From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-yw1-x112d.google.com (mail-yw1-x112d.google.com [IPv6:2607:f8b0:4864:20::112d]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id A2BCE3B2A4 for ; Fri, 3 May 2024 03:22:44 -0400 (EDT) Received: by mail-yw1-x112d.google.com with SMTP id 00721157ae682-6202ad4cae3so587747b3.2 for ; Fri, 03 May 2024 00:22:44 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=aterlo.com; s=google; t=1714720964; x=1715325764; darn=lists.bufferbloat.net; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=U0cG5WsG1i5DojTa5mLYZor9VMr59R97yXL3I9mOhP0=; b=d02dxOIyoo8+IGM5D/CR2IeLnxlWh3Ns0XHarwuH82RxpX1Ds86Vim0BfZ8eJ7fp4Z P8eoYB8yI+ojlNau+mgdpkKvi4evrGbMU4u2nu8Gyz1XrCbi7XmuselwMrk7X//QK07T bcYNfRfg0Ll3mY5e1ZzmoJfLfXr0KtI9DOuA4= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1714720964; x=1715325764; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=U0cG5WsG1i5DojTa5mLYZor9VMr59R97yXL3I9mOhP0=; b=PV7AoU+B4QItQGObcWsnB++hsDZpCRo1b0smKEgZBHyqmwr2DqbXByQuQY0nu/oAEG JdNThtCqnQgK7bDzcJVksVR0ZneCoVHmZdauRIKVEzNzyTwGFKRVDgeE2vPYEQrG1Cuo i3SkBmzLgoaelEzG3fxDWBUkn/4g8D8DvZoBBGb4PiLfxpiWLmPTUthJDNrFm8r5BM9M Oeg8LkZ4iq+hHRDKk2um+gbUjCsih1vDjUhbcUajB/XaGL7Y6nHVa/EkVSQ8vyo3+9Zb 2m4Rmu+eZwgcvWX23ANv+DWJwk9cEaTkeCnARouy9g1cO1mnZXC2UCjFseBk+tgXTkNp PYmQ== X-Gm-Message-State: AOJu0YwqQR7d1Jg4otGspQibwDYz9qxNO5yQ4Q1rWdJcc0q5UpsgbPPv Kw4qghyy+BWyNWO4mYMF/NmGiTQiao46r9rXu95NaFlJyQNCv/Ht2ZGo7/7XCAQJ3N5Qz8Pc0pW USLHcNiVL8Lm6SSzDk02m8q0s/5s9eTZTLV3ehIbhX3YHELeeUp4= X-Google-Smtp-Source: AGHT+IGX2NGEPoH/YS+DsaX3taMdPEgQjmRy55djkl6TclAxVGvqox6Ef8uRogvxTn/UleIoU0vClCsCDF/j7q8Xgk8= X-Received: by 2002:a81:91c2:0:b0:618:822a:a916 with SMTP id i185-20020a8191c2000000b00618822aa916mr1798893ywg.13.1714720963996; Fri, 03 May 2024 00:22:43 -0700 (PDT) MIME-Version: 1.0 References: <77d8f31e-860b-478e-8f93-30cb6e0730ac@auckland.ac.nz> In-Reply-To: <77d8f31e-860b-478e-8f93-30cb6e0730ac@auckland.ac.nz> From: Jeremy Austin Date: Thu, 2 May 2024 23:22:33 -0800 Message-ID: To: Ulrich Speidel Cc: starlink@lists.bufferbloat.net Content-Type: multipart/alternative; boundary="000000000000f2c6fe06178795ca" Subject: Re: [Starlink] =?utf-8?q?It=E2=80=99s_the_Latency=2C_FCC?= X-BeenThere: starlink@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: "Starlink has bufferbloat. Bad." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 03 May 2024 07:22:44 -0000 --000000000000f2c6fe06178795ca Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Comments inline. [image: Company logo] *Jeremy Austin* Sr. Product Manager *Preseem | Aterlo Networks* Book a call: https://app.hubspot.com/meetings/jeremy548 1-833-773-7336 ext 718 <18337737336718> *|* 1-907-803-5422 <19078035422> jeremy@aterlo.com On Thu, May 2, 2024 at 5:48=E2=80=AFPM Ulrich Speidel via Starlink < starlink@lists.bufferbloat.net> wrote: > > Which brings me back to the question of how important bandwidth is. The > answer is: probably more important in the future. We're currently relying > mostly on CDNs for video delivery, but I can't fail but notice the progre= ss > that's being made by AI-based video generation. Four or five years ago, > Gen-AI could barely compose a credible image. A couple of years ago, it > could do video sequences of a few seconds. Now we're up to videos in the > minutes. > > If that development is sustained, you'll be able to tell your personal > electronic assistant / spy to dream up a personalised movie, say an > operatic sci-fi Western with car chases on the Titanic floating in space, > and it'll have it generated in no time starring the actors you like. ETA: > Around 2030 maybe? > > But these things will be (a) data-heavy and (b) aren't well suited for CD= N > delivery because you may be the only one to every see a particular movie, > so you'll either need to move the movie generation to the edge, or you ne= ed > to build bigger pipes across the world. I'm not sure how feasible either > option is. > Why shouldn=E2=80=99t the advances in GPU, CPU and storage substitute for bandwidth? Given how compact symbol libraries are today, why shouldn=E2=80= =99t one be able to generate this high quality, customized movie entirely locally, streaming characters, backgrounds, and kinematics from a large library? Preloading their library is how image generation, transcription, etc. works today. This surely must be less total real-time bandwidth than the alternative, which pre-renders elsewhere. With each target customized, there will be a little incentive to pre-render elsewhere. Offset against this potential bandwidth savings =E2=80=94 clearly the next = step in, say, audiobook generation =E2=80=94 is that eventually we will want full stereoscopic 360 environments a la Apple headsets. Plausibly this is in the 16k range at a minimum. The Vegas Sphere is approximately this resolution. My $.02 Jeremy Austin > --000000000000f2c6fe06178795ca Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Comments inline.=C2=A0

