字符串到浮点转换遇到问题

XC8的浮动变量的精度（？）编译器就是这样的。您不能只按需要添加小数位数，这不是Atof函数的问题。从HTTP://www-MyCHIP.COM/FUMMS/M96562.ASPXO中获取，如果您还没有使用32位浮点，那么这可能会给您更精确的一点。锡安。解决方案取决于你想用这些数字做什么，例如，你真的需要浮点…P.S.你解决了SIM808的电平移位问题吗？



      以下为原文

    The precision for float variables of your XC8 (?) compiler is like that. You cannot just add fractional digits as you want, and this is not a problem of the atof function.
 

taken from http://www.microchip.com/forums/m966542.aspx
 
So if you do not already use 32 bit floats, then that might give you a littlebit more precision.
But think about whether you really need more precision.  The solution depends on what you are trying to do with these numbers, e.g. do you really need floating point ...
 
P.S. Did you solve your level shifter problem for SIM808 ?
 

请注意，切换到32位浮点将导致更慢的代码。



      以下为原文

    And Note that switching to 32bit floats will result in bigger slower code.

如果浮点变量的大小和精度相似，则可以使用32位长，其中11.123456表示11123456。您可以获得6位数的精度值，达到4294（未签名）。可能有固定点数值类型，尽管可能不适用于PICS和XC8。我找到了这个20年的文档：http://www. Microchip .com /下载/Eng/NoppNo/0526E.PDFANHTTP://WW1.Microchip .com…/En/AppNodis/0617PDF.对于PIC24和PIC32有固定的例程：http://www. MicroChIP.COM/SWLabyWeb/Studio.ASPX？产品=固定%20点%20MaTys20LabyLaCube可能满足您的需求：http://www. McCHIP.COM/FUMMS/M801645.ASPXORT项目：http://885.n7.nabbul.com /浮点定点-TD89814HTMLhttp://www. PcList.COM/TeCfRe/MyCHIP/Mth/BasIC.HTM



      以下为原文

    If your floating point variables are similar in size and precision, you might be able to use 32 bit long, where 11.123456 would be represented by 11123456. You can get 6 digit fractional precision for values up to 4294 (unsigned).
 
There may be a fixed point numerical type, although perhaps not for PICs and XC8. I found this 20 year old document:
http://ww1.microchip.com/downloads/en/AppNotes/00526e.pdf
and
http://ww1.microchip.com/.../en/AppNotes/00617.pdf
 
There are fixed point routines for PIC24 and PIC32:
http://www.microchip.com/SWLibraryWeb/product.aspx?product=Fixed%20Point%20Math%20Library
 
Perhaps long double would meet your needs:
http://www.microchip.com/forums/m801645.aspx
 
Other items of interest:
http://microcontrollers.2385.n7.nabble.com/Floating-point-fixed-point-td89814.html
http://www.piclist.com/techref/microchip/math/basic.htm
 

如果浮点变量的大小和精度相似，则可以使用32位长，其中11.123456表示11123456。您可以获得6位数的精度值，达到4294（未签名）。可能有固定点数值类型，尽管可能不适用于PICS和XC8。我找到了这个20年的文档：http://www. Microchip .com /下载/Eng/NoppNo/0526E.PDFANHTTP://WW1.Microchip .com…/En/AppNodis/0617PDF.对于PIC24和PIC32有固定的例程：http://www. MicroChIP.COM/SWLabyWeb/Studio.ASPX？产品=固定%20点%20MaTys20LabyLaCube可能满足您的需求：http://www. McCHIP.COM/FUMMS/M801645.ASPXORT项目：http://885.n7.nabbul.com /浮点定点-TD89814HTMLhttp://www. PcList.COM/TeCfRe/MyCHIP/Mth/BasIC.HTM



      以下为原文

    If your floating point variables are similar in size and precision, you might be able to use 32 bit long, where 11.123456 would be represented by 11123456. You can get 6 digit fractional precision for values up to 4294 (unsigned).
 
There may be a fixed point numerical type, although perhaps not for PICs and XC8. I found this 20 year old document:
http://ww1.microchip.com/downloads/en/AppNotes/00526e.pdf
and
http://ww1.microchip.com/.../en/AppNotes/00617.pdf
 
There are fixed point routines for PIC24 and PIC32:
http://www.microchip.com/SWLibraryWeb/product.aspx?product=Fixed%20Point%20Math%20Library
 
Perhaps long double would meet your needs:
http://www.microchip.com/forums/m801645.aspx
 
Other items of interest:
http://microcontrollers.2385.n7.nabble.com/Floating-point-fixed-point-td89814.html
http://www.piclist.com/techref/microchip/math/basic.htm
 

谢谢大家，这个问题是通过设置32位浮点配置来解决的。我非常感谢你们的帮助。谢谢你的帮助。我用了一个上拉电路，它起作用了。现在，PIC18F480能够读取GSM的响应，并在LCDSmile中显示它：



      以下为原文

    Thank you all.
The problem is solved by setting 32 bit configuration for floating point.
I really appreciate your help.
And about "level shifter problem for SIM808"
Yeah it is solved. Thank you for your help. I used a pull-up circuit for it.It worked. Now the PIC18f4580 is able to read the response from the GSM and is displaying it in the LCDSmile: