diff --git a/cplusplus/examples/test.cpp b/cplusplus/examples/test.cpp
new file mode 100644
index 00000000..fc6b1a8a
--- /dev/null
+++ b/cplusplus/examples/test.cpp
@@ -0,0 +1,390 @@
+/* Test the C++ API. 
+ *
+ * This isn't a full test suite, look in the Python area for that. This is
+ * just supposed to check that the C++ binding is working. 
+ *
+ * compile with:
+ *
+ *      g++ -g -Wall test.cpp `pkg-config vips-cpp --cflags --libs`
+ *
+ * run with:
+ *
+ * 	./a.out ~/pics/k2.jpg ~/pics/shark.jpg --vips-leak
+ * 	valgrind --leak-check=yes ./a.out ~/pics/k2.jpg ~/pics/shark.jpg 
+ * 	rm x.tif
+ *
+ */
+
+/*
+#define VIPS_DEBUG
+#define VIPS_DEBUG_VERBOSE
+ */
+
+#include <stdlib.h>
+
+#include <vips/vips8>
+
+using namespace vips;
+
+bool
+equal_vector( std::vector<double> a, std::vector<double> b )
+{
+	for( unsigned int i = 0; i < a.size(); i++ ) 
+		if( fabs( a[i] - b[i] ) > 0.001 ) {
+			printf( "vectors differ at %d: should be [", i );
+			for( unsigned int i = 0; i < a.size(); i++ ) {
+				if( i > 0 )
+					printf( ", " ); 
+				printf( "%g", a[i] );
+			}
+			printf( "], is [" );
+			for( unsigned int i = 0; i < a.size(); i++ ) {
+				if( i > 0 )
+					printf( ", " ); 
+				printf( "%g", a[i] );
+			}
+			printf( "]\n" );
+
+			return( false );
+		}
+
+	return( true );
+}
+
+bool
+equal_double( double a, double b )
+{
+	if( fabs( a - b ) > 0.001 ) {
+		printf( "doubles differ: should be %g, is %g\n", a, b );
+		return( false );
+	}
+
+	return( true ); 
+}
+
+/* We can't do this with a template, I think we'd need partially-parameterised
+ * template, which is C++11 only.
+ */
+
+/* Only test a few points and only test uchar: we are just testing the C++ 
+ * overloads, we rely on the python test suite for testing the underlying 
+ * vips operators.
+ */
+#define TEST_BINARY( OPERATOR ) \
+void \
+test_binary_##OPERATOR( VImage left, VImage right ) \
+{ \
+	for( int x = 10; x < 30; x += 10 ) {  \
+		std::vector<double> p_left = left.getpoint( x, x );  \
+		std::vector<double> p_right = right.getpoint( x, x );  \
+		std::vector<double> p_result =  \
+			OPERATOR<std::vector<double>,  \
+				std::vector<double>,  \
+				std::vector<double> >(p_left, p_right ); \
+ 		\
+		VImage im_result; \
+		std::vector<double> p_im_result; \
+ 		\
+		/* test: image = image OP image \
+		 */ \
+		im_result = OPERATOR<VImage, VImage, VImage>( left, right ); \
+		p_im_result = im_result.getpoint( x, x ); \
+ 		\
+		if( !equal_vector( p_result, p_im_result ) ) { \
+			printf( #OPERATOR \
+				"(VImage, VImage) failed at (%d, %d)\n", \
+				x, x ); \
+			abort(); \
+		} \
+ 		\
+		/* test: image = image OP vec \
+		 */ \
+		im_result = \
+			OPERATOR<VImage, \
+				VImage, std::vector<double> >( left, p_right );\
+		p_im_result = im_result.getpoint( x, x ); \
+ 		\
+		if( !equal_vector( p_result, p_im_result ) ) { \
+			printf( #OPERATOR \
+				"(VImage, vector) failed at (%d, %d)\n", \
+				x, x ); \
+			abort(); \
+		} \
+ 		\
+		/* test: image = vec OP image \
+		 */ \
+		im_result = \
+			OPERATOR<VImage, std::vector<double>, \
+				VImage>( p_left, right );  \
+		p_im_result = im_result.getpoint( x, x ); \
+ 		\
+		if( !equal_vector( p_result, p_im_result ) ) { \
+			printf( #OPERATOR \
+				"(vector, VImage) failed at (%d, %d)\n", \
+				x, x ); \
+			abort(); \
+		} \
+ 		\
+		/* test: image = image OP double \
+		 */ \
+		for( unsigned int i = 0; i < p_right.size(); i++ ) { \
+			im_result = \
+				OPERATOR<VImage,  \
+					VImage, double>( left, p_right[i] ); \
+			p_im_result = im_result.getpoint( x, x ); \
+ 			\
+			if( !equal_double( p_result[i], p_im_result[i] ) ) { \
+				printf( #OPERATOR \
+					"(VImage, double) failed at " \
+					"(%d, %d)\n", \
+					x, x ); \
+				abort(); \
+			} \
+		} \
+ 		\
+		/* test: image = double OP image  \
+		 */ \
+		for( unsigned int i = 0; i < p_left.size(); i++ ) { \
