Hajime Hoshi
abbb929148
ebiten: Bug fix: Wrong uniform type was used
...
This change also includes PanicOnErrorAtImageAt, which panics when
an error is detected at (*Image).At. This function is only for testing.
Fixes #1349
2020-09-17 20:52:32 +09:00
Hajime Hoshi
442849e120
ebiten: Skip some tests on Wasm
...
They are flagile and sometimes causes crash on GitHub Actions workflow
tests.
2020-08-24 04:10:24 +09:00
Hajime Hoshi
dfc23ee557
ebiten: Skip tests in the short mode
2020-08-19 23:11:54 +09:00
Hajime Hoshi
80d04732c2
Revert "ebiten: Skip heavy tests even on Wasm"
...
This reverts commit 0ee70df466
.
Reason: This didn't affect the results
2020-08-19 23:00:45 +09:00
Hajime Hoshi
9c85b20817
ebiten: Bug fix: Test compilation error
2020-08-19 22:41:36 +09:00
Hajime Hoshi
0ee70df466
ebiten: Skip heavy tests even on Wasm
...
Updates #1313
2020-08-19 22:33:01 +09:00
Hajime Hoshi
213fbd283a
ebiten: Bug fix: Skip the tests correctly
2020-08-19 22:07:15 +09:00
Hajime Hoshi
bebad644ad
ebiten: Refactoring
2020-08-18 23:31:19 +09:00
Hajime Hoshi
a48c0d9f2c
ebiten: Skip some tests on Go1.12 Wasm
2020-08-17 04:15:33 +09:00
Hajime Hoshi
c032d61f93
ebiten: Rename a test case
...
Updates #1269
2020-07-28 14:04:05 +09:00
Hajime Hoshi
da98e67345
mipmap: Bug fix: Wrong level calculation when the indices are empty
...
Fixes #1269
2020-07-28 14:02:51 +09:00
Hajime Hoshi
d22026d366
ebiten: Better test values for CompositeModeMultiply
2020-07-18 21:57:02 +09:00
Hajime Hoshi
6ccb614b08
graphicsdriver/metal: Implement CompositeModeMultiply
...
Updates #410
2020-07-18 21:51:57 +09:00
Hajime Hoshi
f927e09f56
mipmap: Unify DrawImage and DrawTriangles
...
Fixes #909
2020-07-16 02:31:17 +09:00
Hajime Hoshi
a4e669cc39
ebiten: Avoid using SubImage for stretching test
2020-07-15 23:49:26 +09:00
Hajime Hoshi
2a9fd5ef13
mipmap: Avoid using the vertex backend when GopherJS is not used
...
The backend logic is especially for GopherJS. As the backend logic
tends to consume a lot of memory, let's avoid this when possible.
Updates #797
2020-07-14 12:16:37 +09:00
Hajime Hoshi
cec32d4f72
ebiten: Add tests to modify pixels after ReplacePixels
...
Updates #1222
2020-07-03 03:25:47 +09:00
Hajime Hoshi
36515eb1f5
Revert "ebiten: Make ebiten.Image and buffered.Image 1:1"
...
This reverts commit 620981a09a
.
Fixes #1218
Updates #896
2020-06-28 01:01:39 +09:00
Hajime Hoshi
006b518929
ebiten: Skip test failing on macOS
2020-06-27 20:24:32 +09:00
Hajime Hoshi
5c4740c294
ebiten: Improve TestImageColorMCopy to test copying actually works
2020-06-25 22:49:37 +09:00
Hajime Hoshi
198621404f
buffered: Bug fix: Pixel data was not invalidated after DrawImage/Triangles
2020-06-25 13:11:19 +09:00
Hajime Hoshi
20b5be0886
graphicsdriver: Bug fix: Tie-breaking when picking a texel
...
When a texel is picked on texel-borders of a texture, the behavior
(tie-breaking) depends on GPU and unexpected. This change fixes this
issue by shifting 1/512 [texel] when picking a texel up.
Updates #1212
2020-06-24 20:51:16 +09:00
Hajime Hoshi
484473b6d9
graphicscommand: Span to 1/3 pixels
...
The center of pixels is problematic as the behavior depends on GPU.
In order to avoid this, align the vertices with about 1/3 pixels.
Updates #929
Fixes #1171
2020-06-16 03:56:53 +09:00
Hajime Hoshi
0b94e2e036
ebiten: Updates the tests
...
The tests are still disabled as they don't pass yet.
Updates #1171
2020-06-15 02:51:21 +09:00
Hajime Hoshi
6d25f7eaeb
ebiten: Disable TestImageFloatTranslate
...
It looks like the behavior depends on GPU.
Updates #1171
2020-06-14 18:22:28 +09:00
Hajime Hoshi
d74f7d3a77
shareable: Add gaps between images
...
After this change, each image on the texture atlases has a
transparent border around it.
This change removes the hack to round texels not to violate the
source regions.
