CCS C Software and Maintenance Offers
FAQFAQ   FAQForum Help   FAQOfficial CCS Support   SearchSearch  RegisterRegister 

ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

CCS does not monitor this forum on a regular basis.

Please do not post bug reports on this forum. Send them to CCS Technical Support
POW() issues

 
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion
View previous topic :: View next topic  
Author Message
KamPutty
Guest
POW() issues
PostPosted: Fri Aug 18, 2006 4:43 pm     Reply with quote
Hi all,

It seems that there is an issue with version 3.234 and the pow() method.

Here are my results for the following:
Code:

void test1()
{
    float power0=0;
    float power1=0;
    float power2=0;
    float power3=0;
    float power4=0;

    printf("----\r\n");
    power0=pow(10, 0);
    power1=pow(10, 1);
    power2=pow(10, 2);
    power3=pow(10, 3);
    power4=pow(10, 4);

    printf("pow(10, 0)=%f\r\n", power0);
    printf("pow(10, 1)=%f\r\n", power1);
    printf("pow(10, 2)=%f\r\n", power2);
    printf("pow(10, 3)=%f\r\n", power3);
    printf("pow(10, 4)=%f\r\n", power4);
}

.
.
.
....................     power0=pow(10, 0); 
1A7C:  MOVLW  82
1A7E:  MOVWF  x29
1A80:  MOVLW  20
1A82:  MOVWF  x22
1A84:  CLRF   x23
1A86:  CLRF   x24
1A88:  CLRF   x25
1A8A:  CLRF   x26
1A8C:  CLRF   x27
1A8E:  CLRF   x28
1A90:  MOVFF  224,22C
1A94:  MOVFF  223,22B
1A98:  MOVWF  x2A
1A9A:  MOVFF  228,230
1A9E:  MOVFF  227,22F
1AA2:  MOVFF  226,22E
1AA6:  MOVFF  225,22D
1AAA:  MOVLB  0
1AAC:  RCALL  16F2
1AAE:  MOVFF  03,210
1AB2:  MOVFF  02,20F
1AB6:  MOVFF  01,20E
1ABA:  MOVFF  00,20D
....................     power1=pow(10, 1); 
1ABE:  MOVLW  82
1AC0:  MOVLB  2
1AC2:  MOVWF  x21
1AC4:  MOVLW  20
1AC6:  MOVWF  x2A
1AC8:  CLRF   x23
1ACA:  CLRF   x24
1ACC:  MOVLW  7F
1ACE:  MOVWF  x25
1AD0:  CLRF   x26
1AD2:  CLRF   x27
1AD4:  CLRF   x28
1AD6:  MOVFF  224,22C
1ADA:  MOVFF  223,22B
1ADE:  MOVFF  221,229
1AE2:  MOVFF  228,230
1AE6:  MOVFF  227,22F
1AEA:  MOVFF  226,22E
1AEE:  MOVWF  x2D
1AF0:  MOVLB  0
1AF2:  RCALL  16F2
1AF4:  MOVFF  03,214
1AF8:  MOVFF  02,213
1AFC:  MOVFF  01,212
1B00:  MOVFF  00,211
....................     power2=pow(10, 2); 
1B04:  MOVLW  82
1B06:  MOVLB  2
1B08:  MOVWF  x21
1B0A:  MOVLW  20
1B0C:  MOVWF  x2A
1B0E:  CLRF   x23
1B10:  CLRF   x24
1B12:  MOVLW  80
1B14:  MOVWF  x25
1B16:  CLRF   x26
1B18:  CLRF   x27
1B1A:  CLRF   x28
1B1C:  MOVFF  224,22C
1B20:  MOVFF  223,22B
1B24:  MOVFF  221,229
1B28:  MOVFF  228,230
1B2C:  MOVFF  227,22F
1B30:  MOVFF  226,22E
1B34:  MOVWF  x2D
1B36:  MOVLB  0
1B38:  RCALL  16F2
1B3A:  MOVFF  03,218
1B3E:  MOVFF  02,217
1B42:  MOVFF  01,216
1B46:  MOVFF  00,215
....................     power3=pow(10, 3); 
1B4A:  MOVLW  82
1B4C:  MOVLB  2
1B4E:  MOVWF  x21
1B50:  MOVLW  20
1B52:  MOVWF  x2A
1B54:  CLRF   x23
1B56:  CLRF   x24
1B58:  MOVLW  80
1B5A:  MOVWF  x2D
1B5C:  MOVLW  40
1B5E:  MOVWF  x26
1B60:  CLRF   x27
1B62:  CLRF   x28
1B64:  MOVFF  224,22C
1B68:  MOVFF  223,22B
1B6C:  MOVFF  221,229
1B70:  MOVFF  228,230
1B74:  MOVFF  227,22F
1B78:  MOVWF  x2E
1B7A:  MOVLB  0
1B7C:  RCALL  16F2
1B7E:  MOVFF  03,21C
1B82:  MOVFF  02,21B
1B86:  MOVFF  01,21A
1B8A:  MOVFF  00,219
....................     power4=pow(10, 4); 
1B8E:  MOVLW  82
1B90:  MOVLB  2
1B92:  MOVWF  x21
1B94:  MOVLW  20
1B96:  MOVWF  x2A
1B98:  CLRF   x23
1B9A:  CLRF   x24
1B9C:  MOVLW  81
1B9E:  MOVWF  x25
1BA0:  CLRF   x26
1BA2:  CLRF   x27
1BA4:  CLRF   x28
1BA6:  MOVFF  224,22C
1BAA:  MOVFF  223,22B
1BAE:  MOVFF  221,229
1BB2:  MOVFF  228,230
1BB6:  MOVFF  227,22F
1BBA:  MOVFF  226,22E
1BBE:  MOVWF  x2D
1BC0:  MOVLB  0
1BC2:  RCALL  16F2
1BC4:  MOVFF  03,220
1BC8:  MOVFF  02,21F
1BCC:  MOVFF  01,21E
1BD0:  MOVFF  00,21D
.
.
.


