- From: Ken Russell <kbr@google.com>
- Date: Tue, 26 Jan 2021 19:19:07 -0800
- To: Joe Drago <jdrago@netflix.com>
- Cc: Christopher Cameron <ccameron@google.com>, public-colorweb@w3.org
- Message-ID: <CAMYvS2eHf6zYgV32ZrWgnLmi2sEQzgoEN2R1Zx5xMhok-V=0MA@mail.gmail.com>
On Tue, Jan 26, 2021 at 5:33 PM Joe Drago <jdrago@netflix.com> wrote: > To be fair, I could probably argue the exact opposite direction, something > like what Lars said (slightly massaged): > > We could just offer srgb (the normal, SDR current kind), extended SRGB @ X > nits (HDR on), and extended linear SRGB @ X nits (HDR on), and offer > *nothing* else. If you ask for it in an ImageData, you get floats, and > it's up to your fancy WASM/JS to "do things". > Some of our group's partners - specifically Unity - already use SDR linear sRGB for their rendering pipelines on Android. To achieve parity with native applications, the WebGL working group would like to see the option for unorm8, linear sRGB back buffers exposed to the web in some form. I'm not convinced that the options can be reduced to just these three without losing other significant functionality, but would need to study more the texture formats supported in Core Animation, DXGI, etc. -Ken I prefer more dials, but I did like Lars' point about cutting out > degenerate combinations. It could be as simple as error'ing out of > getContext() on certain "bad" combinations, though. > > > > On Tue, Jan 26, 2021 at 5:22 PM Christopher Cameron <ccameron@google.com> > wrote: > >> >> >> On Tue, Jan 26, 2021 at 11:51 AM Joe Drago <jdrago@netflix.com> wrote: >> >> >>> Export / Save is a weird one, but important. Chris' slides will go into >>> detail on this one (ImageData / ImageBitmap), but I just wanted to note >>> that transfer functions play a much different role here, and due to file >>> format limitations, we might want a few more selections here, notably PQ >>> and HLG. I obviously have a bit of a bias with HEIF-derived formats (AVIF , >>> HEIC), but I think 10bpc P3/BT2020 PQ/HLG images are going to be too >>> useful/common as an export format to not have some bidirectional mechanism >>> for those profiles, even if we don't actually do the direct final encoding >>> to any of those formats. >>> >> >> Yeah, I think export/save will need some more parameters. There are some >> obvious things (like, a uint8-p3 canvas should be saved that way), but I >> think many others will need explicit controls. >> >> >>> Being able to ask the canvas for "please give me unorm16 BT2020 PQ" >>> would be an incredible feature in today's world of WASM-compiled encoders. >>> >> >> Yeah, one of the topics to discuss next meeting should probably be adding >> rec2020-pq named color space (it's down there in the "for future >> consideration" section). >> >> As you mentioned in some of the earlier sections, indeed, we will >> probably want to lean in to some more color spaces especially in the >> context of what CSS Color Level 4 offers. >> >> -- I support flexible work schedules, and I’m sending this email now because it is within the hours I’m working today. Please do not feel obliged to reply straight away - I understand that you will reply during the hours you work, which may not match mine. (credit: jparent@)
Received on Wednesday, 27 January 2021 03:19:33 UTC