Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

PIC 18F series performance

Status
Not open for further replies.

JoKKeR

Full Member level 2
Joined
Feb 14, 2008
Messages
130
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,298
Location
Estonia
Activity points
2,281
Hi !

Can 8 bit PIC MCU handle a floating point calculations and how many floating point variables it can operate with ?

PIC need to do multiple floating point calculations (progressive) ...

So far ... im aiming at F18 series for a project, because of performance, but i know little about F10 and F12 series.

My plan is to drive a small 2x16 or 1x16 LCD, to have max 4 ADC-s, some digital IO pins and interrupt function. Probably 18 pin DIP package will do a job.


Thank you !
 

JoKKeR

Full Member level 2
Joined
Feb 14, 2008
Messages
130
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,298
Location
Estonia
Activity points
2,281
blueroomelectronics said:
The 18F1320 is a nice 18pin PIC.
Sure it is ;) . Can it handle multiple floating point calculations ?
 

helios31

Junior Member level 1
Joined
Jul 17, 2009
Messages
15
Helped
3
Reputation
6
Reaction score
0
Trophy points
1,281
Location
France
Activity points
1,364
Hewlett Packard used 4 bit processors on many calculators...
 

zx_sa

Junior Member level 3
Joined
Oct 3, 2007
Messages
25
Helped
6
Reputation
12
Reaction score
0
Trophy points
1,281
Location
UK
Activity points
1,471
You can do floating point calculation on any processor - it's just software that does the calculation. The question is whether the software routines will be fast enough for your application.

You need to give much more information for someone to calculate if the PIC will be fast enough.

If you need to do any sort of real-time filtering of input signals, you should instead look at using a DSP-type processor. PIC does have a DSP range as well, but I've never really looked at them.
 

    JoKKeR

    points: 2
    Helpful Answer Positive Rating

JoKKeR

Full Member level 2
Joined
Feb 14, 2008
Messages
130
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,298
Location
Estonia
Activity points
2,281
I realize the fact that pic can do a floating point calculation, there are float.h library in c compiler, that makes it much more easier.

Since calculations need to be progressive and floating point ones, i thinking about usage of RAM, because this could be the problem.
 

zx_sa

Junior Member level 3
Joined
Oct 3, 2007
Messages
25
Helped
6
Reputation
12
Reaction score
0
Trophy points
1,281
Location
UK
Activity points
1,471
What is your application? What is the mathematical function that you have to apply? How much data throughput (ie kilobytes of data per second)?
 

JoKKeR

Full Member level 2
Joined
Feb 14, 2008
Messages
130
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,298
Location
Estonia
Activity points
2,281
Eeh ... Hard to answer this question.

The application is heat system regulator. It is devise which turns valve depending on sensor data. There is a small DC motor and a gear mechanism.

My task will be to redesign the system.
The old system worked on OP amps.

''I was told that calculations need to be floating point''. Because of 0 Progress and my little experience i may assume that i need to use algorithm ( calculations ) witch are progressive 3 maybe 4 steps. So i need to store more than 3 floating point variables.

So far as i know PIC MCU-s got 128 bit of RAM ... so i got no idea how much RAM will it it take to run application.

About mathematical functions ... i may assume, they are linear. ;)

Sorry for poor explanation of a problem :(
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top