Commit 763e0964 authored by Thomas Steinreiter's avatar Thomas Steinreiter
Browse files

Update README.md

parent 7e4af317
......@@ -104,7 +104,7 @@ Compiler: Intel(R) 64, Version 17.0.0.109 Build 20160721 (MKL 2017.0) on Windows
Be aware the chart has logarithmic scale.
The benchmarks show some interesting facts:
The benchmarks shows some interesting facts:
* The *serial IKJ* implementation is brutally faster than IJK (~30x @2048)
* *Serial IKJ* is the fastest smaller sizes (2-16). *MKL Blas* is the fastest for larger sizes (32-2048).
* The *Parallel OpenMP IKJ implementation* is never the fastest. **Manually parallelizing matrix multiplication is not recommended!**
\ No newline at end of file
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment