Decimal to fraction conversion

[ErnieM]

tico,

(ignores all the hard feelings going back and fourth)

As far as your original problem goes, how do you intend to hold a fractional number in memory? PBP only handles integers (sad but true).

So you are limited to schemes such as working with your values scaled up by say 1,000, so 1/16 would be stored as 62 (1/16 * 1000 and truncated to an integer) and you can divide by 62 to see how many 1/16 pieces it has.

[tico]

tico,

(ignores all the hard feelings going back and fourth)

As far as your original problem goes, how do you intend to hold a fractional number in memory? PBP only handles integers (sad but true).

So you are limited to schemes such as working with your values scaled up by say 1,000, so 1/16 would be stored as 62 (1/16 * 1000 and truncated to an integer) and you can divide by 62 to see how many 1/16 pieces it has.

I thought that I could use the method described in post 2 or 3

divide say 11.5 to get the following results
5/16 Vector1
15/16 Vector2
1 11/16 Vector3
2 5/8 Vector4
3 15/16 Vector5
5 3/4 Vector6Center
7 9/16 Vector7
8 7/8 Vector8
9 13/16 Vector9
10 9/16 Vector10
11 3/16 Vector11

I was hoping that this is possible to do with PBP, If I am wrong plese let me know. I am going to begin writing some code to do this tommorow.


opps forgot the formula..... as in excel
Round ((11.5/2) -((11.5/2)*(0.9)^0.5),3 = 5/16
I am hoping that PBP can process this type of formula, if not its back to the drawing board

[T.Jackson]

Round ((11.5/2) -((11.5/2)*(0.9)^0.5),3 = 5/16
I am hoping that PBP can process this type of formula, if not its back to the drawing board

PBP would have a heart attack with that.

[mister_e]

Yes and no, PBP have it's limit... it's not something new. But decent math approach would save the day.

To make a short story, PBP works with integer, nowhere in the manual you'll find an answer about floating points and or square power.

Thinking never killed... sure you use the easy route... but...

i think few people may need to relax or simply be ban out of here...

[Acetronics2]

Hi, Steve

Why not change the formula to :

16 x 115 / 4 x ( 200 - SQR ( 36 000 )) = 5 000 ( exactly 4721. .... ! PbP will return 5060 here ... )

PbP would do that without any pain ... but not very precisely, as you see !!!

Alain

Ok, ... also just look for possible overflows !!!

[tico]

Oh boy looks like I got over eager with another project, well I will see what I can do to get some sort of formula working in PBP.
Looks like I will be doing alot of debug and lcdout on this one.

[skimask]

Oh boy looks like I got over eager with another project, well I will see what I can do to get some sort of formula working in PBP.
Looks like I will be doing alot of debug and lcdout on this one.

Don't forget about the floating point math routine package that Microchip has (and are at the MeLabs site). Those subroutines will handle the math that PBP won't...but you'll have to learn how to interface with them...

[Acetronics2]

Hi, Tico

I'm curious to see where your formula came out ... and what used for.

looks too much like a triangle solving and inch converting !!!

Alain

PS ... see also here for math ... http://www.micromegacorp.com/pbp.html

a V3 chip is also available ... don't hesitate to browse !!! http://www.micromegacorp.com/umfpu-v3.html