+			im_result = \
+				OPERATOR<VImage, \
+					double, VImage>( p_left[i], right ); \
+			p_im_result = im_result.getpoint( x, x ); \
+ 			\
+			if( !equal_double( p_result[i], p_im_result[i] ) ) { \
+				printf( #OPERATOR \
+					"(double, VImage) failed at " \
+					"(%d, %d)\n", \
+					x, x ); \
+				abort(); \
+			} \
+		} \
+	} \
+}
+
+// eg. double = double + double
+// or image = double + image
+template <typename A, typename B, typename C> 
+A test_add( B left, C right )
+{
+	return( left + right );
+}
+
+template <typename T>
+std::vector<T> operator+(std::vector<T> &v1, const std::vector<T> &v2) 
+{
+	std::vector<T> result( v1.size() );
+
+	for( unsigned int i = 0; i < v1.size(); i++ )
+		result[i] = v1[i] + v2[i];
+
+	return( result ); 
+}
+
+TEST_BINARY( test_add );
+
+template <typename A, typename B, typename C> 
+A test_subtract( B left, C right )
+{
+	return( left - right );
+}
+
+template <typename T>
+std::vector<T> operator-(std::vector<T> &v1, const std::vector<T> &v2) 
+{
+	std::vector<T> result( v1.size() );
+
+	for( unsigned int i = 0; i < v1.size(); i++ )
+		result[i] = v1[i] - v2[i];
+
+	return( result ); 
+}
+
+TEST_BINARY( test_subtract );
+
+template <typename A, typename B, typename C> 
+A test_multiply( B left, C right )
+{
+	return( left * right );
+}
+
+template <typename T>
+std::vector<T> operator*(std::vector<T> &v1, const std::vector<T> &v2) 
+{
+	std::vector<T> result( v1.size() );
+
+	for( unsigned int i = 0; i < v1.size(); i++ )
+		result[i] = v1[i] * v2[i];
+
+	return( result ); 
+}
+
+TEST_BINARY( test_multiply );
+
+template <typename A, typename B, typename C> 
+A test_divide( B left, C right )
+{
+	return( left / right );
+}
+
+template <typename T>
+std::vector<T> operator/(std::vector<T> &v1, const std::vector<T> &v2) 
+{
+	std::vector<T> result( v1.size() );
+
+	for( unsigned int i = 0; i < v1.size(); i++ )
+		result[i] = v1[i] / v2[i];
+
+	return( result ); 
+}
+
+TEST_BINARY( test_divide );
+
+int
+main( int argc, char **argv )
+{
+	GOptionContext *context;
+	GOptionGroup *main_group;
+	GError *error = NULL;
+
+	if( VIPS_INIT( argv[0] ) )
+		vips_error_exit( NULL ); 
+
+	context = g_option_context_new( "" ); 
+
+	main_group = g_option_group_new( NULL, NULL, NULL, NULL, NULL );
+	g_option_context_set_main_group( context, main_group );
+	g_option_context_add_group( context, vips_get_option_group() );
+
+	if( !g_option_context_parse( context, &argc, &argv, &error ) ) {
+		if( error ) {
+			fprintf( stderr, "%s\n", error->message );
+			g_error_free( error );
+		}
+
+		vips_error_exit( NULL );
+	}
+
+	g_option_context_free( context );
+
+	VImage left = VImage::new_from_file( argv[1] ); 
+	VImage right = VImage::new_from_file( argv[2] ); 
+
+{ 
+	printf( "testing constant args ...\n" ); 
+
+	double a[] = { 1.0, 2.0, 3.0 }; 
+	double b[] = { 4.0, 5.0, 6.0 }; 
+
+	std::vector<double> avec( a, a + VIPS_NUMBER( a ) ); 
+	std::vector<double> bvec( b, b + VIPS_NUMBER( b ) ); 
+
+	VImage out = left.linear( avec, bvec ); 
+
+	out.write_to_file( "x.tif" );
+}
+
+{
+	printf( "testing operator overloads ...\n" ); 
+	test_binary_test_add( left, right );
+	test_binary_test_subtract( left, right );
+	test_binary_test_multiply( left, right );
+	test_binary_test_divide( left, right );
+
+	VImage band_one = left[1];
+
+	std::vector<double> point = left(0, 0);
+}
+
+{
+	// write to a formatted memory buffer
+	printf( "testing formatted memory write ...\n" ); 
+
+	size_t size;
+	void *buf;
+	left.write_to_buffer( ".png", &buf, &size );
+	printf( "written to memory %p in png format, %zd bytes\n", buf, size );
+
+	// load from the formatted memory area
+	VImage im = VImage::new_from_buffer( buf, size, "" );
+	printf( "loaded from memory, %d x %d pixel image\n", 
+		im.width(), im.height() ); 
+
+	// write back to a file
+	im.write_to_file( "x.tif" );
+	printf( "written back to x.tif\n" ); 
