Can I save a pointcloud as double precision

[yaseryacoob]

Is your feature request related to a problem? Please describe.
Need to save a cloud as double instead of single (in ascii), not clear how to do it, doesn't seem part of the library?

[Nicholas-Autio-Mitchell]

You could access the underlying points from the xyz attribute of a PyntCloud object, which is a numpy array - then convert it to type np.float64 and save it to disk.
Converting from float32 to float64 isn't actually going to give you any extra precision (how would numpy know what the actual extra decimal points should be?)

If your original data is in double precision, I would suggest just feeding a copy into PyntCloud, as to not lose that original precision

[yaseryacoob]

I ended up doing your second option, feeding in the double precision and maintaining it. It is not unusual for geodata to require keeping double precision. Many thanks! From: Nicholas Mitchell [mailto:notifications@github.com] Sent: Tuesday, February 05, 2019 5:50 AM To: daavoo/pyntcloud Cc: Yaser Yacoob; Author Subject: Re: [daavoo/pyntcloud] Can I save a pointcloud as double precision (#234) You could access the underlying points from the xyz attribute of a PyntCloud object, which is a numpy array - then convert it to type np.float64 and save it to disk. Converting from float32 to float64 isn't actually going to give you any extra precision (how would numpy know what the actual extra decimal points should be?) If your original data is in double precision, I would suggest just feeding a copy into PyntCloud, as to not lose that original precision — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub<#234 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AILI-4x7G2h-ZBvIcdin1IZxYc_RpLXpks5vKWHQgaJpZM4aPn3A>.