Subject:
|
Math.log probs. (Was: Proximity sensor)
|
Newsgroups:
|
lugnet.robotics.rcx.java
|
Date:
|
Mon, 1 Oct 2001 20:21:21 GMT
|
Viewed:
|
2883 times
|
| |
|
|
Hi Paul,
> > Btw. could you use my StringBuffer extension?
> I tried them but it looks as though there are problems with the Math
> methods. In particular Math.log(<I can't remember waht>) returns zero when
> it shouldn't. If you like you could take a look at them. They are all
> written in Java. If you do you should run a test under the Linux emulator.
> The regression directory has exmaples. Look for the ones that use
> Test.assert().
Ok, I couldn't find any examples for which the Math.log
returned 0, either with emulator nor the RCX (I didn't look
too hard :o). Maybe you could supply with some examples?
The basic routine should be ok!
I encountered situations in which the convergence criteria
couldn't be met:
ACCURACY = 0.0000001f;
but changing it to:
ACCURACY = 1e-6f;
did the job. Try that...
That made me wonder; is the double precision really floating
point precision?
Regards,
Martin
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
