Prev: ABC video player
Next: any with Graphing Calculator or Nucalc 3.5, or other good andbad calculators?
From: Autymn D. C. on 20 Sep 2009 04:39
I need ye to find e^(π√67) - 5280^3 - 724, at best precision, and post
the answer here and by which program. Other/Older versions of the
same are welcome. Also tell here your computer if you care. I'm
about to post the below at Apple Support about the bugs in Calculator,
and need the poll list.

-Aut

Re: Calculator App version incorrectly calculating Log and powers
Apple's misCalculator
What version is everyone's Calculator? 4.0.6 is mine and it's full of
floating point bugs, loppoffs as above by order of operations or
operands*, loppoffs of small EEs which wreck the whole equation,
dropped bigger operations/operators when compounded, errors upon
operator fiddling (− ± −; − ± ±; N ^x√y M ^x√y = NM; N ^x√y M ^x√y M ≠
N ^x√y (M ^x√y M); 2 y^x 1023 = +/− 1), no delete for EEs, weird
characters show up in ASCII/Unicode view in Programmer, poor stack
handler (170! *; 170! M+ C MR *; 2 y^x 1023 +/−/×/÷) whereas Magic
Number Machine (older) does 1000! in a flash—its window is a pretty
blend of Calculator's full keypads and Graphing Calculator's equation
editor, and it has drawers for statistics also!—lack of imaginary/
complex and production (or Pi function) domains for (<0)!, and likely
more—which bugs hav ye seen? Here were some bugs in Panther's version
(which was?): http://groups.google.com/group/comp.sys.mac.advocacy/msg/fb81cc7ad7e34338.
By the way, Graphing Calculator was neat, and Grapher is awesom, but
mine hangs and crashs when I set contours (on z=y^(x-1)) on a plot
with axes of -6..6 rather than its default -5..5.

*A workaround is a delta: 1000.1 1/x + 1500 1/x = 1/x. Calculator
doesn't recognise 1000., another bug for above.

For fun I'd like to see a version history for Calculator, if there is
one... X.iv.iv had 4.0.4. And it and every other app needs to be in
Mactracker.

Calculator is woefully inaccurate. e^(π√67) - 5280^3 - 724:

Google: 19.9992676
Grapher 1.0 aka Curvus Pro (2005, 14 digits): 19.99960327148438
Graphing Calculator 1.3, aka NuCalc (14 digits): 19.999816894531
Calculator 4.0.6 (2005, X.iv.xii, 16 digits): 19.9998168945312
Magic Number Machine 1.0.11 (2004, 21 of 24 digits):
19.99999866245422449603
Magic Number Machine 1.0.30 (2009, 20 of 24 digits):
19.99999866245422449622; er, where's the 25th digit it advertises? or
the 23d? Why are the significant figures not the same as the digits?

Three transcendentals are in the first term, and Calculator's off by
1/100000. How is Windows?

Uh oh, in MNM: 0.461632144958362341262! = Not a number. The Pi or
Ghamma function isn't maghic enouh for MNM.
From: Rotten Apple on 20 Sep 2009 14:47

"James Waldby" <no(a)no.no> wrote in message
news:rq-dnX4-JZ9w-CvXnZ2dnUVZ_uCdnZ2d(a)bresnan.com...
> [This topic's irrelevant in several of the listed newsgroups, but I
> didn't trim the list as I don't know where you are reading. You could
> set followups when you have an egregiously long list of newsgroups.]
> [See bc results below]
>
> On Sun, 20 Sep 2009 01:39:27 -0700, Autymn D. C. wrote:
>
>> I need ye to find e^(π√67) - 5280^3 - 724, at best precision, and post
>> the answer here and by which program. Other/Older versions of the same
>> are welcome. [...]
>
>> Google: 19.9992676
>> Grapher 1.0 aka Curvus Pro (2005, 14 digits): 19.99960327148438 Graphing
>> Calculator 1.3, aka NuCalc (14 digits): 19.999816894531 Calculator 4.0.6
>> (2005, X.iv.xii, 16 digits): 19.9998168945312 Magic Number Machine
>> 1.0.11 (2004, 21 of 24 digits): 19.99999866245422449603
>> Magic Number Machine 1.0.30 (2009, 20 of 24 digits):
>> 19.99999866245422449622; er, where's the 25th digit it advertises? or
>> the 23d? Why are the significant figures not the same as the digits?
> ...
>
> Here are some bc 1.06 results per my linux 2.6 system. Probably
> about 45 digits of the 60-digit result are correct, and about 107
> digits of the 120-digit result, by comparison with the leading
> digits of the 180-digit result.
>
> scale=60; pi=4*a(1); e(pi*sqrt(67)) - 5280^3 - 724
> 19.999998662454224506829261312578628508183312503815806586355322
>
> scale=120; pi=4*a(1); e(pi*sqrt(67)) - 5280^3 - 724
> 19.999998662454224506829261312578628508183312503816712633371282105122950998831523502041379242353370629039564711064649074090
>
> scale=180; pi=4*a(1); e(pi*sqrt(67)) - 5280^3 - 724
> 19.9999986624542245068292613125786285081833125038167126333712821051229509988315235020413792423533706290395647152488070416966...
>
> Here are two of your values, lined up with bc's for comparison:
> 19.99999866245422449603
> 19.99999866245422449622
> 19.9999986624542245068292613125786285081833125038167...
>
> --
> jiw