+
+	g_free( buf ); 
+}
+
+{
+	// write to a vanilla memory buffer
+	printf( "testing memory array write ...\n" ); 
+
+	size_t size;
+	void *buf;
+	buf = left.write_to_memory( &size );
+	printf( "written to memory %p as an array, %zd bytes\n", buf, size );
+
+	// load from the memory array
+	VImage im = VImage::new_from_memory( buf, size, 
+		left.width(), left.height(), left.bands(), left.format() );
+	printf( "loaded from memory array, %d x %d pixel image\n", 
+		im.width(), im.height() ); 
+
+	// write back to a file
+	im.write_to_file( "x.tif" );
+	printf( "written back to x.tif\n" ); 
+
+	g_free( buf ); 
+}
+
+{
+	printf( "testing double return from operation ...\n" ); 
+
+	double avg = left.avg(); 
+
+	printf( "left.avg() = %g\n", avg ); 
+}
+
+{ 
+	printf( "testing optional enum args ...\n" ); 
+
+	VImage out = left.embed( 10, 10, 1000, 1000, 
+		VImage::option()->set( "extend", "copy" ) );
+
+	out.write_to_file( "x.tif" );
+}
+
+{ 
+	printf( "testing multiple image return ...\n" ); 
+
+	VImage rows; 
+	VImage cols = left.profile( &rows );
+	rows.write_to_file( "x.tif" );
+	cols.write_to_file( "x.tif" );
+}
+
+{ 
+	printf( "testing interpolators ...\n" ); 
+
+	VInterpolate interp = VInterpolate::new_from_name( "nohalo" );
+
+	VImage out; 
+
+	out = left.resize( 0.2, 
+		VImage::option()->set( "interpolate", interp ) );
+	out.write_to_file( "x.tif" );
+}
+
+{ 
+	printf( "testing new_from_image() ...\n" ); 
+
+	VImage out = left.new_from_image( 128 );
+
+	out.write_to_file( "x.tif" );
+}
+
+	printf( "all tests passed\n" ); 
+
+        return( 0 );
+}
diff --git a/cplusplus/examples/try.cpp b/cplusplus/examples/try.cpp
deleted file mode 100644
index c7c77049..00000000
--- a/cplusplus/examples/try.cpp
+++ /dev/null
@@ -1,93 +0,0 @@
-/* 
- * compile with:
- *
- *      g++ -g -Wall try.cc `pkg-config vips-cpp --cflags --libs`
- *
- */
-
-#define DEBUG
-
-#include <vips/vips8>
-
-using namespace vips8;
-
-int
-main( int argc, char **argv )
-{
-	GOptionContext *context;
-	GOptionGroup *main_group;
-	GError *error = NULL;
-
-	if( vips_init( argv[0] ) )
-		vips_error_exit( NULL ); 
-
-	context = g_option_context_new( "" ); 
-
-	main_group = g_option_group_new( NULL, NULL, NULL, NULL, NULL );
-	g_option_context_set_main_group( context, main_group );
-	g_option_context_add_group( context, vips_get_option_group() );
-
-	if( !g_option_context_parse( context, &argc, &argv, &error ) ) {
-		if( error ) {
-			fprintf( stderr, "%s\n", error->message );
-			g_error_free( error );
-		}
-
-		vips_error_exit( NULL );
-	}
-
-
-{ 
-	VImage in = VImage::new_from_file( argv[1], 
-		VImage::option()->set( "access", VIPS_ACCESS_SEQUENTIAL_UNBUFFERED ) ); 
-	double avg;
-
-	avg = in.avg(); 
-
-	printf( "avg = %g\n", avg ); 
-}
-
-{ 
-	VImage in = VImage::new_from_file( argv[1], 
-		VImage::option()->set( "access", VIPS_ACCESS_SEQUENTIAL_UNBUFFERED ) ); 
-
-	VImage out = in.embed( 10, 10, 1000, 1000, 
-		VImage::option()->set( "extend", VIPS_EXTEND_BACKGROUND )->
-		set( "background", 128 ) );
-
-	out.write_to_file( "embed.jpg" );
-}
-
-{ 
-	VImage in = VImage::new_from_file( argv[1], 
-		VImage::option()->set( "access", VIPS_ACCESS_SEQUENTIAL_UNBUFFERED ) ); 
-	double a[] = { 1.0, 2.0, 3.0 }; 
-	double b[] = { 4.0, 5.0, 6.0 }; 
-
-	std::vector<double> avec( a, a + VIPS_NUMBER( a ) ); 
-	std::vector<double> bvec( b, b + VIPS_NUMBER( b ) ); 
-
-	VImage out = in.linear( avec, bvec ); 
-
-	out.write_to_file( "linear.jpg" );
-}
-
-{ 
-	VImage in = VImage::new_from_file( argv[1], 
-		VImage::option()->set( "access", VIPS_ACCESS_SEQUENTIAL_UNBUFFERED ) ); 
-	VImage out = in.linear( 1, 2 ); 
-
-	out.write_to_file( "linear1.jpg" );
-}
-
-{ 
-	VImage in = VImage::new_from_file( argv[1] ); 
-	VImage out = in.new_from_image( 128 );
-
-	out.write_to_file( "const.jpg" );
-}
-
-	vips_shutdown();
-
-        return( 0 );
-}