtil pendants also don't count loading a value. Matrix4f.mul(Matrix4f) shall be used to utilize an already stored value anyways. Matrix4f.mul(Matrix4f) didn't really exist in FloatUtil. Same is true for Matrix4f.invert(), re-grouped order, i.e. pushing the non-arg variant last. +++ Revised performance numbers from commit 15e60161787224e85172685f74dc0ac195969b51 AMD64 + OpenJDK17 - FloatUtil.multMatrix(a, a_off, b, b_off, dest) is considerable slower than all - Matrix4f.mul(a, b) roughly ~10% faster than FloatUtil.multMatrix(a, b, dest) - Matrix4f.mul(b) roughly ~18% faster than FloatUtil.multMatrix(a, b, dest) (*) - Matrix4f.invert(a) roughly ~ 2% faster than FloatUtil.invertMatrix(..) - Matrix4f.invert() roughly ~ 4% slower than FloatUtil.invertMatrix(..) (*) - Launched: nice -19 scripts/tests-x64.sh RaspberryPi 4b aarch64 + OpenJDK17 - FloatUtil.multMatrix(a, a_off, b, b_off, dest) is considerable slower than all - Matrix4f.mul(a, b) roughly ~ 9% faster than FloatUtil.multMatrix(a, b, dest) - Matrix4f.mul(b) roughly ~14% faster than FloatUtil.multMatrix(a, b, dest) (*) - Matrix4f.invert(a) roughly ~14% faster than FloatUtil.invertMatrix(..) - Matrix4f.invert() roughly ~12% faster than FloatUtil.invertMatrix(..) (*) - Launched: nice -19 scripts/tests-linux-aarch64.sh (*) not a true comparison in feature, as operating on 'this' matrix values for one argument, unavailable to FloatUtil. Conclusion - Matrix4f.mul(..) is considerable faster! - Matrix4f.invert(..) faster, esp on aarch64 And additional Matrix4fb tests using float[16] similar to FloatUtil also demonstrates less performance compared to Matrix4f using dedicated float fields.
* AABBOX: Revert API doc copy/type errors from commit ↵Sven Gothel2023-04-051-7/+7
| | | | 15e60161787224e85172685f74dc0ac195969b51
* TestAWTCardLayoutAnimatorStartStopBug532: Suppress warningsSven Gothel2023-04-051-1/+10
|
* GraphUI: Adopting Vec*f API; Adding Group; Scene + Group are Container, ↵Sven Gothel2023-04-0519-273/+1214
| | | | | | | | | | | | | | | | traversing the PMVMatrix throughout childs (-> see TreeTool). Utilizing the Vec*f (and Matrix4f) API w/ AABBox et al renders our code more clean & safe, see commit 15e60161787224e85172685f74dc0ac195969b51. A Group allows to contain multiple Shapes, hence the PMVMatrix must be traversed accordingly using TreeTool for all operations (draw, picking, win->obj coordinates, ..). Hence Scene + Group are now implementing Container and reuse code via TreeTool and a Shape.Visitor*. This will allow further simplification of user code.
* Math: Complete Matrix4f w/ Vec[234]f and adopt it throughout Quaternion, ↵Sven Gothel2023-04-0546-1374/+6578
| | | | | | | | | | | | | | | | | | | | | | | | | | | Ray, AABBox, Frustum, Stereo*, ... adding hook to PMVMatrix Motivation was to simplify matrix + vector math usage, ease review and avoid usage bugs. Matrix4f implementation uses dedicated float fields instead of an array. Performance didn't increase much, as JVM >= 11(?) has some optimizations to drop the array bounds check. AMD64 + OpenJDK17 - Matrix4f.mul(a, b) got a roughly ~10% enhancement over FloatUtil.multMatrix(a, b, dest) - Matrix4f.mul(b) roughly ~3% slower than FloatUtil.multMatrix(a, b, dest) - FloatUtil.multMatrix(a, a_off, b, b_off, dest) is considerable slower than all - Matrix4f.invert(..) roughly ~3% slower than FloatUtil.invertMatrix(..) RaspberryPi 4b aarch64 + OpenJDK17 - Matrix4f.mul(a, b) got a roughly ~10% enhancement over FloatUtil.multMatrix(a, b, dest) - Matrix4f.mul(b) roughly ~20% slower than FloatUtil.multMatrix(a, b) - FloatUtil.multMatrix(a, a_off, b, b_off, dest) is considerable slower than all - Matrix4f.invert(..) roughly ~4% slower than FloatUtil.invertMatrix(..) Conclusion - Matrix4f.mul(b) needs to be revised (esp for aarch64) - Matrix4f.invert(..) should also not be slower ..
