Re: Change made to atan2()

Xavier Bestel wrote:
> On Fri, 2007-09-28 at 08:37 +0200, Xavier Bestel wrote:
>> On Wed, 2007-09-26 at 09:04 -0500, David Miller wrote:
>>> Hi, Ken pointed out to me that the atan2() function in the LRM differs from the 
>>> C (and other programming languages, matlab etc.) definition.
>>>
>>> currently in the LRM (chap 4) we define atan2()
>>> atan2(x,y) - arc tangent x / y
>>>
>>> This is also the same as the definition in the Verilog 2005 standard (which I 
>>> guess we cut and pasted).
>>>
>>> This function should be defined as:
>>> atan2(y,x) - arc tangent y / x
>>>
>>> Unless anyone strongly objects, I will go ahead and correct this definition.
>> You just repeated the same error :)
>>
>> The function should be defined as:
>> atan2(x,y) - arc tangent y / x
> 
> Sorry, I didn't realise it was just a cosmetic error, I thought it was a
> wrong formula. I'm going to bury myself somewhere now.
> 
> 	Xav

No worries.

There was a whole discussion on the SV reflector, about
how everyone expects "y" to be the "opposite" side of
the triangle and thus the cosmetic change makes it
easier to understand.

(Am I the only one who sees a minus sign - instead of equals =
in the definitions?)

-Geoffrey

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.