HP Forums
TI Nspire and Bad Math - Printable Version

+- HP Forums (https://archived.hpcalc.org/museumforum)
+-- Forum: HP Museum Forums (https://archived.hpcalc.org/museumforum/forum-1.html)
+--- Forum: Old HP Forum Archives (https://archived.hpcalc.org/museumforum/forum-2.html)
+--- Thread: TI Nspire and Bad Math (/thread-105889.html)

TI Nspire and Bad Math - Chuck - 01-12-2007

Don't you just hate bad math? Check out TI's website on the Nspire. It seems we have a new form for the quadratic formula (screen shot #5). Didn't their previous graph of teh parabola clue them in to the wrong solutions? Hmmm?




Re: TI Nspire and Bad Math - Jan - 01-12-2007

Should I laugh or just become angry. I am still not sure. This would not have happened in an old HP advert.

Re: TI Nspire and Bad Math - Frank Boehm (Germany) - 01-12-2007

At least "0" is not offered as a solution <g>

Re: TI Nspire and Bad Math - Gerson W. Barbosa - 01-13-2007

This would not have happened in an old HP advert.

Don't be so sure of that. This is from an old '93 HP folder:


Error in 1993 Equation Writer brochure - Karl Schneider - 01-13-2007

Hah! Good retort, Gerson!

(On the other hand, maybe 1993 isn't "old" in the sense of classic HP; the attention to detail in the product literature was slipping by then.)

The error I see in the algebraic expression is that 2*pi should have been parenthesized, so that the square root would be applied to both numbers.

This is an example of my basic objection to EOS calculators: The expression you see in the display (which is often cropped) just doesn't match the graphical expression on the paper. Therefore, it is difficult to ascertain whether the entered expression was correct, yet the user is expected to trust the final result returned by hitting "=".

The HP-48 Equation Writer helped to avert errors such as these, but it's kind of a chore to enter a graphical expression using the cursor arrows.

-- KS

Edited: 13 Jan 2007, 3:26 p.m.

Re: TI Nspire and Bad Math - Karl Schneider - 01-13-2007

Wow! That's difficult to excuse. The incorrect second result stems from the incorrect second-listed quadratic-equation formula, and clearly does not match the graph, which is correct.

My question is, how did that occur? Wouldn't the "Nspire" automatically generate both correct quadratic-equation formulas? Many calculators have the quadratic equation built in, because it is so applicable to many problems.

Sloppy and sub-mediocre!

-- KS

Re: Error in 1993 Equation Writer brochure - Gerson W. Barbosa - 01-13-2007

Hello, Karl!

You're right. When Jan said "old HP advert" he surely was neaning "prior to 1986". But I found that HP mistake interesting to show here. The same picture appears in old 48 boxes, I think. I placed a pair of parentheses by hand on the brochure. I am not sure I edited the scan I sent to Dave to be included in the next DVD.



Edited: 13 Jan 2007, 4:48 p.m.

Re: TI Nspire and Bad Math - Chuck - 01-13-2007

You can bet it wasn't Nspire that made the error, but rather the calc-operator and screen-capturer made the error. They most likely typed in the supposed quadratic equation to demonstrate the substitution ability (also kind of false advertizing) and made a typo in the formula. I can certainly atest to making typos myself, but I'd try not to do it on a major web page touting thelatest software and calculator. I sent an email to TI to inform them. We'll see how long it takes them to change it, or if they really care at all about it.

Older Errors - Palmer O. Hanson, Jr. - 01-15-2007

Hah! Good retort, Gerson!

... maybe 1993 isn't "old" in the sense of classic HP; the attention to detail in the product literature was slipping by then.)

The HP-41C Financial Decisions Pac Addendum Card (00041-90273 3/80) printed in March 1980 provided two corrections which applied to both the original manual (00041-90042) and to the Revision B manual printed in January 1980. The first correction explains that "the prompts for the Modified Internal Rate of Return (MIRR) were inadvertently reversed". The second correction explains that "The Bonds program does not correctly calculate the yield of a short-term bond" and provides a method to remedy the problem.

Even earlier, the HP-27 Owner's Handbook (00027-90012 which does not have a date on the card) provided corrections to three pages of the manual (00027-90001) which was printed in January 1976.

Maybe things weren't so pristine in the olden days after all.

And don't forget the HP41 games pac - Gene - 01-15-2007

misspelled a "shields" prompt as "sheilds"

Mistakes happen. Even in the PPC ROM.

Re: Older Errors - Les Wright - 01-15-2007

When I was student I had occasion to actually try to use the Enzyme Kinetics program in the Chemistry User Solutions Book. Discovered after getting wonky results for the Hofstee plot option that several steps intended to transform certain regression coefficients were just plain wrong in the listing and, I have discovered now that I have a wand, in the barcode.

Recently, I have taken interest in the Root Finder code in the HP41 Standard Applications Pac. The code works, but it seems to include redundant steps (i.e storing interim results that are never again recalled) and has a lot of inefficient filler. It uses the Illinois modification of the secant method, and as such is a pretty good routine. I have easily shaved it down to 59 steps from the original 75, and have easily adapted it to the 11C. It is shorter and seems to work better than the adaptation of Newton's Method that is offered in the 11C manual.

I don't believe everything was perfect in the golden olden days. As for the PPC manual--how could it avoid some mistakes, as the thing is a monster!


Re: Older Errors - Les Wright - 01-15-2007

Whoops! The Illinois method is a modification of the false position method, not the (admittedly quite similar) secant method.