add link to wigner-benchmark

0382 · May 8, 2024 · ab26d3b · ab26d3b
1 parent 738dd4c
commit ab26d3b
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 3 deletions.
diff --git a/README.md b/README.md
@@ -22,7 +22,7 @@ double x = wigner_6j(dj1, dj2, dj3, dj4, dj5, dj6);
 ### Limitation 
 
 For quite large quantum number, the package will give wrong answer, since it use float number arithmetic.
-However, it is trustworthy for most real real world numerical calculation system. For `f9j`, it works at least about `Jmax = 25`.
+Please see [wigner-benchmark](https://github.com/0382/wigner-benchmark) for the error estimate and performance benchmark.
 
 ## API
 

diff --git a/README_zh.md b/README_zh.md
@@ -21,8 +21,8 @@ double x = wigner_6j(dj1, dj2, dj3, dj4, dj5, dj6);
 
 ### 不足
 
-由于使用了浮点数计算，这个包在较大的角动量时会给出错误的结果。不过在大部分现实体系的数值计算中，它还是值得信赖的。
-例如，对于`f9j`计算，它在至少`Jmax = 25`时还是正确的。
+由于使用了浮点数计算，这个包在较大的角动量时会给出错误的结果。
+详细的误差和性能测试，请看：[wigner-benchmark](https://github.com/0382/wigner-benchmark)。
 
 ## 提供的函数
 ```cpp