[pdal] how to work around issue 225

[Steven M. Ottens]

Howard Butler schreef op 4/8/2014 5:24 PM:
> On Apr 8, 2014, at 10:17 AM, Steven M. Ottens <steven at minst.net> wrote:
>
>> Since I need to unzip the LAZ files anyway, I use a static output.xml which always takes /home/geodan/output.las and let the script unzip files to output.las and cleanup after import
> Was a LAZ with filters.cache slower than converting to LAS, writing to disk, and then loading the subsequent LAS file? I would think it shouldn't be...
I ran into problems with the bigger LAZ files, see 
https://github.com/PDAL/PDAL/issues/270#issuecomment-39711054 so I 
switched to unzipping them first. Unzipping is rather fast and the 
crunching was progressing unlike with pure LAZ files.

However I just did a quick check with the new pipelines and there seems 
to be a slight advantage for direct reading of LAZ now. I have another 
30000 LAZ files to read later, I'll do that with straight LAZ and we'll 
see what will be faster in the long run :)

regards,
Steven