diff --git a/ref_app/src/app/benchmark/images/app_benchmark_pi_agm.docx b/ref_app/src/app/benchmark/images/app_benchmark_pi_agm.docx
deleted file mode 100644
index b07c49176..000000000
Binary files a/ref_app/src/app/benchmark/images/app_benchmark_pi_agm.docx and /dev/null differ
diff --git a/ref_app/src/app/benchmark/images/app_benchmark_pi_agm.pdf b/ref_app/src/app/benchmark/images/app_benchmark_pi_agm.pdf
deleted file mode 100644
index 733c585fd..000000000
Binary files a/ref_app/src/app/benchmark/images/app_benchmark_pi_agm.pdf and /dev/null differ
diff --git a/ref_app/src/app/benchmark/readme.md b/ref_app/src/app/benchmark/readme.md
index daae809f6..199d9a3f4 100644
--- a/ref_app/src/app/benchmark/readme.md
+++ b/ref_app/src/app/benchmark/readme.md
@@ -48,13 +48,11 @@ $\pi$ using a Gauss AGM method with help
 from the [`decwide_t`](https://github.com/ckormanyos/real-time-cpp/blob/master/ref_app/src/math/wide_decimal/decwide_t.h)
 template class.
 
-Differing performance classes are compared for
-the $8$-bit MICROCHIP(R) AVR controller of the Arduino
+There are strikingly differing performance classes
+for the $8$-bit MICROCHIP(R) AVR controller of the Arduino
 and the $32$-bit ARM(R) 8 controller
 of the BeagleBone Black Edition, Rev. C.
 
-![](images/app_benchmark_pi_agm.pdf)
-
 The $\pi$ calculation requires approximately
 $470~\text{ms}$ and $1.5~\text{ms}$,
 respectively, on these two microcontroller systems.