From: Steinar H. Gunderson <sgunderson@bigfoot.com>
Date: Thu, 3 Apr 2014 20:54:13 +0000 (+0200)
Subject: Loosen up the 0.499 vs. 0.501 subpixel resample test.
X-Git-Tag: 1.1~5
X-Git-Url: https://git.sesse.net/?p=movit;a=commitdiff_plain;h=233ddaa1c94e8c11ef73740b4d9c1aa31a97dd5c

Loosen up the 0.499 vs. 0.501 subpixel resample test.

Seemingly these limits were a bit too tight for something that's
actually supposed to be approximate.
---

diff --git a/resample_effect_test.cpp b/resample_effect_test.cpp
index 9a28013..e90f19e 100644
--- a/resample_effect_test.cpp
+++ b/resample_effect_test.cpp
@@ -334,11 +334,11 @@ TEST(ResampleEffectTest, ReadHalfPixelFromLeftAndScale) {
 	// Check that we are (almost) the same no matter the rounding.
 	ASSERT_TRUE(resample_effect->set_float("left", 0.499f));
 	tester.run(out_data, GL_RED, COLORSPACE_sRGB, GAMMA_LINEAR);
-	expect_equal(expected_data, out_data, dst_width, 1);
+	expect_equal(expected_data, out_data, dst_width, 1, 1.5f / 255.0f, 0.4f / 255.0f);
 
 	ASSERT_TRUE(resample_effect->set_float("left", 0.501f));
 	tester.run(out_data, GL_RED, COLORSPACE_sRGB, GAMMA_LINEAR);
-	expect_equal(expected_data, out_data, dst_width, 1);
+	expect_equal(expected_data, out_data, dst_width, 1, 1.5f / 255.0f, 0.4f / 255.0f);
 }
 
 TEST(ResampleEffectTest, Zoom) {