From: "Nathan Shipley" Received: from mail-pg1-f179.google.com ([209.85.215.179] verified) by media-motion.tv (CommuniGate Pro SMTP 6.1.0) with ESMTPS id 7215346 for AE-List@media-motion.tv; Mon, 17 Dec 2018 16:49:39 +0100 Received: by mail-pg1-f179.google.com with SMTP id j10so5109343pga.1 for ; Mon, 17 Dec 2018 07:58:04 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to; bh=Dp0G8ftOOM0nWdEschjJRwAbht/bn8dwy2WIUntYDCA=; b=K33EdPel1gHO+JuLTqinxmmjaDm5HcdLYKpDjQ2oHCHVS36EmIzNwGpb23+itfwwro HXurhSx4eyKA8H+IzGgtK0tWkFLkTupWIcE00xE7uEKimhmMjmCbuXSDifQurKpprgny WjrclUiUS2I6YNiIlAouHyn7XKpjBD7DbhDU1qQcboRW1rghdIRIb9gSqt2jbbvhRa2A 8dE3huYVZhfhyAKP7Ifl3gfWtHJ7kuUDmHx3FqiPJEgy5g3Dq/BruDzIvIOtaAcR5Z7R 4Vi3iu1P8dMQ2Dl7uJt4YCNF0yapTYJP/LAkkRik+oG2OUGwojhbVL4yb/JRYQJMNthB CUsg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to; bh=Dp0G8ftOOM0nWdEschjJRwAbht/bn8dwy2WIUntYDCA=; b=TGIknStvPRK1q2S2bdZGYEGgLmQh2d593SIVr1FVpUCKOqVeptJ7yBuxyq7WxZgugY BCzqTE1g2JHe5o7YIswnEDbs1P6u4eoT33RyM7a9f+V5IGtaM29ADcTZpKei+nlGZZ9F OcyZ/oievWyZOd5UTWnHZdrefAwkMKHSMauZBsm2z2x630CerwoFyvHSFGvYwThrSHjT thRMS+eGZuzFcL68rFoGwvzl8anzcVIB1zs1UpYUPfO8HiLnd2jCTcpTO+4vvts/kYdF AW9fzYcvfIshTWdwfCHDnY2i0I0hcWs6AaI55g0fxF3WWi8XZjYplL9k8oNJWHXlhYen eEYQ== X-Gm-Message-State: AA+aEWaBT4qbhWKmz9sEBFqvoV0oYnbQh32/9LoOMcuc9kIrhRGgp642 mhTHJ6orleuTg/TH7hpIiU7w4Yd3uPFq5sCg9bME5g== X-Google-Smtp-Source: AFSGD/UCa9FwdlURxM3YIrN2LGxF3NnXlCCDoThmiN9jBvkNgkyT4PDYEE0r1RP4KRlDmIrOcAGCO+RdHZEaU1RYHks= X-Received: by 2002:a63:e545:: with SMTP id z5mr12597867pgj.195.1545062282058; Mon, 17 Dec 2018 07:58:02 -0800 (PST) MIME-Version: 1.0 References: In-Reply-To: Date: Mon, 17 Dec 2018 07:57:50 -0800 Message-ID: Subject: Re: [AE] 8 bpc effect on a 16 bpc project To: After Effects Mail List Content-Type: multipart/alternative; boundary="0000000000007a61f0057d39d8f4" --0000000000007a61f0057d39d8f4 Content-Type: text/plain; charset="UTF-8" It's there, but I wonder if it's not that noticeable because AE seems to be adding some dithering/noise to reduce the obviousness of the 8-bit effect. Here's an example where I see it. Try this in a 16-bit AE project: - In a 1920x1080 comp, make a new solid - Apply Gradient Ramp, default black to white settings - Apply Levels, set gamma to 0.25 - Apply another Levels, set gamma to 4.00 Those two gamma adjustments are the inverse of each other and should balance each other out so the original ramp should (theoretically) be unchanged. At this point, however, you should still see SOME banding in the dark part of the ramp if you zoom in. Not nearly as bad as if the project was set to 8-bit, but there. Try switching to 8-bit and you'll see the banding turn very obvious. Now, with the project set back to 16-bit, apply an 8-bit effect like Card Wipe or Motion Tile and put it in between the two Levels effects on your solid. The banding suddenly changes and gets ugly. The second levels effect is now applying a gamma of 4 to an 8-bit image instead of the 16-bit one from before. If you mouse over and look at the color values in the info window (especially in the dark areas of the ramp), you'll see that there are only a few values as you move the mouse around. The interesting part is that the image gets suddenly noisy/grainy. AE must be adding some dithering or noise the reduce the visual impact of an 8 bit effect in a 16-bit project. The same way you might add a little noise to a subtle gradient with banding. It doesn't do this when the project is set to 8-bit. Is the noise/dithering possibly why you're not seeing anything when you put the motion tile effect in? Hopefully that's helpful, - Nathan On Mon, Dec 17, 2018 at 5:35 AM Nick Fox-Gieg wrote: > My understanding is the 8bpc penalty only applies to the output of that > single effect; the effects before and after can resume processing at their > higher bit depth. (Can anybody confirm?) > > Nick > > On Mon, Dec 17, 2018, 5:21 AM Roei Tzoref >> Hey Y'all >> >> in Ae, I am using a 8bit effect on a 16bit effect project (for example - >> motion tile), I am not getting a visual penalty for doing that and >> everything looks exactly the same. I have tested it though makig a gradient >> and pushing the levels to it to get a banding which 16 bpc project removes. >> nomatter where I put the motion tile effect, I am not getting a 8bpc >> result. so how is using an 8bit effect on a 16bit effect actually make a >> difference? can anyone show an example? >> >> Thanks >> > --0000000000007a61f0057d39d8f4 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
It's there, but I wonder if it's not that noticeab= le because AE seems to be adding some dithering/noise to reduce the obvious= ness of the 8-bit effect.

