Improve Color8 documentation

This documents in which situations Color8 is most suited and its
precision limitations.

1 files changed, 2 insertions, 1 deletions

diff --git a/modules/gdscript/doc_classes/@GDScript.xml b/modules/gdscript/doc_classes/@GDScript.xml
index 718bf3f6d4..85ff5080f0 100644
--- a/modules/gdscript/doc_classes/@GDScript.xml
+++ b/modules/gdscript/doc_classes/@GDScript.xml
@@ -18,12 +18,13 @@
 			<param index="2" name="b8" type="int" />
 			<param index="3" name="a8" type="int" default="255" />
 			<description>
-				Returns a [Color] constructed from red ([param r8]), green ([param g8]), blue ([param b8]), and optionally alpha ([param a8]) integer channels, each divided by [code]255.0[/code] for their final value.
+				Returns a [Color] constructed from red ([param r8]), green ([param g8]), blue ([param b8]), and optionally alpha ([param a8]) integer channels, each divided by [code]255.0[/code] for their final value. Using [method Color8] instead of the standard [Color] constructor is useful when you need to match exact color values in an [Image].
 				[codeblock]
 				var red = Color8(255, 0, 0)             # Same as Color(1, 0, 0).
 				var dark_blue = Color8(0, 0, 51)        # Same as Color(0, 0, 0.2).
 				var my_color = Color8(306, 255, 0, 102) # Same as Color(1.2, 1, 0, 0.4).
 				[/codeblock]
+				[b]Note:[/b] Due to the lower precision of [method Color8] compared to the standard [Color] constructor, a color created with [method Color8] will generally not be equal to the same color created with the standard [Color] constructor. Use [method Color.is_equal_approx] for comparisons to avoid issues with floating-point precision error.
 			</description>
 		</method>
 		<method name="assert">