umapd51@sund.cc.ic.ac.uk (W.A.C. Mier-Jedrzejowicz) (07/28/90)
Eric Schweitzer wrote that he called HP and they said there were "no errata" with the HP48 version D. Well, strictly speaking, they are probably right, if by "errata" they mean "errata sheets", but I bet that's not the question Eric was asking! The "HP 48SX Anomalies and Affected Firmware Revision Levels" sheet lists anomalies in versions A, B and C, plus one anomaly in version D. The anomaly, erratum, error or bug in revision D is that if you enter a SYMBOLIC complex number in the form '(A,<B)', where < is the 'angle' symbol (not a less than sign), then the HP48 returns this as 'B*COS(A)+i*B*SIN(A)' instead of 'A*COS(B)+i*A*SIN(B)'. This ONLY happens if you enter a symbolic complex number, and only if you enter it in polar form. A simple solution is to enter symbolic complex numbers in the rectangular form '(X,Y)'. Actually, I have never heard of a user wanting to use symbolic polar entry, so this should not be a problem to most users. As for the anomaly lists, or "errata sheets", not all revision C units come with them, certainly not in Europe. Can anyone report if ANY revision D units have been received with this sheet? Of course, you can get a copy through your friendly User Club, can't you? Is it time I posted another update to my User Club list? Wlodek Mier-Jedrzejowicz, Space & Atmospheric Physics, Imperial College, London BITNET: UMAPD51@VAXA.CC.IC.AC.UK usenet: see header Disclaimer: I neither speak nor write for IC or HP, at least not in this communication.
kevinc@tekig5.PEN.TEK.COM (Kevin E Cosgrove) (07/30/90)
In article <1990Jul27.224836.22140@cc.ic.ac.uk> umapd51@cc.ic.ac.uk writes: > >The anomaly, erratum, error or bug in revision D is that if you enter a >SYMBOLIC complex number in the form '(A,<B)', where < is the 'angle' >symbol (not a less than sign), then the HP48 returns this as >'B*COS(A)+i*B*SIN(A)' instead of 'A*COS(B)+i*A*SIN(B)'. This ONLY >happens if you enter a symbolic complex number, and only if you enter it >in polar form. A simple solution is to enter symbolic complex numbers in >the rectangular form '(X,Y)'. Actually, I have never heard of a user >wanting to use symbolic polar entry, so this should not be a problem to >most users. Rats! Now I wish I'd waited just a little longer to get my 48. The reason I have the calculator is that I wanted easier treatment of complex numbers and matricies. I have found myself in situations where entering symbolic angle notation would have been the easiest route. Does anyone out there know HP's policy on upgrading a ROM for a customer that needs performance in the area of a bug? Thanks.... Kevin Cosgrove <tektronix.TEK.COM!tessi!{ bucket | escargot }!cathouse!kevinc>
matt@comtec.UUCP (matt hoffman) (07/31/90)
In article <1990Jul27.224836.22140@cc.ic.ac.uk>, umapd51@sund.cc.ic.ac.uk (W.A.C. Mier-Jedrzejowicz) writes: > > As for the anomaly lists, or "errata sheets", not all revision C units > come with them, certainly not in Europe. Can anyone report if ANY > revision D units have been received with this sheet? Of course, you can > get a copy through your friendly User Club, can't you? Is it time I > posted another update to my User Club list? I purchased an HP 48 on June 14, 1990. It did not contain an errata sheet describing the "symbolic-complex" error. It did come with an extra manual page (00048-90063) titled, "Important HP 48SX Operating Tips". This page suggested that flag -3 be cleared before NXEQ is executed. It also included a short program for inverting an 8x8 or larger matrix. .....Matt Hoffman.....