FastMathTestPerformance results misleading

2011-03-31 Thread Bill Rossi
This test shows that Math.sin() is faster than FastMath.sin(), but this is misleading because the test operates over the domain 0 < x <1, whereas legal arguments to sin() are all finite numbers. In particular, its when |x| is large that performance of the Math trig functions really drops off

Re: [Math] FastMath Performance

2011-01-26 Thread Bill Rossi
On Wed, 26 Jan 2011, Gilles Sadowski wrote: On Wed, Jan 26, 2011 at 12:12:47PM +, sebb wrote: On 26 January 2011 11:05, Gilles Sadowski wrote: Hi. I'd wish to make it clear that I referred to this statement: The Javadoc for FastMath says that it is a replacement for StrictMath, whi

Re: [Math] FastMath Performance

2011-01-25 Thread Bill Rossi
On Tue, 25 Jan 2011, luc.maison...@free.fr wrote: Hi Gilles, - "Gilles Sadowski" a écrit : Hello. [...] Note: this appears to be at least as quick as StrictMath on Sun Java 1.6 in a crude test The performance comparison for "FastMath" should be against "Math", as "StrictMath is

Re: [Math] BigDecimal to power

2011-01-21 Thread Bill Rossi
You can compare results to those produced by the org.apache.commons.math.dfp package. On Fri, 21 Jan 2011, sebb wrote: On 21 January 2011 13:19, Gilles Sadowski wrote: Hello. When proposing code, I think that you might get better attention by posting to the "dev" ML. I changed the licen

Re: [MATH] generating -0.0 rather than 0.0 when incrementing a negative number

2011-01-20 Thread Bill Rossi
I think having an explicit special case might be the best way to solve this, in terms of performance. The test can be put inside the existing if block so you only pay the price of the comparison when rounding up. On Wed, 19 Jan 2011, sebb wrote: On 19 January 2011 18:28, Ted Dunning wr

Re: [math] unit tests for FastMath

2010-08-28 Thread Bill Rossi
On Sat, 28 Aug 2010 18:25:23 -0400, wrote: > Luc Maisonobe wrote: > > Hi all, > > > > I have started integrating the FastMath class from MATH-375. I have not > > committed anything for now as I am first fixing numerous checkstyle > > errors (more than 300) and wanted to have one clean commit to

Re: [math] software grant for MATH-375 ?

2010-08-19 Thread Bill Rossi
Ok. I will get that done today. On Thu, 19 Aug 2010, Luc Maisonobe wrote: Hi Bill, About two months ago, you proposed a patch to improve elementary functions accuracy and speed to the apache commons-math component at our Jira issue tracker . A

Re: [math] elementary functions.

2010-06-09 Thread Bill Rossi
No it had not occured to me. I can look into that. On Wed, 9 Jun 2010, James Carman wrote: On Wed, Jun 9, 2010 at 7:55 AM, Bill Rossi wrote: 2.) I also think there there is the perception that the existing methods are good enough, so its probably not a high priority. Fair enough. Have

Re: [math] elementary functions.

2010-06-09 Thread Bill Rossi
ty. On Wed, 9 Jun 2010, Phil Steitz wrote: Luc Maisonobe wrote: Le 08/06/2010 23:07, Bill Rossi a écrit : Luc, I'm OK with providing an ICLA or SGA or both. A bit of history on the project to help you decide what is appropriate: While working for my employer some last year we noticed som

Re: [math] elementary functions.

2010-06-08 Thread Bill Rossi
nobe wrote: Le 08/06/2010 23:07, Bill Rossi a écrit : Luc, I'm OK with providing an ICLA or SGA or both. A bit of history on the project to help you decide what is appropriate: While working for my employer some last year we noticed some of these functions were slow. I quickly wrote a c

Re: [math] elementary functions.

2010-06-08 Thread Bill Rossi
Luc, I'm OK with providing an ICLA or SGA or both. A bit of history on the project to help you decide what is appropriate: While working for my employer some last year we noticed some of these functions were slow. I quickly wrote a couple of faster replacements. Seperately I decided to t

Re: [math] elementary functions.

2010-06-08 Thread Bill Rossi
#x27;d be great to see it make it into the language itself (in about 10 years as these things work). On Tue, Jun 8, 2010 at 11:39 AM, Ted Dunning wrote: Bill, This looks like nice work. Have you considered contributing the code back to Java itself? On Tue, Jun 8, 2010 at 8:29 AM, Bill Rossi

Re: [math] elementary functions.

2010-06-08 Thread Bill Rossi
esults and compare them to an aribitrary precision math library. On Tue, 8 Jun 2010, James Carman wrote: Have you tried looking at StrictMath? On Tue, Jun 8, 2010 at 10:44 AM, Ted Dunning wrote: Bill, Which functions do you have? Anything more than the standard sin, cos, exp and log? On

[math] elementary functions.

2010-06-08 Thread Bill Rossi
Greetings, I have developed over the past year a set of elementary functions similar to those in java.lang.Math, but with the following characteristics: * Higher performance. * Better accuracy. Results are accurate to slightly more that +/- 0.5 ULP. * Pure Java. The standard Math class is i