Return-Path: Received: from mail-gg0-f169.google.com ([209.85.161.169] verified) by media-motion.tv (CommuniGate Pro SMTP 4.2.10) with ESMTP-TLS id 4745161 for AE-List@media-motion.tv; Wed, 13 Jun 2012 08:02:16 +0200 Received: by ggm4 with SMTP id 4so127906ggm.28 for ; Tue, 12 Jun 2012 23:04:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; bh=o6qZ6y2Oph8Q9TsKRyqJezXwFQ0N5GHvwpp5XmJJgto=; b=aEZW85/fqRTpU5+OWC5RKeeKBxSZsUVGncve9rVK53Z6WVJiQNjMq/m/OjgCu/AoDu YRaKD/0QOZiXrSfNaauWwX3PDs97f3dx8+NFpgzyXufi7jrwvoK6+9phjWYE6QEq6hNb pzId9UFN8fSMKQ2w6p1hmqfW5EhCVM8gVLomY+Jj3NxLbxdkTOY9LemDRhWGAJEJMf4w Mxp2tKcPAZdR5BZBLO0FMR97m++6lnryNklWTzvwvNodYNDutVo39ckyldGz0ovOqqDq ZHFCCOXfXho9d8vCs5EyAIhWs3yEAvctlAG6eR/X/aFU26EjR5+sN5ASB8DSZyvMfVqv LcWQ== MIME-Version: 1.0 Received: by 10.50.104.167 with SMTP id gf7mr9847669igb.38.1339567496532; Tue, 12 Jun 2012 23:04:56 -0700 (PDT) Received: by 10.64.17.200 with HTTP; Tue, 12 Jun 2012 23:04:56 -0700 (PDT) In-Reply-To: References: Date: Tue, 12 Jun 2012 23:04:56 -0700 Message-ID: Subject: Re: [AE] (OT) An example of a PC version of a Mac Pro with today's tech for AE From: Darby Edelen To: After Effects Mail List Content-Type: multipart/alternative; boundary=e89a8f2357cdc8921204c2545cb9 --e89a8f2357cdc8921204c2545cb9 Content-Type: text/plain; charset=ISO-8859-1 > > Parts of the program may be running one process, but using many cores - > that's hyperthreading. And you will see acceleration from that with both > physical and virtual cores. > That is not my understanding of hyperthreading. As far as I understand hyperthreading shares the execution blocks on a physical core to push through 2 threads simultaneously instead of 1 with fancy scheduling trickery. If a physical core would've been entirely occupied with Process A but it instead must share resources with Process B then Process A would take longer to complete than it would have without hyperthreading. However, if Process A doesn't fully utilize the core then the remainder can work on Process B and you get a net gain in performance with hyperthreading. One core running multiple threads simultaneously is hyperthreading, where as I don't believe it is possible to have one thread executing on multiple cores simultaneously. All of these threads are getting us tied into quite the knot :) --e89a8f2357cdc8921204c2545cb9 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Parts of the program may be running one process, bu= t using many cores - that's hyperthreading. And you will see accelerati= on from that with both physical and virtual cores.

That is not my understanding of hyperthreading.= =A0 As far as I understand hyperthreading shares the execution blocks on a = physical core to push through 2 threads simultaneously instead of 1 with fa= ncy scheduling trickery.

If a physical core would've been entirely occupied with Process A b= ut it instead must share resources with Process B then Process A would take= longer to complete than it would have without hyperthreading.=A0 However, = if Process A doesn't fully utilize the core then the remainder can work= on Process B and you get a net gain in performance with hyperthreading.
One core running multiple threads simultaneously is hyperthreading, whe= re as I don't believe it is possible to have one thread executing on mu= ltiple cores simultaneously.

All of these threads are getting us tie= d into quite the knot :)
--e89a8f2357cdc8921204c2545cb9--