Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #38400 > unrolled thread

Has Android different floats?

Started byj4n bur53 <janburse@fastmail.fm>
First post2018-06-24 20:41 +0200
Last post2018-06-26 10:29 -0700
Articles 8 — 4 participants

Back to article view | Back to comp.lang.java.programmer

Contents

  Has Android different floats? j4n bur53 <janburse@fastmail.fm> - 2018-06-24 20:41 +0200
    Re: Has Android different floats? Arne Vajhøj <arne@vajhoej.dk> - 2018-06-24 15:21 -0400
      Re: Has Android different floats? bursejan@gmail.com - 2018-06-24 13:31 -0700
        Re: Has Android different floats? bursejan@gmail.com - 2018-06-24 13:35 -0700
          Re: Has Android different floats? bursejan@gmail.com - 2018-06-24 13:44 -0700
          Re: Has Android different floats? Eric Sosman <esosman@comcast-dot-net.invalid> - 2018-06-24 17:28 -0400
            Re: Has Android different floats? bursejan@gmail.com - 2018-06-24 16:08 -0700
              Re: Has Android different floats? bursejan@gmail.com - 2018-06-26 10:29 -0700

#38400 — Has Android different floats?

Fromj4n bur53 <janburse@fastmail.fm>
Date2018-06-24 20:41 +0200
SubjectHas Android different floats?
Message-ID<pgool7$bru$1@news.albasani.net>
Dear All,

Running a simple test.
** is Math.pow(),
^ is BigInteger.pow() and then BigInteger.doubleValue().

On Java:

   ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
   15-0.0
   16-0.0
   17-0.0
   18-0.0
   19-0.0
   20-0.0
   21-0.0
   22-0.0
   23-1.6777216E7
   24-0.0
   25-0.0
   26-0.0
   27-0.0
   28-0.0
   29-0.0
   30-0.0

On Adroid:

   ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
   15-0.0
   16-0.0
   17-0.0
   18-0.0
   19-0.0
   20-0.0
   21-0.0
   22-0.0
   23-0.0
   24-0.0
   25-0.0
   26-0.0
   27-0.0
   28-0.0
   29-1.7592186044416E13
   30-0.0

Has Android different floats?
Tested jdk1.8.0_152 against Android 8.0.0

Best Regards

[toc] | [next] | [standalone]

#38401

FromArne Vajhøj <arne@vajhoej.dk>
Date2018-06-24 15:21 -0400
Message-ID<pgor09$19ak$1@gioia.aioe.org>
In reply to#38400
On 6/24/2018 2:41 PM, j4n bur53 wrote:
> Running a simple test.
> ** is Math.pow(),
> ^ is BigInteger.pow() and then BigInteger.doubleValue().
> 
> On Java:
> 
>    ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
>    15-0.0

>    23-1.6777216E7

>    29-0.0
>    30-0.0
> 
> On Adroid:
> 
>    ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
>    15-0.0

>    23-0.0

>    29-1.7592186044416E13
>    30-0.0
> 
> Has Android different floats?
> Tested jdk1.8.0_152 against Android 8.0.0

ARM supposedly use standard IEEE 754 FP.

But I wonder if usage of strictfp would change your results.

Arne

[toc] | [prev] | [next] | [standalone]

#38402

Frombursejan@gmail.com
Date2018-06-24 13:31 -0700
Message-ID<b4f7eefd-bde4-4b77-9dbf-b417091061ed@googlegroups.com>
In reply to#38401
You cannot use the strictfp keyword in front
of Math.pow(). Thats only for the basic operations.

Math.pow() itself is implemented in JDK 1.8 as follows:

    public static double pow(double a, double b) {
        return StrictMath.pow(a, b); // default impl. delegates to StrictMath
    }

Am Sonntag, 24. Juni 2018 21:21:57 UTC+2 schrieb Arne Vajhøj:
> On 6/24/2018 2:41 PM, j4n bur53 wrote:
> > Running a simple test.
> > ** is Math.pow(),
> > ^ is BigInteger.pow() and then BigInteger.doubleValue().
> > 
> > On Java:
> > 
> >    ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
> >    15-0.0
> 
> >    23-1.6777216E7
> 
> >    29-0.0
> >    30-0.0
> > 
> > On Adroid:
> > 
> >    ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
> >    15-0.0
> 
> >    23-0.0
> 
> >    29-1.7592186044416E13
> >    30-0.0
> > 
> > Has Android different floats?
> > Tested jdk1.8.0_152 against Android 8.0.0
> 
> ARM supposedly use standard IEEE 754 FP.
> 
> But I wonder if usage of strictfp would change your results.
> 
> Arne

[toc] | [prev] | [next] | [standalone]

#38403

Frombursejan@gmail.com
Date2018-06-24 13:35 -0700
Message-ID<90082b6a-1547-4825-83f5-34f0762daad4@googlegroups.com>
In reply to#38402
The test code would be:

        for (int i=22; i<25; i++) {
           System.out.println("Math.pow(10,"+i+")="+Math.pow(10,i));
        }

The result (without strictfp modifier to main()) is:
   
   Math.pow(10,22)=1.0E22
   Math.pow(10,23)=1.0000000000000001E23
   Math.pow(10,24)=1.0E24

The result (with strictfp modifier to main()) is:

   Math.pow(10,22)=1.0E22
   Math.pow(10,23)=1.0000000000000001E23
   Math.pow(10,24)=1.0E24

Because invoking Math.pow() is not a basic operator +, *, /, etc..