The OP certainly will never get an answer from CSMA, where they can only
tell you the Mac is best, not how to actually make it do what you need it to
do!

From: Autymn D. C. on 23 Sep 2009 16:09
On Sep 20, 11:47 am, "Rotten Apple" <rot...(a)pple.com> wrote:
> The OP certainly will never get an answer from CSMA, where they can only
> tell you the Mac is best, not how to actually make it do what you need it to
> do!

What do you "need" the Mac to do?

-Aut
aware of Ncalc and arbitrary precision calculators

Nobody has Graphing Calculator or Nucalc 3.5? How about Maple,
Matlab, Mathematica, etc.?
From: Rotten Apple on 23 Sep 2009 16:33
I'm not the OP.

--

"I never mentioned that I couldn't afford to buy a Mac.
I said I couldn't afford to buy any computer." -- Dave Fritzinger

"Autymn D. C." <lysdexia(a)sbcglobal.net> wrote in message
news:a4eaa0ca-6e47-48ec-ab79-c6ed13788f86(a)q40g2000prh.googlegroups.com...
On Sep 20, 11:47 am, "Rotten Apple" <rot...(a)pple.com> wrote:
> The OP certainly will never get an answer from CSMA, where they can only
> tell you the Mac is best, not how to actually make it do what you need it
> to
> do!

What do you "need" the Mac to do?

-Aut
aware of Ncalc and arbitrary precision calculators

Nobody has Graphing Calculator or Nucalc 3.5? How about Maple,
Matlab, Mathematica, etc.?


From: Marc Heusser on 27 Sep 2009 08:45
In article
<a3ae4efe-452e-432e-aca5-7f9a62667124(a)x6g2000prc.googlegroups.com>,
"Autymn D. C." <lysdexia(a)sbcglobal.net> wrote:

> I need ye to find e^(ｹﾃ67) - 5280^3 - 724, at best precision, and post
> the answer here and by which program.

Best is too long, here are the first 1000 places from Mathematica
7.0.1.0 on Mac OS X 10.6.1 on a MBP:

In[12]:= N[E^(Pi 67^(1/2)) - 5280^3 - 724, 1000]

Out[12]= ¥
19.9999986624542245068292613125786285081833125038167126333712821051229¥
5099883152350204137924235337062903956471524880704169663194152335947423¥
7247878165181328956918107617412298168050068584051548935546606492552193¥
2150684101786225435785261996215140852650105109056320696104349773580556¥
5423417889465000855352953229796386873263540393090632856130098056111179¥
6376267919912546165980132765175353482799070432483463787362975675595004¥
8758183126688753303422158498741419589744239191248900047790604995377262¥
8719643718294875677352508842970083897879288693711471012745612064501981¥
7900708888252632370683489799641789076547364930333194096515431772093639¥
3509913260861897462692518520383245713817639149089489615977154391845954¥
5788509481967128676509628925325666129452798918072739477172785208043565¥
0367941668080591540115540604771752801288638441429093557529037161883462¥
6186410377295994348525206437170608432714483896371653340100257298269499¥
2066249754746018061399727932220589972306069513239283192319407039999536¥
551231630421101084968

The same calculated with machine precision, which is all you can ask for
in usual programmes:

In[13]:= E^(Pi 67^0.5) - 5280^3 - 724

Out[13]= 19.9998

Bear in mind that this is what you do:

In[14]:= E^(Pi 67^0.5)


Out[14]= 1.47198*10^11

HTH

Marc

--
remove bye and from mercial to get valid e-mail
<http://www.heusser.com>
 |  Next  |  Last
Pages: 1 2
Prev: ABC video player
Next: any with Graphing Calculator or Nucalc 3.5, or other good andbad calculators?