* GraphUI: Use Shape.ZAscendingComparator in Scene, to be reused across typsSven Gothel2023-03-312-29/+33
|
* GraphUI: Relocate com.jogamp.graph.ui.gl.* -> com.jogamp.graph.ui.*, resolve ↵Sven Gothel2023-03-3126-72/+71
| | | | | | GL/VK abstraction at a later time differently Actual GPU rendering toolkit dependency can be abstracted differently, i.e. GPU <- { GL, VK } etc.
* Graph Shader: Complete the USE_DISCARD logic avoiding output set after ↵Sven Gothel2023-03-3012-36/+67
| | | | discard, even though technically allowed (ignored after discard)
* GraphUI Shape: DragZoom: Allow resize_sxy_min down to 1/2%, need to find a ↵Sven Gothel2023-03-301-1/+1
| | | | | | | | | | good way to auto-deteminate this one (dpi, ..) Type animation UISceneDemo03 requires this to resize a single glyph, i.e. 0.03f was too big. 1% might be OK, but then .. maybe not. In general, a dpi autodetermination of a minimal scale/size might be OK, assuming user sits in front of a screen and not having a multi-screen circus arena setup ;-)
* GraphUI: Fix debug-box and allow API access in Shape (off, thickness ↵Sven Gothel2023-03-307-34/+79
| | | | fractional to box size) and Scene for all Shapes.
* GraphUI Demos: Cover more std arguments via GraphUIDemoArgsSven Gothel2023-03-303-104/+53
|
* Graph GLSL: Enable 'discard' in fragment shader w/o ↵Sven Gothel2023-03-3014-55/+64
| | | | | | | | | GLRendererQuirks.GLSLBuggyDiscard to avoid overdraw of such regions. Historically we disabled `discard` due to an old NV tegra2 compiler bug, which caused the compiler to freeze. Today we no more seem to have this GLSL compiler issue, i.e. GLRendererQuirks.GLSLBuggyDiscard never gets set.
* GraphUI Demo: UISceneDemo20 use Scene.screenshot(..), drop own codeSven Gothel2023-03-301-31/+3
|
* GraphUI Demo UISceneDemo03: Add 1 space padding to demo-text, make ↵Sven Gothel2023-03-301-10/+36
| | | | screenshots, add launch screen (visible @ '-wait')
* GraphUI Scene: Add screenshot(..) method for convenienceSven Gothel2023-03-301-4/+37
|
* GraphUI: Add GlyphShape representing a single Font.Glyph as a GraphShape; ↵Sven Gothel2023-03-304-114/+417
| | | | | | | | | | | | | | | | | | Use w/ UISceneDemo03 Type Animation... A list of GlyphShape can be created via GlyphShape.processString(..), which preserves all details incl. intended original unscaled position and its kerning. Whitespace or contourless Glyphs are dropped. A GlyphShape is represented in font em-size [0..1] unscaled. +++ UISceneDemo03 Type Animation - Using GlyphShape and apply scaling via its Shape.setScale() - Recalc fontScale per used text - Refined 'arrival' criteria and smoothing out near target w/ speed-up rotation - Using GraphUIDemoArgs to parse common commandline demo options
* GraphUI Shape.setTransform(..): Scale before rotate to preserve target-size ↵Sven Gothel2023-03-301-14/+14
| | | | in rotation
* GraphUI Shape: Refine API docSven Gothel2023-03-301-8/+8
|
* Graph*: API doc refinementSven Gothel2023-03-302-1/+8
|
* Vec2f, Vec3f: Return this @ set(..), add set(float[]) and add(float, ..)Sven Gothel2023-03-302-22/+60
|
* GraphUI Shape: Rename rotOrigin -> rotPivotSven Gothel2023-03-281-18/+18
|
* GraphUI Shape.setTransform(): Fix !sameScaleRotatePivot && hasRotate && ↵Sven Gothel2023-03-281-2/+2
| | | | | | !hasRotPivot: Don't scale pivot back from center-pivot, same as w/ hasRotPivot A copy & paste bug ..
* GraphUI Scene: Added API doc argsSven Gothel2023-03-281-6/+6
|
* Graph Font: TypecastFont: getGlyphBoundsFU() + processString(): If ↵Sven Gothel2023-03-281-4/+6
| | | | isWhitespace() continue w/ AABBox resize, but don't earmark this (whitespace) glyph for next left_glyph (kerning).
