diff --git a/headers/private/interface/moreUTF8.h b/headers/private/interface/moreUTF8.h
index c45948caf9..8fee6e25f2 100644
--- a/headers/private/interface/moreUTF8.h
+++ b/headers/private/interface/moreUTF8.h
@@ -1,22 +1,20 @@
 #ifndef __MOREUTF8
 #define __MOREUTF8
 
+#include <stdio.h>
 
 static inline bool
 IsInsideGlyph(uchar ch)
 {
 	return (ch & 0xC0) == 0x80;
+//	return (ch & 0x80);
 }
 
-
 static inline uint32
-UTF8NextCharLen(const char *text)
+UTF8NextCharLenUnsafe(const char *text)
 {
 	const char *ptr = text;
-	
-	if (ptr == NULL || *ptr == 0)
-		return 0;
-	
+
 	do {
 		ptr++;
 	} while (IsInsideGlyph(*ptr));
@@ -24,6 +22,15 @@ UTF8NextCharLen(const char *text)
 	return ptr - text;
 }
 
+static inline uint32
+UTF8NextCharLen(const char *text)
+{
+	if (text == NULL || *text == 0)
+		return 0;
+
+	return UTF8NextCharLenUnsafe(text);
+}
+
 static inline uint32
 UTF8PreviousCharLen(const char *text, const char *limit)
 {
@@ -41,17 +48,39 @@ UTF8PreviousCharLen(const char *text, const char *limit)
 	return text - ptr;
 }
 
+// TODO: use this function in other places of this file...
+static inline uint32
+count_utf8_bytes(uchar ch)
+{
+	// the number of high bits set until the first
+	// unset bit determine the count of bytes used for
+	// this glyph from this byte on
+	uchar bit = 1 << 7;
+	uint32 count = 1;
+	if (ch & bit) {
+		bit = bit >> 1;
+		while (ch & bit) {
+			count++;
+			bit = bit >> 1;
+		}
+	}
+	return count;
+}
+
 static inline uint32
 UTF8CountBytes(const char *text, uint32 numChars)
 {
-	const char *ptr = text;
-
-	while (numChars) {
-		ptr += UTF8NextCharLen(ptr);
-		numChars--;
+	if (text) {
+		// iterate over numChars glyphs incrementing ptr by the
+		// number of bytes for each glyph, which is encoded in
+		// the first byte of any glyph.
+		const char *ptr = text;
+		while (numChars--) {
+			ptr += count_utf8_bytes(*ptr);
+		}
+		return ptr - text;
 	}
-
-	return ptr - text;
+	return 0;
 }
 
 static inline uint32
diff --git a/src/servers/app/ServerFont.cpp b/src/servers/app/ServerFont.cpp
index 4e92541300..268904fe54 100644
--- a/src/servers/app/ServerFont.cpp
+++ b/src/servers/app/ServerFont.cpp
@@ -294,7 +294,7 @@ ServerFont::GetGlyphShapes(const char charArray[], int32 numChars) const
 	if (!face)
 		return NULL;
 	
-	FT_Set_Char_Size(face, 0, int32(fSize) * 64, 72, 72);
+	FT_Set_Char_Size(face, 0, int32(fSize * 64), 72, 72);
 	
 	Angle rotation(fRotation);
 	Angle shear(fShear);
@@ -344,7 +344,7 @@ ServerFont::GetEscapements(const char charArray[], int32 numChars,
 	if (!face)
 		return NULL;
 	
-	FT_Set_Char_Size(face, 0, int32(fSize) * 64, 72, 72);
+	FT_Set_Char_Size(face, 0, int32(fSize * 64), 72, 72);
 	
 	Angle rotation(fRotation);
 	Angle shear(fShear);
@@ -408,7 +408,7 @@ ServerFont::GetEscapements(const char charArray[], int32 numChars,
 	if (!face)
 		return false;
 
-	FT_Set_Char_Size(face, 0, int32(fSize) * 64, 72, 72);
+	FT_Set_Char_Size(face, 0, int32(fSize * 64), 72, 72);
 
 	// UTF8 handling...this can probably be smarter
 	// Here is what I do in the AGGTextRenderer to handle UTF8...
@@ -428,13 +428,14 @@ ServerFont::GetEscapements(const char charArray[], int32 numChars,
 
 		uint16* glyphIndex = (uint16*)convertedBuffer;
 		// just to be sure
-		numChars = convertedLength / sizeof(uint16);
+		numChars = min_c((uint32)numChars, convertedLength / sizeof(uint16));
 
 		for (int i = 0; i < numChars; i++) {
 			FT_Load_Char(face, glyphIndex[i], FT_LOAD_NO_BITMAP);
-//			widthArray[i] = float(face->glyph->metrics.width / 64) / fSize;
-			widthArray[i] = ((float)face->glyph->metrics.horiAdvance / 64.0) / fSize;
-			widthArray[i] += is_white_space(glyphIndex[i]) ? delta.space : delta.nonspace;
+			if (face->glyph) {
+				widthArray[i] = ((float)face->glyph->metrics.horiAdvance / 64.0) / fSize;
+				widthArray[i] += is_white_space(glyphIndex[i]) ? delta.space : delta.nonspace;
+			}
 		}
 	}
 	delete[] convertedBuffer;