Fixes #1194
Updates #1171
2020-06-14 18:05:41 +09:00
Hajime Hoshi
96fa0565e4
buffered: Remove mutex and use sync/atomic for performance
...
This change also enables to remove the optimization at
(*buffered.Image).ReplacePixels.
// This commit w/ the optimization
BenchmarkImageDrawOver-8 60225 19241 ns/op
// This commit w/o the optimization
BenchmarkImageDrawOver-8 66567 17700 ns/op
// The previous w/ the optimization
BenchmarkImageDrawOver-8 62355 19580 ns/op
// The previous w/o the optimization
BenchmarkImageDrawOver-8 54460 22768 ns/op
Updates #1137
2020-06-14 01:16:23 +09:00
Hajime Hoshi
5db4a55a07
ebiten: Check the source images are not disposed at DrawTriangles and DrawTrianglesWithShader
2020-06-13 18:12:21 +09:00
Hajime Hoshi
81159b8538
ebiten: Skip some slow tests on GopherJS
...
Fixes #1138
2020-04-19 02:09:02 +09:00
Hajime Hoshi
1d701577d6
buffered: Use the pending pixels when possible at At
...
(*Image).At can be unnecessarily slow since this tries to get
pixels from GPU. This change reduces the chance to read GPU by
using its pending pixels when possible.
Fixes #1137
2020-04-17 21:34:49 +09:00
Hajime Hoshi
19333c51e4
ebiten: Use internal/testing for tests
2020-04-01 18:35:12 +09:00
Hajime Hoshi
c36d4678a3
mipmap: Bug fix: DrawTriangles with a scale-only color matrix should be optimized
...
Fixes #1101
2020-03-17 02:48:29 +09:00
Hajime Hoshi
875a529708
graphics: Allow ReplacePixels on a sub-image
...
Fixes #980
2020-02-16 22:18:07 +09:00
Hajime Hoshi
aa6fc67736
Revert "graphics: Avoid all copying pixels"
...
This reverts commit c60a32a479
.
Reason: This breaks backward comptibility and it was not obvious how to fix examples.
2019-11-16 02:04:46 +09:00
Hajime Hoshi
c60a32a479
graphics: Avoid all copying pixels
...
This is a breaking change: ReplacePixels now takes the ownership
of the given pixels.
Fixes #983
2019-11-16 01:51:36 +09:00
Hajime Hoshi
1b011e3864
graphics: Bug fix: unexpected panic at Set with an out-of-range position
2019-10-12 02:28:55 +09:00
Hajime Hoshi
2621681a2a
graphics: Add TestImageDrawTrianglesAndMutateArgs
2019-09-18 23:52:11 +09:00
Hajime Hoshi
e434869dd7
graphics: More precise enlarging
...
Before this change, enlarging a 1px image with x4095 scale was not
accurate and edges were missing. This change fixes this issue.
Updates #611
2019-07-31 22:02:39 +09:00
Hajime Hoshi
b210339786
graphics: Use 'negative' mipmap when enlarging a too small image
...
This is a hack to render edges correctly.
This works only when the filter is nearest.
Fixes #611
2019-07-30 23:03:55 +09:00
Hajime Hoshi
d25d184ed7
graphics: Bug fix: wrong for loops in TestImageMipmap*
2019-07-30 19:19:30 +09:00
Hajime Hoshi
db1030a6d4
graphics: Add TestImageMipmapNegativeDet
2019-07-30 14:11:04 +09:00
Hajime Hoshi
0793d35c40
graphicscommand: Improve debug messages
...
Fixes #900
2019-07-19 23:35:13 +09:00
Hajime Hoshi
66b00c6f87
graphics: Remove println
2019-07-15 03:12:53 +09:00
Hajime Hoshi
a7e7e007ca
restorable: Bug fix: PutQuadVertices should take the source image
...
Fixes #898
2019-07-15 03:10:03 +09:00
Hajime Hoshi
4ab97ad250
graphics: Add TestImageZeroSizedMipmap
2019-04-25 10:51:59 +09:00
Hajime Hoshi
3cb9d18fc4
graphics: Bug fix: Too small mipmap
...
Bug: #839
2019-04-07 05:29:46 +09:00
Hajime Hoshi
5ed6565d1d
graphics: Bug fix: SubImage's SubImage's bounds was wrong
2019-02-21 23:44:57 +09:00
Hajime Hoshi
8934f8296f
graphics: Bug fix: Don't touch pixelsToSet when disposing a subimage
...
This change also removes finalizer for subimages since this is not
necessary.
Fixes #823
2019-02-21 23:07:11 +09:00
Hajime Hoshi
cbb0a1da95
graphicsdriver: Refactoring
...
This improved FPS on mobiles a little bit. Maybe now texture2D is
called only when necessary, but not sure.
2019-02-17 17:37:11 +09:00
Hajime Hoshi
1c3113e763
graphics: Add TestImageDrawTrianglesWithSubImage
2019-02-16 22:37:44 +09:00