significant digits for J-index and M-index?

[Elena Miranda]

Hello M-Tex users,

I'm using M-Tex 3.5.0, and I'm working with imported EBSD data. When I calculate my ODF (de la Vallee Poussin, halfwidth 15º), I follow up with J- and M-index calculations, and I get values of 3.1894 for the J-index and 0.0359 for the M-index. I was wondering if all the digits of calculated J-index and M-index values are significant? I want to figure out how to report this correctly in a manuscript.

Thanks,

--Elena Miranda

[Ralf Hielscher]

Hi Elena,

the J-index depends heavily on the kernel halfwidth. Once the kernel halfwidth is fixed the calculation is quite accurate. You can check how reliable the the calculation is by taking two times randomly half the ebsd data and computing the values. In MTEX 4.0 this is done by

ebsd_small = ebsd.subSample(numberOfPoints)

I hope this helps,

Ralf.

[Elena Miranda]

Hi Ralf,

Thank you for your response. I'm running MTEX 3.5, and I was wondering if you could tell me what this same command is in 3.5 instead? Thank you very much.

--Elena

[Luiz Fernando Grafulha Morales]

Hi Elena

I think you can try to  use 10° instead of 15° during your ODF calculations, this is the value normally used if you want to compare your datasets with pole figures from other authors, your CPO will be more clear, but still not that strong to the point to look "artificial"

In addition, most of the people when discussing about J-index, they use one decimal value (so in your case, would be 3.2), so dont worry too much about super-precision


cheers


Luiz