[FFmpeg-devel] [PATCH 1/7] lavu/pixfmt: add AV_PIX_FMT_RGBA32

[Timo Rothenpieler]

On 11.10.2024 23:19, Lynne via ffmpeg-devel wrote:
> On 11/10/2024 23:06, Michael Niedermayer wrote:
>> On Thu, Oct 10, 2024 at 07:32:45AM +0200, Lynne via ffmpeg-devel wrote:
>>> This format is useful for doing certain lossless transforms on images,
>>> RCT in particular, which require you to escalate the size from 16 to
>>> 32 bits to avoid overflows.
>>>
>>> APIchanges will be done alongside when comitting.
>>> ---
>>>   libavutil/pixdesc.c | 27 +++++++++++++++++++++++++++
>>>   libavutil/pixfmt.h  |  4 ++++
>>>   2 files changed, 31 insertions(+)
>>
>> this needs at least am update to fate:
>> also iam not sure about the name rgba32 instead of rgba128
> 
> Yeah, me neither. The thing is, the float equivalent of the format is 
> RBGAF32.
> I don't mind RGBA128, but are there any other suggestions?

It'd definitely be confusing to have RGB32 and RGBA32 to be something so 
completely different.
The Float-One can kinda get away with it by the nature of being float 
and different from the get go, though it's definitely also a bit confusing.