I do specify -i 20 in my scripts (that was a typo just in my original email). 
Does it still work ok for you with an interval of 20 instead of 100?


I’m actually doing more tests just now. I’ve found that gdal_contour is 
significantly faster with a far smaller output shape file size if I first 
reproject to EPSG:3785 using gdalwarp, e.g
gdalwarp -s_srs EPSG:4326 -t_srs EPSG:3785 -r bilinear srtm_36_02.tif 
srtm_36_02_rp.tif



Then contour extraction with
gdal_contour -a elev -snodata 32767 -i 20 srtm_36_02_rp.tif 
srtm_36_02_contours.shp



Interestingly if I gdalwarp then gdal_merge the contour processing is still 
pretty fast (but with the artifacts of bilinear warping). 
But if I first merge, then reproject with gdalwarp then the processing is very 
slow, or fails due to the ~4GB limit


Not entirely sure what is going on, but I’ll put together a script to 
demonstrate the different issues.


Tom




> On Mar 24, 2014, at 1:41 PM, Jukka Rahkonen <[email protected]> wrote:
> 
> 
> Jukka Rahkonen <jukka.rahkonen <at> mmmtike.fi> writes:
> 
> 
>> 
>> 
>> Tom Beddard <tom <at> subblue.com> writes:
>> 
>> 
>> 
>> 
>>> Can anyone else confirm this behaviour when extracting contours from
>>> 
>>> 
>>> merged SRTM tiles?
>> 
>> 
>> 
> 
> Can you check this command, it is perhaps not that you actually used:
> gdal_contour -a elev -snodata 32767 -i uk-3785.tif uk-contour.shp
> 
> 
> My GDAL gives FAILURE: Neither -i nor -fl are specified.
> With -i 100 I have managed to run your commands. Result was 620 MB shapefile
> with 135020 linestrings and it took several minutes to run it but it worked
> for me.
> 
> 
> -Jukka Rahkonen-
> 
> 
> 
> 
> 
> 
> _______________________________________________
> gdal-dev mailing list
> [email protected]
> http://lists.osgeo.org/mailman/listinfo/gdal-dev
> 
> 
> 
_______________________________________________
gdal-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/gdal-dev

Reply via email to