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

Question about sprintf and float variables

 
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion
View previous topic :: View next topic  
Author Message
ccdart



Joined: 12 Jan 2005
Posts: 3
Location: Seattle, WA, USA

View user's profile Send private message

Question about sprintf and float variables
PostPosted: Tue Jan 25, 2005 4:02 pm     Reply with quote

Here is a short test program:

// fuses, etc. not shown
void main(void)
{
signed int8 c=100;
float value=0;
char ascout[20];

value = (float)c;

sprintf(ascout,"%f",value);
}

Now, I'm using MPLAB 7.00 with CCS as the designated tool suite. The CCS version (IDE/PCB/PCM) is 3.215

When I run the above program in the MPLAB simulator and examine the variable contents afterwards, I see:

c = 0x64 (as expected)
value = "100.0000" which, read from memory, is 0x85, 0x48, 0x00, 0x00 (again as expected)
BUT:
ascout contains "99.999999" (!) [followed by terminating 0x00, as expected].

If I change the sprintf specification to "%3.0f", ascout is then "99". If, instead, I use "%3.1f", ascout then contains "99.9".

I tried adding this line: value = value + 0.000001; before the sprintf. Now I get the (predictable, at least) result of "100.000001". Grrr...

It appears to me as if the conversion from signed int8 to float is working okay, but sprintf is not. Of course, what I want is an ascii string of "100.000000"

What must I do to get this result?
TIA
Ttelmah
Guest







Re: Question about sprintf and float variables
PostPosted: Tue Jan 25, 2005 4:26 pm     Reply with quote

ccdart wrote:
Here is a short test program:

// fuses, etc. not shown
void main(void)
{
signed int8 c=100;
float value=0;
char ascout[20];

value = (float)c;

sprintf(ascout,"%f",value);
}

Now, I'm using MPLAB 7.00 with CCS as the designated tool suite. The CCS version (IDE/PCB/PCM) is 3.215

When I run the above program in the MPLAB simulator and examine the variable contents afterwards, I see:

c = 0x64 (as expected)
value = "100.0000" which, read from memory, is 0x85, 0x48, 0x00, 0x00 (again as expected)
BUT:
ascout contains "99.999999" (!) [followed by terminating 0x00, as expected].

If I change the sprintf specification to "%3.0f", ascout is then "99". If, instead, I use "%3.1f", ascout then contains "99.9".

I tried adding this line: value = value + 0.000001; before the sprintf. Now I get the (predictable, at least) result of "100.000001". Grrr...

It appears to me as if the conversion from signed int8 to float is working okay, but sprintf is not. Of course, what I want is an ascii string of "100.000000"

What must I do to get this result?
TIA

Unfortunately, you won't.
The problem is that when you use floating point values, you only have about 6.5 digits of 'real' accuracy (actually 1 in 8388608). With three digits in front of the DP, anything beyond the third digit after the DP, is really just 'noise', inherent in the conversion processes. You are asking too much of a 4 byte float. If you know that the values will never exceed a known value, then consider instead using a 32 bit integer, and just 'scaling' the result, by adding a suitable decimal point to the output string. This gives just over 9 digits of accuracy available.
If you limit yourself to 3 decimal places, and add 0.00049 to the value, you will get nice repeatable results.

Best Wishes
ccdart



Joined: 12 Jan 2005
Posts: 3
Location: Seattle, WA, USA

View user's profile Send private message

PostPosted: Tue Jan 25, 2005 4:36 pm     Reply with quote

Okay, that's pretty much what I was expecting to hear. I'll try your suggestion ASAP... if true, it should be good enough.
ccdart



Joined: 12 Jan 2005
Posts: 3
Location: Seattle, WA, USA

View user's profile Send private message

PostPosted: Tue Jan 25, 2005 4:46 pm     Reply with quote

Yup! That does it. I also tried four decimal places and adding .000049, and that seems to work as well. (Though I wonder if that's treading on thin ice...)

Thanks again for the solution!
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