|
From: | Gonzalo Garramuno |
Subject: | Re: [Openexr-devel] [Offtopic]| half lossy datatype |
Date: | Wed, 30 Apr 2014 17:03:31 -0300 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.2.0 |
On 30/04/14 16:30, Paul Miller wrote:
Thanks. That was part of the problem. However, I still don't get a clean transition from float to half uint16. There are differences in the result between calling exr_halflt2uint( uint16_t ) and doing exr_half2float(uint16_t) + exr_flt2uint(uint32_t).pow() doesn't like negative values, and I've seen different results on different machines. You should clip negatives before calling it.
My guess is that it is because conversion from float to half is lossy.
[Prev in Thread] | Current Thread | [Next in Thread] |