From dsmythe@haystack.mit.edu Tue Mar 30 00:15:16 2004 Received: from franklin.haystack.mit.edu (franklin.haystack.mit.edu [192.52.61.78]) by mail.mpifr-bonn.mpg.de (8.8.8+Sun/8.8.8) with SMTP id AAA21135 for ; Tue, 30 Mar 2004 00:35:59 +0200 (MET DST) Received: from dls-pc.haystack.mit.edu (dls-pc.haystack.edu [192.52.61.27]) by franklin.haystack.mit.edu (8.12.8/8.12.8) with ESMTP id i2TMZt00021387 for ; Mon, 29 Mar 2004 17:35:55 -0500 Message-Id: <5.2.1.1.2.20040329170505.00b20d90@franklin> X-Sender: dls@franklin X-Mailer: QUALCOMM Windows Eudora Version 5.2.1 Date: Mon, 29 Mar 2004 17:15:16 -0500 To: walter Alef From: Dan Smythe Subject: Fwd: Mk4 decoder firmware Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Content-Length: 1621 X-Spam-Status: No, hits=-4.0 required=4.0 tests=AWL,BAYES_10 version=2.55 X-Spam-Level: X-Spam-Checker-Version: SpamAssassin 2.55 (1.174.2.19-2003-05-19-exp) Status: R X-Status: N X-KMail-EncryptionState: X-KMail-SignatureState: Walter, Here is a Mark 4 Decoder Status report from Roger Cappallo. ... Dan Date: Mon, 29 Mar 2004 17:02:09 -0500 To: dls@haystack.mit.edu From: Roger Cappallo Subject: Mk4 decoder firmware Cc: bec@haystack.mit.edu Hi Dan, The current state of the decoder firmware revision is that it is about 99.9% healthy. During checkout Brian Corey discovered one mode of phase cal extraction -- one which extracts two tones from each of four signals simultaneously -- that has a slight glitch. If the last tone frequency is one hertz, and there are more than 43 characters in the command line, it will acknowledge with some bogus characters. The above problem only occurs when addressing via the field system, not manually. Looking with a digital scope, it appears that the spurious characters really are originating within the decoder. On the other hand, looking at the decoder internal buffer, it thinks it is sending the write acknowledgement (a $). Somewhere between the code that does the output buffering and what comes out on the UART lines, the spurious characters are being generated. It is perplexing. Since this problem only happens (to our knowledge) in an artificial test case that has little useful application, we could go ahead and release the code as is. The corrections to the old phase cal extraction code are significant in some cases, and there is substantial new functionality in terms of pcal modes (one signal by 8 tones, two signals with 4 tones, and 4 signals with 2 tones each), as well as the new code to support state statistics gathering and display. Regards, Roger