Ravi versus Lua versus Luajit performance test again

[Dibyendu Majumdar <mobile@...>]

Hi,

With more opcodes now being compiled I was able to run another test.
The test programs are in:

https://github.com/dibyendumajumdar/ravi/tree/master/ravi-tests

The results are below.

C:\github\ravi\ravi-tests>\lua-5.3.0\src\build\Release\lua.exe fannkuchen.lua 11
556355
Pfannkuchen(11) = 51
elapsed time in secs 60.775

C:\github\ravi\ravi-tests>..\build\Release\lua.exe  fannkuchen.lua 11
556355
Pfannkuchen(11) = 51
elapsed time in secs 31.107

C:\github\ravi\ravi-tests>..\build\Release\lua.exe  fannkuchen.ravi
556355
Pfannkuchen(1) = 51
elapsed time in secs 11.226

C:\github\ravi\ravi-tests>\luajit\luajit.exe fannkuchen.lua 11
556355
Pfannkuchen(11) = 51
elapsed time in secs 4.703

---------------------------------------

The fannkuchen.ravi has the parameter 11 hard-coded as the compiler
does not yet allow typed parameters.

The 2 and 3rd tests are both JIT compiled - but the 3rd uses static
typing and Ravi array specialization to improve performance.

Thanks and Regards

Dibyendu