TI 84 Plus Calculator sPi

[ixtapalapaquetl]

You are added onto my hero list. The use of the getTime function was smart. Is it confirmed that it is not possible on a 83/84?

Great work.

More efficient coding might be determined by the method in which we calculate Pi itself. Of course I will experiment, and clean it up. So this can be our first version.

This was all you brother. I just added a little patch at the end!

I am almost positive that the 89/92 (and likely Inspire) are the only ones with Clocks. So 83/84/86 are likely out of the running. Regarding the pi calculations, I seem to recall some infinite series which converge quite rapidly... it will take me a few days to dig anything up. I'll let you know if I find anything.

And Roger_D25, you should learn to love your inner nerd! Don't hide it! Let it sing loudly and let it sing off key!

[Alpha]

i know you can program in C for the Ti-89/92 as they are the same cpu, a 68k based chip and run much faster then the 82-86 conterparts. the only way to get time and your algorithm working is to add a counter alongside the calcutions and then have an average number based on your own clock as to how fast it works. because the 84 doesnt have an internal clock, you have no real easy way to track that.

this is one of my favorite algorithms and with it and some of my own programming skills i have able to find pi to an easy 150 digits

http://en.wikipedia.org/wiki/Bailey-...louffe_formula

[Rmrdr13]

My Ti-83 plus did it in about 1'29" give or take a few tenths.

It doesn't have a clock so i had to use a stopwatch.

[ixtapalapaquetl]

More good news! According to TI website, the "84 Plus family" has the clock function (enabled using MODE) and the requisite Gettime function! http://education.ti.com/educationpor...bid=4&topid=60 I don't have access to one, so someone else will have to tweak the syntax.

[Knight]

More good news! According to TI website, the "84 Plus family" has the clock function (enabled using MODE) and the requisite Gettime function! http://education.ti.com/educationpor...bid=4&topid=60 I don't have access to one, so someone else will have to tweak the syntax.

I have been playing with the getTime function on my Ti 84. If I enter getTime on my screen, the result in in the form "{xx xx xx}." The x's have would be replaced with numbers that represent hours, minutes, and seconds. I can't seem to assign it to a variable as well. (Didn't think so anyway.)

[Knight]

This will for sure work with a 83/84.

I got a time of 42/43 seconds. I even checked it with the Windows clock. It is for sure accurate to one second.

ClrHome
startTmr->H
1->S
0->C

For(A,1,4000,2)
(4/(S*A))->B
B+C->D
D->C
(-1*S)->S
End

checkTmr(H)->J

Disp "Pi ="
Disp (C+0.0005)    
Disp "Time:",J
Output(1,1,"")

[ixtapalapaquetl]

This will for sure work with a 83/84.

I got a time of 42/43 seconds. I even checked it with the Windows clock. It is for sure accurate to one second.

Congratulations! Mission accomplished! Now I want to know why the hell my trusty 89 so damn slow!!!

[Knight]

Congratulations! Mission accomplished! Now I want to know why the hell my trusty 89 so damn slow!!!

I will try to get my friends 89 for sure now. I would like to try the method you used, and the one that I used. (getTime vs. startTmr)

[FrogBite]

Is it possible to convert this equation to the TI?



That is: 2 plus the area between 0 and the sqrt of 8 of the above equation

[Knight]

Is it possible to convert this equation to the TI?



That is: 2 plus the area between 0 and the sqrt of 8 of the above equation

It is possible to integrate that function.

2+fnInt((sqt(4+(x-(sqt(8)/2))^2))-sqt(2),x,0,sqt(8)) sqt - Is for square root, ^2 is something raised to the second power.

I believe that you need to use >200 for 'r' to get a closer answer.

fnInt is for 83/84 calculators.
The 89 and otters have the actual integral sign.

I however got an answer of 4.09 by using that equation.



Edit: My friend with a 89 got a time of ~110s with my method. Strange that my 84 is faster.

[bluep3ace]

SUCCESS! This is so effin cool!

There are a few 89-related syntax modifications to your original code, but all I really did was add in a couple gettime()'s and converted the difference (in {hr,min,sec} format) to seconds in order to prevent negative values from popping up. Note that the Clock must be manually enabled: Home -> F1 -> B and then set time and date. Unfortunately, the Clock does not seem to work on Virtual TI, so we can't run the emulator realtime on our computers...

spi89()
Prgm
ClrIO
getTime()->ti
1->s
0->c
For a,1,4000,2
4/(s*a)->b
b+c->d
d->c
-1*s->s
EndFor
getTime()->tf
tf-ti->dt
3600*dt[1]+60*dt[2]+dt[3]->tpi
Disp "pi="
Disp c+.0005
Disp "time="
Disp tpi
EndPrgm

Check it out!

102 seconds!!! Does this qualify as the current WR? Time for some memory tweaks!!!

Great work Knight! Obviously, feel free to modify any ugly code - my programming skills are limited to TI calculators, and thus may be substandard!

iirc pi = 3.141592653589.........
error in the code? or internal rounding?
still, nice job on the WR

[Knight]

iirc pi = 3.141592653589.........
error in the code? or internal rounding?
still, nice job on the WR

It is a fault in the code. Only 4000 fractions were added up to produce pi. That is roughly accurate to 4 decimal places. It would take 20,000 or more fractions to correctly calculate pi to 8-10 decimal places.

For the sake of saving time, the value of our calculated pi was left to be inaccurate. pi != pi

[FrogBite]

It is possible to integrate that function.

2+fnInt((sqt(4+(x-(sqt(8)/2))^2))-sqt(2),x,0,sqt(8)) sqt - Is for square root, ^2 is something raised to the second power.

I believe that you need to use >200 for 'r' to get a closer answer.

fnInt is for 83/84 calculators.
The 89 and otters have the actual integral sign.

I however got an answer of 4.09 by using that equation.



Edit: My friend with a 89 got a time of ~110s with my method. Strange that my 84 is faster.

graph it and add two to the area value from 0 to sqrt(8). You get the same value of pi as saved on the calc.

sorry there was an sign mistake in the equation. should work now.

[ZOMGVTEK]

I got my TI-83+ 8Mhz (6Mhz Stock clock) running safe at 19Mhz with a few tweaks, its wicked fast, but sometimes while grahing, it throws artifacts...

I think i need more voltage, and a little better cooling.

[Knight]

I got my TI-83+ 8Mhz (6Mhz Stock clock) running safe at 19Mhz with a few tweaks, its wicked fast, but sometimes while grahing, it throws artifacts...

I think i need more voltage, and a little better cooling.

I might try to make my calculator slower, so that I can cheat at some games.

Hax0r?

[71 (Bryan)]

My goal will be to over clock my calculator.

i was saying that to my cousin just last weak, there must be a way.
if any on finds out how, let me know

[Knight]

i was saying that to my cousin just last weak, there must be a way.
if any on finds out how, let me know

Here you go.

http://www.ticalc.org/hardware/overclocking/

Pick your calculators number and click its boost link.