update README

performance comparison/philipphenkel/README.md

@@ -1,19 +1,5 @@
-# Environment
-CPU: intel i7 4930k<br>
-OS: windows 7 enterprise sp1, 64 bit<br>
-Compiler: visual c++ 2015<br>
-Compiler Option: release, x64, full optimization<br>
-Iteration: 600000 (because of memory leak, test [test_henkel_ctor_and_dtor](../header/henkel.hpp) by yourself) and 10000000<br>
-Thread Count: 12<br>
-Boost: 1.60.0
-# Result
-	henkel
-		each spends 420012 nanoseconds on test_henkel_ctor_and_dtor (600000 iteration)
-		each spends 51897 nanoseconds on test_henkel_all_N (10000000 iteration)
-		each spends 1803 nanoseconds on test_henkel_billion (10000000 iteration)
-# Compare To Fdhvdu
-henkel's threadpool can only compare to Fdhvdu's CThreadPool. (because CThreadPool_Ret cannot use "all" function)<br>
-About construction and destruction time, henkel's threadpool is slower than Fdhvdu's threadpool<br>
-About waiting all job to complete, henkel's threadpool is slower than Fdhvdu's CThreadPool<br>
-About given a number of jobs, henkel's threadpool is slower than Fdhvdu's CThreadPool<br><br>
-As a result, you should use Fdhvdu's CThreadPool in any condition
+# Compile command
+	g++ -std=c++14 -O3 test.cpp ../src/philipphenkel.cpp -lpthread -lboost_thread -lboost_system
+# Warning
+Do not use this.<br>
+It gets stuck when runing out_100000.