Path: utzoo!utgpu!watserv1!watmath!att!drutx!cbnewsb!druhi.att.com!neal
From: neal@druhi.att.com (XGPB30000-McBurnettND(DR9225)289)
Newsgroups: comp.sys.handhelds
Subject: Re: HP-48 Comparison Test
Keywords: Calendar, clock accuracy, bundling software with I/O package
Message-ID: <332@cbnewsb.ATT.COM>
Date: 16 Mar 90 23:23:58 GMT
References: <39230@apple.Apple.COM> <326@cbnewsb.ATT.COM> <1990Mar13.232026.10379@cc.ic.ac.uk>
Sender: news@cbnewsb.ATT.COM
Reply-To: neal@druhi.att.com (XGPB30000-McBurnettND(DR9225)289)
Distribution: comp.sys.handhelds,comp.sys.hp
Organization: AT&T Bell Labs, Denver CO
Lines: 50

In article <1990Mar13.232026.10379@cc.ic.acuk>, umapd51@suna.cc.ic.ac.uk
(W.A.C. Mier-Jedrzejowicz) writes:
> This is a followup to Neal McBurnett's followup to...
> Anyway, the year 4000 should be leap year - the rule is years
> divisible by 4 are leap - but century years are not - except
> century years whose first two digits are divisible by 4.

What I said was that the year 4000 "should not be" a leap year.  This
is based on the true length of the year, which we now know much more
accurately than Pope Gregory did.  The rule "should" be changed so that
years divisible by 4000 would not be leap years, but years divisible by
20000 would be.

I don't know if any formal decision (by who? the UN?) has been made
to change the rule, and I doubt that it would make sense to come to
such a decision so long before it really matters.  I doubt it would
make sense for HP to include code to take it into account.  What I
do believe is that it is stretching things to say that the HP
calendar is accurate over an 8400 year span, since 95% of that span
has not yet occurred - who knows what calendar we will be using
a decade or century from now, to say nothing of 8 millenia!
The claim of 8400 years was not made by HP, it was an minor comment
by someone who posted a wonderful review of the calculator.  I do
wish that the Julian dates from 1 AD to 1582 were dealt with.  Of
course it is straightforward to supplement HP's functions with my own,
but I have to use a different date format which can handle years BC
(maybe any negative date should be interpreted as simply representing
a negative year?)

> As for automatic accuracy adjustments to TIME and DATE, there is a
> utility to do that - included on the floppy disk you get if you buy an
> I/O kit instead of just the cable. My favourite is still the HP-41
> on this one, since it had functions to set the correction, but was
> smaller than the HP-71B and did not have the time bugs of the 71.

Oh boy - I can't wait to get the kit.  What I would really like to do
is hook the calculator up to the modem signal from the National Institute
of Standards and Technology (only a local call for me...) which allows
a computer to sync to within about 1 ms. accuracy.

But why is this "EXACT"-like utility bundled with a hardware purchase?
There seem to be at least two programs included in the I/O kit which
would be useful with or without the serial line (the other one being
the hp28->hp48 transfer code).  I hope that HP puts the "EXACT"-like utility
on their bbs (and on this newsgroup), like they did for the transfer code.

I agree with the others, though, the HP48 is way ahead of the HP28, which
was way ahead of the competition.

-Neal McBurnett // AT&T Bell Labs, Denver // neal@druhi.att.com, att!druhi!neal