James Jackson-South
2032ec2430
Merge branch 'master' into GradientBrush
9 years ago
James Jackson-South
fcafc2e044
Merge pull request #548 from woutware/Performance-improvements-1
Performance improvements in the PNG filtering (roughly 20% gain on my machine).
9 years ago
woutware
12ec82b25f
Processed review comments and did similar refactoring to the PNG decoding.
9 years ago
woutware
3659d8690d
Merge branch 'Performance-improvements-1' of https://github.com/woutware/ImageSharp into Performance-improvements-1
9 years ago
woutware
a06600380a
Minor performance refactorings.
Moved the division out of the loop, division is expensive.
9 years ago
James Jackson-South
ab130ec127
Merge branch 'master' into Performance-improvements-1
9 years ago
Anton Firsov
995ccf28dd
Merge pull request #547 from brianpopow/feature/indentify-memallocation-fix
fix for unnecessary memory allocation, if Identify is used to retrieve the metadata of an image only
9 years ago
woutware
555524d473
Removed some overhead in the Bgra32.ToVector4 and PackFromVector4 methods.
It's now also more inline with the Rgba32 implementation.
9 years ago
popow
8db7cd5955
added parameter metadataOnly to InitializeDerivedData: If true, memory allocation will not be done
9 years ago
woutware
5f599e6227
Performance improvements.
Ditched the use of modulo (%). Modulo is slow, and if you use it to just get the lower 8 bits (1 byte), you can just directly cast to byte.
Also moved one if < out of the loop for speed. The less branching the better.
Similarly replaced an if < 128 by cast to sbyte and taking its abs value.
Performance gain in writing to PNG file seems to be roughly 20% (release mode, 1000x1000 bitmap).
In ImageFramesCollectionTests I've set the culture to en-US to ensure getting english exception texts.
9 years ago
Unknown
1cac740739
Merge branch 'master' into GradientBrush
9 years ago
Unknown
c8d0256e85
remove not implemented Polygon Brush from documentation for now.
9 years ago
Unknown
d29b6017ab
implement GradientRepetitionModes
9 years ago
Unknown
75dab524dd
improve performance on distance calculation
9 years ago
Unknown
a6e04024f8
optimization of EllipticGradientBrush
- precalculate anything that's independent of the pixel coordinate
9 years ago
Unknown
e837349952
first implementation of an elliptical gradient brush
9 years ago
popow
bab1728eea
fix for unnecessary memory allocation, if only the metadata of the image will be read
9 years ago
James Jackson-South
05f240874d
Merge pull request #539 from Jjagg/ifc-docs
Improve IImageFrameCollection doc comments
9 years ago
James Jackson-South
103fa5ca94
Merge branch 'master' into ifc-docs
9 years ago
James Jackson-South
e90548cf0d
Merge pull request #541 from carbon/cq-1
Improve CQ, Round 2
9 years ago
James Jackson-South
3c5bb3cc5f
Merge branch 'master' into cq-1
9 years ago
James Jackson-South
1deb2c59e3
Merge pull request #543 from woutware/pr/540
Pr/540 Added IPixel.PackFromBgra32 definition/implementations.
9 years ago
Unknown
94f1698807
implement radial gradient brush.
9 years ago
Unknown
72ae5fa6ba
#542 : refactor to prepare for other gradients
- move to GradientBrushes namespace
- add abstract base class for Gradient Brushes, that implements the GetGradientSegment implementtion and the color picking.
9 years ago
Unknown
d43c70acb5
fix documentation typo
9 years ago
Unknown
d977ecfecc
#542 : rename file to match other files in the solution (include generic)
9 years ago
Unknown
393444c3ff
#542 : fix rounding issues in tests
9 years ago
Unknown
1cd8656b46
#542 : use bigger files for the tests to "hide" rounding issues:
The current implementation lacks accurrancy and prefers to be fast.
For usual color gradients that shouldn't matter: They're quite smooth, for the black-white gradient on very small length's it's easy to spot these rounding errors as they affect whole pixels one can count manually.
9 years ago
Unknown
eafdfead39
fix whitespacing
9 years ago
Unknown
0a642ffea7
#542 : define debuggerDisplay of ColorStop,
9 years ago
Unknown
35d0f56557
#542 : add test creating black-white patterns by gradients
9 years ago
Unknown
28457c8768
cleanup whitespace
9 years ago
Unknown
7a3ba8af3c
#542 : add tests for multi-color gradients
somehow these don't look correct yet, containing hard edges sometimes.
9 years ago
Unknown
94dbb743f7
#542 : fix implementation of non-axial gradients and add tests
the theory lacks color checks yet that should be added, but it produces output images for visual control
9 years ago
Unknown
90293cc6b7
#542 : add test for vertical gradient
9 years ago
Unknown
ce72683866
#542 : reduce test output image sizes
test images don't have to be that big for axial gradients.
It's sufficient to show they're constant across the axis and correct along the axis.
9 years ago
Peter Amrehn
6dd544d5aa
fix code styling issues
9 years ago
James Jackson-South
a351b14359
Add additional tests plus cleanup
9 years ago
James Jackson-South
d4c9a4cd46
Use expressions
9 years ago
James Jackson-South
cbbae51be7
Fix braces
9 years ago
James Jackson-South
b10bc7fefb
Merge remote-tracking branch 'upstream/master' into pr/540
9 years ago
Jason Nelson
3acc080a9c
Merge branch 'master' into cq-1
9 years ago
James Jackson-South
3f444f4637
Merge pull request #540 from woutware/issue-532
Issue-532: Added IPixel.PackFromArgb32 and ToArgb32.
9 years ago
Unknown
1e587bc689
#542 : use struct for ColorStops
as proposed by @antonfirsov : improving readability and memory locality.
9 years ago
James Jackson-South
3440a7decf
Fix System.Drawing.Bridge
9 years ago
Unknown
6d878441e7
#542 : fix test: indices are 0-based, so bottom left pixel is one smaller
9 years ago
Unknown
e1ee9b0b63
FIX bug in BrushApplicator when applying BlendPercentage
thanks @tocsoft for investigation and finding the bug.
He proposed to just replace < by <= in line 78, but that would only shift the problem to values > 1.
Those values should not be used from a semantic point, but it's not forbidden in a float value.
My fix here keeps the <, but adds an else path that uses the original value from scanline[i]. This adds an else to the code (which technically adds a jump after the then-part), but omits the multiplication.
The simple solution @tocsoft proposed might be faster, but should be preferred if and only if BlendPercentage can be guaranteed to be <=1.
9 years ago
Jason Nelson
a58ca2dab9
Format SimdUtilsTests, part 2
9 years ago
Jason Nelson
2745174027
Format SimdUtilTests
9 years ago
Jason Nelson
53045ff357
Format tests
9 years ago