* GraphUI Demo UISceneDemo03: Add kerning and compensate on label 0/0 origin ↵Sven Gothel2023-03-281-8/+22
| | | | using Glyph bounds minY()
* GraphUI Demo UISceneDemo02: Use em-sized label w/ scale.Sven Gothel2023-03-281-8/+8
|
* GraphUI: Decouple GraphShape from Shape, i.e. allow future Shape w/o ↵Sven Gothel2023-03-2818-303/+465
| | | | Graph/GLRegion
* GraphUI Button: Simplify addShapeToRegion()Sven Gothel2023-03-282-15/+15
|
* GraphUI Label: Add getScaledLineHeight() and refine setFontScale(), i.e. ↵Sven Gothel2023-03-281-3/+21
| | | | only act and markDirty() on change
* GraphUI Label: Remove glyphVisitor's debug code (exception handling)Sven Gothel2023-03-281-10/+2
|
* GraphUI: Label: Enforce bottom-left origin @ 0/0 for good drag-zoom ↵Sven Gothel2023-03-281-3/+4
| | | | experience (working sticky edge)
* GraphUI: Shape: Fix setTransform(): Scale around center and rotate around ↵Sven Gothel2023-03-281-26/+40
| | | | pivot-point or center; Assume Shape origin to be bottom-left for drag-zoom. Removes questionable scale-hack.
* GraphUI: Shape.dispatchMouseEvent(): Rename dsx/dsxy -> sx/sy as they ↵Sven Gothel2023-03-281-7/+7
| | | | denominate whole scale-factor, not their delta
* Graph RenderState: Add setColorStatic(float[]) variantSven Gothel2023-03-281-0/+3
|
* Graph Font.GlyphVisitor*: Pass 'char symbol' to visitor, passing full ↵Sven Gothel2023-03-285-11/+12
| | | | text-processing information
* Graph Font.Glyph: Add getFont() (the owner)Sven Gothel2023-03-282-0/+8
|
* GraphUI Demos: Remove unused GPUUISceneTextAnim01Sven Gothel2023-03-281-303/+0
|
* Graph FontScale: Simplify names, fix API doc and add unit testSven Gothel2023-03-2811-36/+173
|
* Graph Font: Rename getBBox*() -> getBounds*() to preserve a common semantic nameSven Gothel2023-03-276-17/+17
|
* Graph: Fix Font.getGlyphBoundsFU(): FU of advance is requested here.Sven Gothel2023-03-271-1/+1
| | | | Regression from commit a5d593478afa2298282a0624b2490fde84c3a292
* Quaternion: Shorten toString() type nameSven Gothel2023-03-271-1/+2
|
* GraphUI Demos: (Text) Type Animation ...Sven Gothel2023-03-222-0/+697
|
* UISceneDemo0[01]: Useless minor editing (can't move animator lower as sadly ↵Sven Gothel2023-03-222-2/+1
| | | | used in our window dtor event handler)
* Add Vec2f and Vec3f, to support a more simple OO vector-math usage, e.g. for ↵Sven Gothel2023-03-222-0/+528
| | | | | | | | animation etc Implementation borrowed my 'gfxbox2' C++ project <https://jausoft.com/cgit/cs_class/gfxbox2.git/tree/include/pixel/pixel3f.hpp#n29> and its layout from OpenAL's Vec3f.
* Graph: Remove OutlineShape.Visitor* as being replaced by Font.GlyphVisitorSven Gothel2023-03-222-24/+1
| | | | (We may ressurect them if needed for a future use case)
* Graph Font Processing: Use Font.GlyphVisitor instead of ↵Sven Gothel2023-03-227-35/+90
| | | | OutlineShape.Visitor, allowing to use the Glyph (information).
* Graph Font Processing: Produce a 'whitespace' Glyph w/ an OutlineShape (box) ↵Sven Gothel2023-03-223-5/+76
| | | | to allow better handling of such non-contour symbols.
* GraphUI Demos: Adopt to shapesSharpness -> oshapesSharpness renameSven Gothel2023-03-227-7/+7
|
* GraphUI Shape: Make access private where possible & reasonable, synchronize ↵Sven Gothel2023-03-226-42/+55
| | | | dirty and in draw(..) the whole dirty-validate() region.draw()
* GraphUI Scene: getStatusText(..) drop unset values, be more semantic ↵Sven Gothel2023-03-221-5/+19
| | | | sensitive (msaa -> smsaa)
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-04 02:44:45 +0000