Hi Even,
I have 4gb of ram on this machine and task manager reported max usage
was around 2gb. I wasn't using gdal_merge because in the past that
hasn't produced a seamless mosaic. That was some years ago though so I
should probably try again.
I tried again specifying 300 instead of 500 for memory usage and it
worked, topping out at 2.9gb of ram usage according to sysinternals
Process Explorer. So I'm guessing that the first time the 2gb figure I
was seeing in ram usage was what gdal was already using, not what it
asked for and failed to get.
thanks,
matt wilkie
--------------------------------------------
Geographic Information,
Information Management and Technology,
Yukon Department of Environment
10 Burns Road * Whitehorse, Yukon * Y1A 4Y9
867-667-8133 Tel * 867-393-7003 Fax
http://environmentyukon.gov.yk.ca/geomatics/
--------------------------------------------
Even Rouault wrote:
The only idea I have is pretty obvious and I guess you'd already have it, you
probably require more memory than available. The --config GDAL_CACHEMAX
500 -wm 500 options require at least 1 GB. Otherwise it's a more subtle bug.
Unfortunately it's not very practical to try to reproduce it as your input
datasets are pretty big to download.
I've downloaded the first two files and observed that they are at the same
resolution and projection so you could probably use gdal_merge.py instead
_______________________________________________
gdal-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/gdal-dev
