On Thu, Jan 6, 2011 at 7:57 PM, Micha Silver <[email protected]> wrote: >> I agree with you that CRS handling should be improved. Having >> 'unknown' CRS is more useful than assigning a bogus CRS and then >> trying to project it to some other CRS. I think we should remove >> on-the-fly projections option and _always_ do the reprojection of > > Hmm.... Some thoughts: > Won't OTF projection of raster always be a heavy computational process, > slowing down rendering?
On-th-fly raster reprojection is not (IMHO) very expensive. The complexity depends on the desired output quality (e.g. what type of interpolation is used) and raster size. I would expect that most of the raster rendering time is fetching of source data, the warping itself is simple. I thought also about adding some kind of unobtrusive feedback from map canvas indicating possible problems - similar to how browsers nowadays complain about missing plugin or ask whether to save password. This feedback could include both errors (e.g. failed to reproject some parts of layer) and warnings/suggestions (e.g. raster rendering is slow due missing pyramids, layer has unknown CRS). QGIS could also suggest the user to reproject his rasters/vectors for faster rendering. > Second: To which CRS will this automatic OTF reproject to? The situation now > is that a new project always starts in EPSG:4326. There's no way AFAIK to > cause QGIS to always start in some user chosen CRS (other than creating a > default project in your chosen CRS, then change the QGIS shortcut to always > start that project.) > Suppose I have some big LandSAT rasters in UTM, and vector layers in some > other projection. If OTF was always "on" I'd be reprojecting both the > LandSAT and all the vectors into Lon/Lat WGS84. I think it's reasonable to start new project with unknown CRS by default and set CRS from the first loaded layer. Of course an option to override the default project CRS would be convenient for some people working mostly with one CRS. > THird: Having OTF automatic still won't help users with data that lacks a > CRS. No, but at least they can get a notification that a layer has unknown CRS and thus can be aligned incorrectly. Regards Martin _______________________________________________ Qgis-developer mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/qgis-developer
