X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Tue, 04 Oct 2005 14:24:14 -0400 Message-ID: X-Original-Return-Path: Received: from mail.pshift.com ([63.166.217.30] verified) by logan.com (CommuniGate Pro SMTP 5.0c3) with ESMTP id 747394 for lml@lancaironline.net; Tue, 04 Oct 2005 14:20:19 -0400 Received-SPF: none receiver=logan.com; client-ip=63.166.217.30; envelope-from=colyncase@earthlink.net Received: from ccaselt (unverified [206.229.199.193]) by mail.pshift.com (Vircom SMTPRS 4.1.361.21) with SMTP id for ; Tue, 4 Oct 2005 14:19:15 -0400 X-Original-Message-ID: <00ef01c5c910$1fc5bec0$3b11020a@nvidia.com> From: "colyncase on earthlink" X-Original-To: "Lancair Mailing List" References: Subject: [LML] non certified X-Original-Date: Tue, 4 Oct 2005 11:19:12 -0700 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_00EC_01C5C8D5.72CC7B80" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 This is a multi-part message in MIME format. ------=_NextPart_000_00EC_01C5C8D5.72CC7B80 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable why I don't want an SAP. A very famous computer chip manufacturer with a large installed base and = annual unit sales in the 10's of millions still has to rely on this = heuristic for quality control: In the field, every new system configuration ultimately encounters some = bug. For example, the wiz-bang joy-stick that worked fine in 1995 = doesn't work anymore because the timing changed just enough that some = formerly impossible lock-up condition occurs. It only happens with = this joystick, when running a certain game app, when in power mode 4 AND = a disk access occurs just when you depress the trigger. The QA process is to add a machine to the testing lab which runs this = hardware configuration with this application and a disk = flogger/automated button pusher in the background. Somewhere there is = a room with hundreds of these machines in it, each representing some = unique failure mode. Mind you this all happens after massive amounts of = automated testing pre-silicon. If there were 1,000 SAPs installed per year that would represent .1% of = the market covered by this effort. SAPs are a great idea as long as you can find the money for massive = testing and are willing to expense a significant number of = failure-finding aircraft. ....but we're talking about a world where no one has done a full spin = test program on either of the leading contenders for Lancair de-ice. Colyn ------=_NextPart_000_00EC_01C5C8D5.72CC7B80 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable  
why I don't want an SAP.
 
A very famous computer chip = manufacturer with a=20 large installed base and annual unit sales in the 10's of millions still = has to=20 rely on this heuristic for quality control:
 
In the field, every new system = configuration=20 ultimately encounters some bug.   For example, the wiz-bang = joy-stick=20 that worked fine in 1995 doesn't work anymore because the timing changed = just=20 enough that some formerly impossible lock-up condition = occurs.   It=20 only happens with this joystick, when running a certain game app, when = in power=20 mode 4 AND a disk access occurs just when you depress the = trigger.
 
The QA process is to add a machine to = the testing=20 lab which runs this hardware configuration with this application and a = disk=20 flogger/automated button pusher in the background.   Somewhere = there=20 is a room with hundreds of these machines in it, each representing some = unique=20 failure mode.  Mind you this all happens after massive amounts of = automated=20 testing pre-silicon.
 
If there were 1,000 SAPs installed per = year that=20 would represent .1% of the market covered by this effort.
 
SAPs are a great idea as long as you = can find the=20 money for massive testing and are willing to expense a significant = number of=20 failure-finding aircraft.
....but we're talking about a world = where no one=20 has done a full spin test program on either of the leading contenders = for=20 Lancair de-ice.
 
 
Colyn
 
------=_NextPart_000_00EC_01C5C8D5.72CC7B80--