My most recent encounter with the 'too big data' was over the aerial flyover data of the Hycroft mine. The data provided by the survey company was beautiful. The point density was phenomenal and the area covered was huge. The only problem? It wouldn't triangulate in Maptek Vulcan.
Don't worry, I am an old hand at Vulcan and managed to triangulate the entire solid using the external executable. Now the problem is that I can't load the triangulation into Vulcan and use it in a reasonable fashion. The triangulation loads but has so many points that I can't rotate it smoothly. The topography moves in such big fits and starts that I can't see what is going on.
Time to filter the triangulation. I'm pretty good with the Maptek Vulcan software so I can manage this too. After a few iterations I finally have the triangulation file size filtered down to 10 MB from 100 MB. Now the triangulation loads, rotates and looks pretty good but I feel cheated. We paid a lot of money for the point cloud provided by the aerial survey. Why can't I use all those points? Hopefully, the 64-bit version of Vulcan will help with that.
The 64-bit version of Maptek Vulcan should allow users to load large datasets like my huge topography triangulation. This data will still use the same processor but allow more points to be held in memory. Will this limitation on processing power be the new software limitation? Now that I can load truly huge datasets will they just be too slow to effectively work with?
64-bit Vulcan is a step in the right direction but is there something more that could be done to create better mine modeling software?
I also worry that, with no actual release date, Maptek means to delay the release of the version because it is not truly complete. Only time will tell.