Re: [gdal-dev] Any difference between VRT SimpleSource and no-op ComplexSource?
On Thu, Dec 21, 2023, at 15:51, Even Rouault wrote: It only uses a ComplexSource when it needs to, that is to say when there is nodata or a mask band. I didn't realize this. I think I'm hitting a corner case with -separate, which is practically the only way I've ever used it: Ah, indeed, I now see the issue in that mode. Fixed per https://github.com/OSGeo/gdal/pull/8986 Even -- http://www.spatialys.com My software is free, but my time generally not. ___ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev
Re: [gdal-dev] Any difference between VRT SimpleSource and no-op ComplexSource?
Hi, About -separate and multi-band inputs, see https://gdal.org/programs/gdalbuildvrt.html#cmdoption-gdalbuildvrt-separate “Starting with GDAL 3.8, all bands of each input file are added as separate VRT bands, unless -b is specified to select a subset of them. Before GDAL 3.8, only the first band of each input file was placed into a new VRT band, and -b was ignored.” -Jukka Rahkonen- Lähettäjä: gdal-dev Puolesta Lauren?iu Nicola via gdal-dev Lähetetty: torstai 21. joulukuuta 2023 16.09 Vastaanottaja: gdal-dev@lists.osgeo.org Aihe: Re: [gdal-dev] Any difference between VRT SimpleSource and no-op ComplexSource? On Thu, Dec 21, 2023, at 15:51, Even Rouault wrote: It only uses a ComplexSource when it needs to, that is to say when there is nodata or a mask band. I didn't realize this. I think I'm hitting a corner case with -separate, which is practically the only way I've ever used it: $ gdalbuildvrt test.vrt T35TMK_20230520T090559_B02_10m.jp2 && cat test.vrt 0...10...20...30...40...50...60...70...80...90...100 - done. PROJCS["WGS 84 / UTM zone 35N",GEOGCS["WGS 84",DATUM["WGS_1984",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]],AUTHORITY["EPSG","6326"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","4326"]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",27],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",50],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","32635"]] 3.9996e+05, 1.e+01, 0.e+00, 5.4000e+06, 0.e+00, -1.e+01 Gray T35TMK_20230520T090559_B02_10m.jp2 1 2 4 8 16 $ gdalbuildvrt -separate test.vrt T35TMK_20230520T090559_B02_10m.jp2 && cat test.vrt 0...10...20...30...40...50...60...70...80...90...100 - done. PROJCS["WGS 84 / UTM zone 35N",GEOGCS["WGS 84",DATUM["WGS_1984",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]],AUTHORITY["EPSG","6326"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","4326"]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",27],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",50],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","32635"]] 3.9996e+05, 1.e+01, 0.e+00, 5.4000e+06, 0.e+00, -1.e+01 T35TMK_20230520T090559_B02_10m.jp2 1 Otherwise it will use SimpleSource. There is a performance hit in using ComplexSource when it is not needed. In 3.8, there has been an optimization to improve ComplexSource performance in common nodata-only scenarios: cf https://github.com/OSGeo/gdal/commit/cd67491d3909f5f05160fa209609b38d973c42de Oh, that's great! PS: are there good reasons for not allowing -separate for multi-band inputs (by including all bands) or propagating the input block size if it's the same for all of them? Laurentiu ___ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev
Re: [gdal-dev] Any difference between VRT SimpleSource and no-op ComplexSource?
On Thu, Dec 21, 2023, at 15:51, Even Rouault wrote: > It only uses a ComplexSource when it needs to, that is to say when there is > nodata or a mask band. > I didn't realize this. I think I'm hitting a corner case with -separate, which is practically the only way I've ever used it: $ gdalbuildvrt test.vrt T35TMK_20230520T090559_B02_10m.jp2 && cat test.vrt 0...10...20...30...40...50...60...70...80...90...100 - done. PROJCS["WGS 84 / UTM zone 35N",GEOGCS["WGS 84",DATUM["WGS_1984",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]],AUTHORITY["EPSG","6326"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","4326"]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",27],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",50],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","32635"]] 3.9996e+05, 1.e+01, 0.e+00, 5.4000e+06, 0.e+00, -1.e+01 Gray T35TMK_20230520T090559_B02_10m.jp2 1 2 4 8 16 $ gdalbuildvrt -separate test.vrt T35TMK_20230520T090559_B02_10m.jp2 && cat test.vrt 0...10...20...30...40...50...60...70...80...90...100 - done. PROJCS["WGS 84 / UTM zone 35N",GEOGCS["WGS 84",DATUM["WGS_1984",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]],AUTHORITY["EPSG","6326"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","4326"]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",27],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",50],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","32635"]] 3.9996e+05, 1.e+01, 0.e+00, 5.4000e+06, 0.e+00, -1.e+01 T35TMK_20230520T090559_B02_10m.jp2 1 > Otherwise it will use SimpleSource. There is a performance hit in using > ComplexSource when it is not needed. In 3.8, there has been an optimization > to improve ComplexSource performance in common nodata-only scenarios: cf > https://github.com/OSGeo/gdal/commit/cd67491d3909f5f05160fa209609b38d973c42de Oh, that's great! PS: are there good reasons for not allowing -separate for multi-band inputs (by including all bands) or propagating the input block size if it's the same for all of them? Laurentiu___ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev
Re: [gdal-dev] Any difference between VRT SimpleSource and no-op ComplexSource?
Laurențiu, I noticed that gdalbuildvrt with no options produces VRTs that use ComplexSource, when SimpleSource would be sufficient. Is there any meaningful performance difference between the two? I know I should measure myself, I just didn't get to try it yet. It only uses a ComplexSource when it needs to, that is to say when there is nodata or a mask band. Otherwise it will use SimpleSource. There is a performance hit in using ComplexSource when it is not needed. In 3.8, there has been an optimization to improve ComplexSource performance in common nodata-only scenarios: cf https://github.com/OSGeo/gdal/commit/cd67491d3909f5f05160fa209609b38d973c42de Even -- http://www.spatialys.com My software is free, but my time generally not. ___ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev