[Gdal-dev] Bug in OGDI Driver when accessing tiles (FAILURE: 1: End of selection)

[Frank Warmerdam]

Reiner Beck wrote:
> Well, looking at the amount of fopen/close in the RPF driver for the 
> different image parts, there is certainly a heavy impact on how one 
> accesses the data. In my case I have to do it always on a tile basis 
> (normally 256x256 pixel chunks). As I stated in my previous post, the 
> order in which the tiles are accessed are "a bit random". This is to 
> speed up processing on my side.
> However, I could re-order each request to access the tiles in a 
> different order, for example start with the first row of tiles and so on.

Reiner,

I haven't analysed it carefully, but I am not aware of any sigificant win
from careful ordering of tile requests.  I think requesting bigger tiles
would likely help if it could be accomplished in your context.

> Do you think it would speed up the reading process if I re-order the 
> tile requests?
> 
> BTW, do you know how the NITF files in a CADRG dataset are compressed?

Yes, they use the VQ compression as far as I know.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGF, http://osgeo.org