diff --git a/tests/float/string_format_fp30.py b/tests/float/string_format_fp30.py
new file mode 100644
index 000000000..77b2a5288
--- /dev/null
+++ b/tests/float/string_format_fp30.py
@@ -0,0 +1,41 @@
+def test(fmt, *args):
+    print('{:8s}'.format(fmt) + '>' +  fmt.format(*args) + '<')
+
+test("{:10.4}", 123.456)
+test("{:10.4e}", 123.456)
+test("{:10.4e}", -123.456)
+#test("{:10.4f}", 123.456)
+#test("{:10.4f}", -123.456)
+test("{:10.4g}", 123.456)
+test("{:10.4g}", -123.456)
+test("{:10.4n}", 123.456)
+test("{:e}", 100)
+test("{:f}", 200)
+test("{:g}", 300)
+
+test("{:10.4E}", 123.456)
+test("{:10.4E}", -123.456)
+#test("{:10.4F}", 123.456)
+#test("{:10.4F}", -123.456)
+test("{:10.4G}", 123.456)
+test("{:10.4G}", -123.456)
+
+test("{:06e}", float("inf"))
+test("{:06e}", float("-inf"))
+test("{:06e}", float("nan"))
+
+# The following fails right now
+#test("{:10.1}", 0.0)
+
+print("%.0f" % (1.750000 % 0.08333333333))
+# Below isn't compatible with single-precision float
+#print("%.1f" % (1.750000 % 0.08333333333))
+#print("%.2f" % (1.750000 % 0.08333333333))
+#print("%.12f" % (1.750000 % 0.08333333333))
+
+# tests for errors in format string
+
+try:
+    '{:10.1b}'.format(0.0)
+except ValueError:
+    print('ValueError')