The polynomial fits are never accurate over the whole radius interval. Therefore, one optimises them for the borders because the direct comparison of lines in the picture with the (perfectly straight) borders is most critical. The PTLens maintainer optimises only the borders according to his website, so old LensFun data in particular was created this way. I try to include a line closer to the centre in the fit. Be that as it may, you have to make a compromise.
Therefore, I think that a perfectly made crop 1.5 measurement is better for 1.5 sensors than a perfectly made crop 1.0 measurement. The question is: how was LensFun designed? Does it allow multi-crop entries? ------------------------------------------------------------------------------ Put Bad Developers to Shame Dominate Development with Jenkins Continuous Integration Continuously Automate Build, Test & Deployment Start a new project now. Try Jenkins in the cloud. http://p.sf.net/sfu/13600_Cloudbees _______________________________________________ Lensfun-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/lensfun-users
