# Anybody interested in a Quadratic Solver?

Show 40 post(s) from this thread on one page
Page 3 of 4 First 1234 Last
• Dec 18th 2007, 03:50 AM
janvdl
Quote:

Originally Posted by DavidB
What kind of a tool are you planning to turn it into?
I might be interested in helping with it.

Well currently it features a quadratic solver, logs to any base, it can convert expressions into log form, a rad to deg converter and vice versa, compound interest, a basic trig calculator(rad and deg supported), and the decimal to binary converter. :)

If you can help I'd appreciate it. (I'm using Delphi 7 by the way)
• Dec 18th 2007, 07:06 PM
topsquark
Quote:

Originally Posted by DivideBy0
Hold on, how could I only just have noticed it?!

Some cubics have three solutions, yet that formula will always only yield 1! :o

However in those cases it does work :)

Well, that's easy to fix. Once janvdl has one solution to the cubic he can program his solver to do the division to reduce it to a quadratic, which his program can already do. ;)

-Dan
• Dec 20th 2007, 07:26 AM
janvdl
Quote:

Originally Posted by topsquark
Well, that's easy to fix. Once janvdl has one solution to the cubic he can program his solver to do the division to reduce it to a quadratic, which his program can already do. ;)

-Dan

I am starting work on the cubic solver today. I am not able to post a lot on MHF anymore because the ADSL cap on my line is finished :mad:

But I'll keep everyone informed on how it's progressing. ;)
• Dec 24th 2007, 03:25 AM
janvdl
The cubic solver seems to be working okay now. I spent a bit of time on the exponent function of the program, because it didn't allow for a negative base, but it's fixed now. ;)

Anyway, i still face one problem. Lets say we only have the function $x^3 = 0$ then the co-efficients of $x^2$ and $x$ are zero. But then i am faced with "Division by Zero" errors. Does anyone know how to make the application ignore such errors?
• Dec 30th 2007, 06:07 AM
janvdl
Okay, this is like a Beta Version for now, as the cubic solver still has a few bugs that need to be sorted out, and it could do with some enhancement.

Nevertheless, let me know what you guys think of this version.

• Jan 1st 2008, 12:34 AM
earboth
Quote:

Originally Posted by janvdl
Okay, this is like a Beta Version for now, as the cubic solver still has a few bugs that need to be sorted out, and it could do with some enhancement.

Nevertheless, let me know what you guys think of this version.

Good morning and a happy New Year!

I've played around with your program and it works fine. Congratulation!

Only one point: Please check the procedures calculating angles in degrees and in radians. In my opinion you maybe have mixed them up.
• Jan 1st 2008, 01:19 AM
earboth
Quote:

Originally Posted by earboth
...In my opinion you maybe have mixed them up.

Hello,

only to demonstrate what I meant:
• Jan 1st 2008, 02:26 AM
janvdl
Quote:

Originally Posted by earboth
Good morning and a happy New Year!

I've played around with your program and it works fine. Congratulation!

Only one point: Please check the procedures calculating angles in degrees and in radians. In my opinion you maybe have mixed them up.

May you have a happy and prosperous new year too, Earboth :)

Thanks for the spot, I'll recheck the code.
• Jan 9th 2008, 05:07 AM
janvdl
Well here's Math Tool 3.4 :cool:

With a brand new look as well as a graphing tool(Thanks Earboth!) :)

• Jan 9th 2008, 05:17 AM
colby2152
Quote:

Originally Posted by janvdl
Well here's Math Tool 3.4 :cool:

With a brand new look as well as a graphing tool(Thanks Earboth!) :)

Wow, that's pretty good janvdl! You should look into expanding your function grapher for your next edition!
• Jan 9th 2008, 05:19 AM
janvdl
Quote:

Originally Posted by colby2152
Wow, that's pretty good janvdl! You should look into expanding your function grapher for your next edition!

