Wilco (Wilco.Dijkstra.delete@this.ntlworld.com) on March 6, 2015 5:04 am wrote:
[...]
> Btw while on the topic of benchmarks, you might want to edit your recent article about CoreMark-Pro
> - the one subtest you claimed is most representative spends almost all of its time (~80%) in one
> tiny string function. So it's trivial to claim 5x gains with library optimizations...

Funny given that EEMBC AndEBenchPro, which contains CoreMark Pro, seems to be using some
dedicated Intel libraries, in particular for string handling.

armeabi-v7a:
total 6132
90124 Jun 25 2014 libcoremark_selector.so
276904 Jun 25 2014 libcoremark.so
2668588 Jun 25 2014 libmain3d.so
1384148 Jun 25 2014 libmithdp.so
1847720 Jun 25 2014 libmithsp.so

x86:
total 13344
69700 Jun 25 2014 libcoremark_intel_atom2.so
87796 Jun 25 2014 libcoremark_intel_atom.so
87844 Jun 25 2014 libcoremark_intel_core2.so
96388 Jun 25 2014 libcoremark_intel_core4.so
38144 Jun 25 2014 libcoremark_selector.so
243412 Jun 25 2014 libcoremark.so
2712664 Jun 25 2014 libmain3d.so
1059184 Jun 25 2014 libmithdp_intel_atom2.so
1074052 Jun 25 2014 libmithdp_intel_atom.so
1099152 Jun 25 2014 libmithdp_intel_core2.so
1095056 Jun 25 2014 libmithdp_intel_core4.so
1368780 Jun 25 2014 libmithdp.so
636244 Jun 25 2014 libmithsp_intel_atom2.so
646992 Jun 25 2014 libmithsp_intel_atom.so
684348 Jun 25 2014 libmithsp_intel_core2.so
692620 Jun 25 2014 libmithsp_intel_core4.so
1928552 Jun 25 2014 libmithsp.so

Are going to find out this is similar to the AnTuTu/icc story?