Path: csiph.com!x330-a1.tempe.blueboxinc.net!newsfeed.hal-mli.net!feeder3.hal-mli.net!newsfeed.hal-mli.net!feeder1.hal-mli.net!npeer03.iad.highwinds-media.com!news.highwinds-media.com!feed-me.highwinds-media.com!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail From: MitchAlsup Newsgroups: comp.arch Subject: Re: Single Thread Performance Date: Sun, 5 Feb 2012 21:35:53 -0800 (PST) Organization: http://groups.google.com Lines: 23 Message-ID: <1083982.844.1328506553065.JavaMail.geo-discussion-forums@vbtr6> References: Reply-To: comp.arch@googlegroups.com NNTP-Posting-Host: 72.177.28.22 Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Trace: posting.google.com 1328506553 5193 127.0.0.1 (6 Feb 2012 05:35:53 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: Mon, 6 Feb 2012 05:35:53 +0000 (UTC) Cc: Unspecified In-Reply-To: Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=72.177.28.22; posting-account=H_G_JQkAAADS6onOMb-dqvUozKse7mcM User-Agent: G2/1.0 X-Google-Web-Client: true Xref: x330-a1.tempe.blueboxinc.net comp.arch:5777 On Saturday, February 4, 2012 10:24:23 AM UTC-6, Unspecified wrote: > I was wondering if Developed Economies are considering > pursuing Single Thread Performance as worthwhile. Single threaded performance is pretty much tapped out. Given the current co= res one might be able to get 5%-to-15% more out of a single thread at doubl= e the cost, area, and power. So, the economies of scale and basic limits have turned the tables, so that= one can get more performance from 2,3,4,6,8,... cores than one can get fro= m an equivalent amount of area/power/engineering into single threaded archi= tecture. Drawing in the Graphics is a way to allow the CPU core count to remain fair= ly low, and lower the system costs for the lower scale of the market. Here,= one trades a number of CPU cores for a number of Graphics engines (n.e. co= res). Mixing the exact numbers with the available dies space and on-die cah= ce to build the best cost performer at the box level. Realistically, I can see no difference in performance on my new 6-core syst= em compared to the 4-core system it replaces. Mitch