All credit to Earboth for the function grapher. He gave me the complete source code to compile into the application. :)
• Jan 9th 2008, 08:57 AM
Jhevon
i guess we can't call it a quadratic solver anymore... what would be a good name?
• Jan 9th 2008, 10:31 AM
Jhevon

my thoughts:

cubic solver: i still don't like the fact that it gives only one solution and leaves it to the user to find the other two, which he/she may have to do long division to find. this will be a pain, especially if the first root given is complex or irrational

logarithms: it cannot find the natural log...that's like the most common log in calculus. we must fix this. you need to tell it how to recognize e. i don't know the best way to do this for this utility. it may be difficult to code the limit definition and tell the tool how to evaluate it. it may suffice to just input the number, to like 30 decimal places or something, i don't know. how did you code pi?

compound interest: i won't go as far as to say perfect, i didn't play around with this feature that much, but it seems to do the job. the only suggestion i have off-hand is why don't you teach it to solve for any of the unknowns. currently, it can only solve for A, but what if i had A and wanted to solve for the interest rate?

degrees/radians: you still have these mixed up, i think you forgot earboth's suggestion here. by the way, the program cannot recognize "pi," i guess we have to use the decimal expansion. how many decimal places did you use to define pi?

trig calculator: perfect so far. you may want to teach it how to do the inverse trig functions, if possible. as well as code secant, cosecant and cotangent, if you have time

binary converter: perfect. you may want to teach it to go the other way though, that is, from binary to decimal

grapher: the picture is too small. i like how you graphed the derivative and second derivative as well. but maybe there should be an option to turn this feature off, probably not necessary though

About: nice! i like the what's new section. it gives us a report on how the utility has evolved

Math Help Forum: who doesn't like MHF?
• Jan 9th 2008, 10:43 AM
colby2152
Quote:

Originally Posted by Jhevon
i guess we can't call it a quadratic solver anymore... what would be a good name?

MHF QuickSolver
• Jan 9th 2008, 10:06 PM
janvdl
Quote:

Originally Posted by Jhevon

Thank You :D
Quote:

Originally Posted by Jhevon
cubic solver: i still don't like the fact that it gives only one solution and leaves it to the user to find the other two, which he/she may have to do long division to find. this will be a pain, especially if the first root given is complex or irrational

I found the other two formulae for the other 2 roots, but it works with complex roots and is much different from the first formula that DivideBy0 gave me. And I dont really know how to work with that "i" thing in complex roots.
Quote:

Originally Posted by Jhevon
logarithms: it cannot find the natural log...that's like the most common log in calculus. we must fix this. you need to tell it how to recognize e. i don't know the best way to do this for this utility. it may be difficult to code the limit definition and tell the tool how to evaluate it. it may suffice to just input the number, to like 30 decimal places or something, i don't know.

That will be easy :)
Quote:

Originally Posted by Jhevon
how did you code pi?

function Pi (Extended)
begin
Result := 3.1415... (I'm not going to type all 30 decimal places here) :D
end;
Quote:

Originally Posted by Jhevon
compound interest: i won't go as far as to say perfect, i didn't play around with this feature that much, but it seems to do the job. the only suggestion i have off-hand is why don't you teach it to solve for any of the unknowns. currently, it can only solve for A, but what if i had A and wanted to solve for the interest rate?

That is very possible, and not hard to do either, just requires a bit of work.
Quote:

Originally Posted by Jhevon
degrees/radians: you still have these mixed up, i think you forgot earboth's suggestion here. by the way, the program cannot recognize "pi," i guess we have to use the decimal expansion. how many decimal places did you use to define pi?

I looked it up on the internet and my formulae look correct. But I will purposefully "mix them up" in the next version. Or I will post the code here. Chances are I did something stupid again.
Quote:

Originally Posted by Jhevon
trig calculator: perfect so far. you may want to teach it how to do the inverse trig functions, if possible. as well as code secant, cosecant and cotangent, if you have time

Tried doing so. Never arrived at the original value again. I have no idea why that happened. It's already programmed, just hidden.
Quote:

Originally Posted by Jhevon
binary converter: perfect. you may want to teach it to go the other way though, that is, from binary to decimal

Sure, I dont think that would be too hard.
Quote:

Originally Posted by Jhevon
grapher: the picture is too small. i like how you graphed the derivative and second derivative as well. but maybe there should be an option to turn this feature off, probably not necessary though

I kind of wanted everything neatly in one size form, but I have to agree, the picture is a little small.
Quote:

Originally Posted by Jhevon
About: nice! i like the what's new section. it gives us a report on how the utility has evolved

Thought you might find it interesting :D
Quote:

Originally Posted by Jhevon
Math Help Forum: who doesn't like MHF?

I quite agree. :D
Show 40 post(s) from this thread on one page
Page 3 of 4 First 1234 Last