Skip to main content
9 events
when toggle format what by license comment
Jun 12, 2023 at 2:44 vote accept icegray
Jun 27, 2023 at 11:05
Jun 9, 2023 at 8:08 answer added CL. timeline score: 0
Jun 8, 2023 at 6:18 comment added icegray Hi @mkennedy, I tested the transform to 4301 as your suggestion. Both results are same (Win10/Linux). 141.5188010 38.27249734. Next, I tested the transform from 4301 to 4612 (This transform requires a grid shift). There are a difference. Win10:41.5153035,38.2754871 <--correct / Linux:141.5153035,38.2754912 <-- Incorrect. Then, I removed the TKY2JGD.gsb file from Windows, for test. In this case the Win result changed to same as Linux (Incorrect value). So I guess, the Win PostGIS has used TKY2JGD correctly but the Linux doesn't use it. ...
Jun 7, 2023 at 11:46 comment added icegray Hi Ian, Thank you for your comment. Actually the tky2jgd grid-shift are quite a little (around 1-20m), but it's official grid-shift from Japan government. So it's important for Japanese GIS...
Jun 6, 2023 at 16:50 comment added mkennedy Some sort of grid shift is occurring. If you just unproject the original coordinates to Tokyo GCS latitude-longitude, you get: 141.5188010 38.27249734 so quite a bit different than the output. You are correct that the windows and linux implementations should give the same results. Maybe first check whether the unprojection results match (use 4301) to confirm that it's the ntv2-based tfm that's different.
Jun 6, 2023 at 11:05 comment added Ian Turton see gis.stackexchange.com/a/327456/79 - are you sure that it matters?
Jun 6, 2023 at 10:15 history edited Kadir Şahbaz CC BY-SA 4.0
added 13 characters in body
S Jun 6, 2023 at 9:08 review First questions
Jun 6, 2023 at 10:15
S Jun 6, 2023 at 9:08 history asked icegray CC BY-SA 4.0