X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Fri, 24 Aug 2007 16:52:50 -0400 Message-ID: X-Original-Return-Path: Received: from jrcda.com ([206.130.116.53] verified) by logan.com (CommuniGate Pro SMTP 5.1.11) with ESMTP id 2281838 for lml@lancaironline.net; Fri, 24 Aug 2007 03:37:35 -0400 Received-SPF: none receiver=logan.com; client-ip=206.130.116.53; envelope-from=hwasti@lm50.com Received: from [192.168.1.101] (cbl-238-61.conceptcable.com [207.170.238.61] (may be forged)) (authenticated bits=0) by jrcda.com (8.12.11.20060308/8.12.11) with ESMTP id l7O7auKC008002 for ; Fri, 24 Aug 2007 01:36:57 -0600 X-Original-Message-ID: <46CE8A8D.1030405@lm50.com> X-Original-Date: Fri, 24 Aug 2007 00:36:45 -0700 From: Hamid Wasti User-Agent: Thunderbird 2.0.0.6 (Windows/20070728) MIME-Version: 1.0 X-Original-To: Lancair Mailing List Subject: Re: [LML] Re: fallability in the digital age References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Colyn Case on earthlink wrote: > I have to disagree that you would emit detailed terrain /for lack > of processing/ resources. At this point, the amount of 3d processing > power available in a single chip is huge. Everything you say is possible if you are willing to use Windows or Lunix and an industry standard interface like DirectX. Are you willing to trust your life to a PFD that runs Windows? > (disclaimer: I'm a chip architect for nVidia) The chips we are > currently making.... The capabilities of the chips you are currently making are unavailable to certified avionics unless nVidia is willing to share all the details of the chips with the customer (it was a non-starter the last time I inquired) AND willing to foot the likely 8 figure bill to certify the proprietary source code running on the GPU (Graphics Processor Unit) to DO178 standards. As Brent mentioned, the FAA is getting wiser to all the semantics and hand waving people were able to get away with in the past. There have been some presentations and position papers out of the FAA in recent months that point to a rather dark future for PC based systems in the cockpit, especially when it comes to offloading a bunch of processing to the GPU. This stems in a large part from a lack of visibility into the internal working of the GPU and GPU code. > So I think it's going to happen and someone will do a decent job > of it. That may well come true in the experimental world, but not in the certified world. Not without a complete turnabout in the attitude of the FAA and graphics chip manufacturers like nVidia and neither side has even the slightest incentive to change. Regards, Hamid