This is my results from hyperterminal:

Quote:

pow(10, 0)= 1.00
pow(10, 1)= 10.00
pow(10, 2)= 99.99 <--- Whats THIS?!!!
pow(10, 3)=1000.00
pow(10, 4)=9999.99 <--- Whats THIS?!


This is killing me. I just spent about 1 hour tracking this error down and it's not my error! Sad Sad Sad

What am I supposed to do?! Apart from writing my own pow method....



~Kam (^8*
Guest
PostPosted: Fri Aug 18, 2006 5:11 pm     Reply with quote
Welcome to CCS world!

I've spent a total of 8 hours this week trying to figure out a few different problems, one was the help file's fault, 2 were compiler faults.

I think the best way to go about it is don't completely 'trust' the compiler. Get ready to get a datasheet and work it out.... or an asm reference in your case.
Guest
PostPosted: Fri Aug 18, 2006 5:52 pm     Reply with quote
I should just write it all in asm and forgo the compiler! Confused

Well, I'm off to Hawaii...no compiler issues for the next week!

~Kam (^8*
jma_1



Joined: 08 Feb 2005
Posts: 147
Location: Wisconsin

View user's profile Send private message

PostPosted: Sat Aug 19, 2006 11:27 am     Reply with quote
Greetings,

Look at floating point representation of numbers in embedded systems (and all computers) for the problem. Slight errors in the representation and storage of floating point numbers is what you are seeing. The compiler and assembly language are correct. Rewriting in assembly will produce the same results.

If you change your floating point operation to integers (int16 or int32 w/ CCS), the output will be what you expect. As an alternative, round the result of the pow() and/or limit the decimal places.
Display posts from previous:   
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion All times are GMT - 6 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group