Here's an example where I = see it.=C2=A0 Try this in a 16-bit AE project:

- In a 19= 20x1080 comp, make a new solid
- Apply Gradient Ramp, default bla= ck to white settings
- Apply Levels, set gamma to 0.25
= - Apply another Levels, set gamma to 4.00

Those tw= o gamma adjustments are the inverse of each other and should balance each o= ther out so the original ramp should (theoretically) be unchanged.=C2=A0 At= this point, however, you should still see SOME banding in the dark part of= the ramp if you zoom in.=C2=A0 Not nearly as bad as if the project was set= to 8-bit, but there.=C2=A0 Try switching to 8-bit and you'll see the b= anding turn very obvious.=C2=A0=C2=A0

Now, w= ith the project set back to 16-bit, apply an 8-bit effect like Card Wipe or= Motion Tile and put it in between the two Levels effects on your solid.=C2= =A0 The banding suddenly changes and gets ugly.

Th= e second levels effect is now applying a gamma of 4 to an 8-bit image inste= ad of the 16-bit one from before.=C2=A0 If you mouse over and look at the c= olor values in the info window (especially in the dark areas of the ramp), = you'll see that there are only a few values as you move the mouse aroun= d.

The interesting part is that the image gets sud= denly noisy/grainy.=C2=A0 AE must be adding some dithering or noise the red= uce the visual impact of an 8 bit effect in a 16-bit project.=C2=A0 The sam= e way you might add a little noise to a subtle gradient with banding.=C2=A0= It doesn't do this when the project is set to 8-bit.

Is the noise/dithering possibly why you're not seeing anything = when you put the motion tile effect in?

Hopefully that's = helpful,
=C2=A0- Nathan

On Mon, Dec 17, 2018 at 5:35 AM Nick Fox-Gieg <AE-List@media-motion.tv> wrote:<= br>
My understanding is the 8bpc penalty only applies to the output of that s= ingle effect; the effects before and after can resume processing at their h= igher bit depth. (Can anybody confirm?)

Nick

O= n Mon, Dec 17, 2018, 5:21 AM Roei Tzoref <AE-List@media-motion.tv wrote:
=
Hey Y'all

in Ae, I am using a 8bit effect on a 16bit ef= fect project (for example - motion tile),=C2=A0 I am not getting a visual p= enalty for doing that and everything looks exactly the same. I have tested = it though makig a gradient and pushing the levels to it to get a banding wh= ich 16 bpc project removes. nomatter where I put the motion tile effect, I = am not getting a 8bpc result. so how is using an 8bit effect on a 16bit eff= ect actually make a difference? can anyone show an example?

Thanks
--0000000000007a61f0057d39d8f4--