=


On Thu, May 2, 2024 at 5:48=E2=80=AFPM Ulr= ich Speidel via Starlink <starlink@lists.bufferbloat.net> wrote:
=20


Which brings me back to the question of how important bandwidth is. The answer is: probably more important in the future. We're currently relying mostly on CDNs for video delivery, but I can't fail but notice the progress that's being made by AI-based video generation. Four or five years ago, Gen-AI could barely compose a credible image. A couple of years ago, it could do video sequences of a few seconds. Now we're up to videos in the minutes.=C2=A0

If that development is sustained, you'll be able to tell your personal electronic assistant / spy to dream up a personalised movie, say an operatic sci-fi Western with car chases on the Titanic floating in space, and it'll have it generated in no time starring the actors you like. ETA: Around 2030 maybe?

But these things will be (a) data-heavy and (b) aren= 9;t well suited for CDN delivery because you may be the only one to every see a particular movie, so you'll either need to move the movie generation to the edge, or you need to build bigger pipes across the world. I'm not sure how feasible either option is.

<= /blockquote>

<snip>

Why shouldn=E2=80=99t the ad= vances in GPU, CPU and storage substitute for bandwidth? Given how compact = symbol libraries are today, why shouldn=E2=80=99t one be able to generate t= his high quality, customized movie entirely locally, streaming characters, = backgrounds, and kinematics from a large library? Preloading their library = is how image generation, transcription, etc. =C2=A0works today. This surely= must be less total real-time bandwidth than the alternative, which pre-ren= ders elsewhere. With each target customized, there will be a little incenti= ve to pre-render elsewhere.

Offset against this potential bandwidth savings =E2=80=94 clearly the n= ext step in, say, audiobook generation =E2=80=94 is that eventually we will= want full stereoscopic 360 environments a la Apple headsets. Plausibly thi= s is in the 16k range at a minimum. The Vegas Sphere is approximately this = resolution.

My $.02
Jeremy Austin

--000000000000f2c6fe06178795ca--
3D"Company
Jeremy Austin
Sr. P= roduct Manager
Preseem | Aterl= o Networks
Book a call:=C2=A0https:/= /app.hubspot.com/meetings/jeremy548