Return-Path: Received: from gateway14.websitewelcome.com ([67.18.70.2] verified) by media-motion.tv (CommuniGate Pro SMTP 4.2.10) with ESMTP id 4744913 for AE-List@media-motion.tv; Wed, 13 Jun 2012 00:45:38 +0200 Received: by gateway14.websitewelcome.com (Postfix, from userid 5007) id 6B9F57F5A0728; Tue, 12 Jun 2012 17:48:18 -0500 (CDT) Received: from ham02.websitewelcome.com (unknown [173.192.111.51]) by gateway14.websitewelcome.com (Postfix) with ESMTP id 5D9447F5A0702 for ; Tue, 12 Jun 2012 17:48:18 -0500 (CDT) Received: by ham02.websitewelcome.com (Postfix, from userid 666) id 53CA39812F8D5; Tue, 12 Jun 2012 17:48:18 -0500 (CDT) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on ham02.websitewelcome.com X-Spam-Flag2999: NO X-Spam-Level2999: X-Spam-Status2999: "No, score=-1.9 required=5.0 tests=BAYES_00,HTML_MESSAGE autolearn=ham version=3.3.1 Received: from alpina.websitewelcome.com (alpina.websitewelcome.com [74.54.176.2]) by ham02.websitewelcome.com (Postfix) with ESMTP id E89DF9812F876 for ; Tue, 12 Jun 2012 17:48:17 -0500 (CDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=default; d=crishdesign.com; h=Received:From:Mime-Version:Content-Type:Subject:Date:In-Reply-To:To:References:Message-Id:X-Mailer:X-BWhitelist:X-Source:X-Source-Args:X-Source-Dir:X-Source-Sender:X-Source-Auth:X-Email-Count:X-Source-Cap; b=V1MWHONHFBZBhY6AhdhU5fsmOQ8W6s3l0RWqA4d6CymTr6+7CxlexMIBJck8l436l3zZk/Jo9e1g8xmFfQ1Qz0HNEyPPH8DdsIAUkOx3VA2xXQbthQo/18qzEXDb+21o; Received: from [199.21.106.94] (port=58530 helo=[192.168.0.7]) by alpina.websitewelcome.com with esmtpsa (TLSv1:AES128-SHA:128) (Exim 4.69) (envelope-from ) id 1SeZsn-0007K1-I2 for AE-List@media-motion.tv; Tue, 12 Jun 2012 17:48:17 -0500 From: Chris Meyer Mime-Version: 1.0 (Apple Message framework v1278) Content-Type: multipart/alternative; boundary="Apple-Mail=_D9E94D7D-3B7F-46E7-9FB5-8CDCC9F53E36" Subject: Re: [AE] (OT) An example of a PC version of a Mac Pro with today's tech for AE Date: Tue, 12 Jun 2012 16:48:16 -0600 In-Reply-To: To: After Effects Mail List References: Message-Id: <0F6FA53A-17EF-4D90-B053-93098F2C0B75@crishdesign.com> X-Mailer: Apple Mail (2.1278) X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - alpina.websitewelcome.com X-AntiAbuse: Original Domain - media-motion.tv X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - crishdesign.com X-BWhitelist: no X-Source: X-Source-Args: X-Source-Dir: X-Source-Sender: ([192.168.0.7]) [199.21.106.94]:58530 X-Source-Auth: chris@crishdesign.com X-Email-Count: 1 X-Source-Cap: Y3Jpc2h3ZWI7Y3Jpc2h3ZWI7YWxwaW5hLndlYnNpdGV3ZWxjb21lLmNvbQ== --Apple-Mail=_D9E94D7D-3B7F-46E7-9FB5-8CDCC9F53E36 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 On Jun 12, 2012, at 4:05 PM, Stephen van Vuuren wrote: > I=92ve found that buy RAM for HT cores paying nice dividends for AE = performance =96 i.e. 4G per total core including virtual allows MP = renders that fail with 3G or 2GB per core. Assigning more RAM per process in multiproc rendering makes perfect = sense, especially for large format work. Our gentle caution is to not = overestimate how many processes are actually contributing to a multiproc = render. > Even with the hit of HT MP renders, it still is much faster than a = single core rendering. Agree with that too. Esp. since CS5, when launching multiproc rendering = got a lot more efficient. > So I disagree to only buy RAM for physical cores but agree that buy = that max you can afford and install. If you are assigning more cores to AE that are physically on your CPU, = you're not getting anything additional out of multiproc rendering = compared to stopping at the number of physical cores.=20 If you are also buying RAM to support those virtual cores, you're = potentially wasting money. But aside from that from that particular narrow scope, more RAM helps = with caching, previews, and large formats, and more cores - both = physical and virtual - help the portions of AE that are multithreaded. crawling out from under the hood - Chris --Apple-Mail=_D9E94D7D-3B7F-46E7-9FB5-8CDCC9F53E36 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=windows-1252
On Jun 12, 2012, at 4:05 PM, Stephen van = Vuuren wrote:

I=92ve = found that buy RAM for HT cores paying nice dividends for AE performance = =96 i.e. 4G per total core including virtual allows MP renders that fail = with 3G or 2GB per core.

Assigning = more RAM per process in multiproc rendering makes perfect sense, = especially for large format work. Our gentle caution is to not = overestimate how many processes are actually contributing to a multiproc = render.

Even with the hit of HT = MP renders, it still is much faster than a single core rendering. =

Agree with = that too. Esp. since CS5, when launching multiproc rendering got a lot = more efficient.

So I disagree to only = buy RAM for physical cores but agree that buy that max you can afford = and = install.

I= f you are assigning more cores to AE that are physically on your CPU, = you're not getting anything additional out of multiproc rendering = compared to stopping at the number of physical = cores. 

If you are also buying RAM to = support those virtual cores, you're potentially wasting = money.

But aside from that from that particular = narrow scope, more RAM helps with caching, previews, and large formats, = and more cores - both physical and virtual - help the portions of AE = that are multithreaded.

crawling out from under = the hood -
Chris

= --Apple-Mail=_D9E94D7D-3B7F-46E7-9FB5-8CDCC9F53E36--