Am Sonntag, 24. Juni 2018 22:31:44 UTC+2 schrieb burs...@gmail.com:
> You cannot use the strictfp keyword in front
> of Math.pow(). Thats only for the basic operations.
> 
> Math.pow() itself is implemented in JDK 1.8 as follows:
> 
>     public static double pow(double a, double b) {
>         return StrictMath.pow(a, b); // default impl. delegates to StrictMath
>     }
> 
> Am Sonntag, 24. Juni 2018 21:21:57 UTC+2 schrieb Arne Vajhøj:
> > On 6/24/2018 2:41 PM, j4n bur53 wrote:
> > > Running a simple test.
> > > ** is Math.pow(),
> > > ^ is BigInteger.pow() and then BigInteger.doubleValue().
> > > 
> > > On Java:
> > > 
> > >    ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
> > >    15-0.0
> > 
> > >    23-1.6777216E7
> > 
> > >    29-0.0
> > >    30-0.0
> > > 
> > > On Adroid:
> > > 
> > >    ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
> > >    15-0.0
> > 
> > >    23-0.0
> > 
> > >    29-1.7592186044416E13
> > >    30-0.0
> > > 
> > > Has Android different floats?
> > > Tested jdk1.8.0_152 against Android 8.0.0
> > 
> > ARM supposedly use standard IEEE 754 FP.
> > 
> > But I wonder if usage of strictfp would change your results.
> > 
> > Arne

[toc] | [prev] | [next] | [standalone]

#38404

Frombursejan@gmail.com
Date2018-06-24 13:44 -0700
Message-ID<c9789a02-8696-4353-940a-bae36b9ed75f@googlegroups.com>
In reply to#38403
You can try the same Java code on a Android
device. The result will be (Android 8.0.0):

   Math.pow(10,22)=1.0E22
   Math.pow(10,23)=9.999999999999999E22
   Math.pow(10,24)=1.0E24 

I used AIDE-IDE to compile and run the Java
directly on the device.

But the above is not the same test case as
in my initial post. My initial post also

tests BigInteger.pow() and BigInteger.doubleValue().

Am Sonntag, 24. Juni 2018 22:35:50 UTC+2 schrieb burs...@gmail.com:
> The test code would be:
> 
>         for (int i=22; i<25; i++) {
>            System.out.println("Math.pow(10,"+i+")="+Math.pow(10,i));
>         }
> 
> The result (without strictfp modifier to main()) is:
>    
>    Math.pow(10,22)=1.0E22
>    Math.pow(10,23)=1.0000000000000001E23
>    Math.pow(10,24)=1.0E24
> 
> The result (with strictfp modifier to main()) is:
> 
>    Math.pow(10,22)=1.0E22
>    Math.pow(10,23)=1.0000000000000001E23
>    Math.pow(10,24)=1.0E24
> 
> Because invoking Math.pow() is not a basic operator +, *, /, etc..
> 
> Am Sonntag, 24. Juni 2018 22:31:44 UTC+2 schrieb burs...@gmail.com:
> > You cannot use the strictfp keyword in front
> > of Math.pow(). Thats only for the basic operations.
> > 
> > Math.pow() itself is implemented in JDK 1.8 as follows:
> > 
> >     public static double pow(double a, double b) {
> >         return StrictMath.pow(a, b); // default impl. delegates to StrictMath
> >     }
> > 
> > Am Sonntag, 24. Juni 2018 21:21:57 UTC+2 schrieb Arne Vajhøj:
> > > On 6/24/2018 2:41 PM, j4n bur53 wrote:
> > > > Running a simple test.
> > > > ** is Math.pow(),
> > > > ^ is BigInteger.pow() and then BigInteger.doubleValue().
> > > > 
> > > > On Java:
> > > > 
> > > >    ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
> > > >    15-0.0
> > > 
> > > >    23-1.6777216E7
> > > 
> > > >    29-0.0
> > > >    30-0.0
> > > > 
> > > > On Adroid:
> > > > 
> > > >    ?- between(15,30,N), X is 10**N-10^N, write(N-X), nl, fail; true.
> > > >    15-0.0
> > > 
> > > >    23-0.0
> > > 
> > > >    29-1.7592186044416E13
> > > >    30-0.0
> > > > 
> > > > Has Android different floats?
> > > > Tested jdk1.8.0_152 against Android 8.0.0
> > > 
> > > ARM supposedly use standard IEEE 754 FP.
> > > 
> > > But I wonder if usage of strictfp would change your results.
> > > 
> > > Arne

[toc] | [prev] | [next] | [standalone]

#38405

FromEric Sosman <esosman@comcast-dot-net.invalid>
Date2018-06-24 17:28 -0400
Message-ID<pgp2es$ls9$1@dont-email.me>
In reply to#38403
On 6/24/2018 4:35 PM, bursejan@gmail.com wrote:
> The test code would be:
> 
>          for (int i=22; i<25; i++) {
>             System.out.println("Math.pow(10,"+i+")="+Math.pow(10,i));
>          }
> 
> The result (without strictfp modifier to main()) is:
>     
>     Math.pow(10,22)=1.0E22
>     Math.pow(10,23)=1.0000000000000001E23
>     Math.pow(10,24)=1.0E24
> 
> The result (with strictfp modifier to main()) is:
> 
>     Math.pow(10,22)=1.0E22
>     Math.pow(10,23)=1.0000000000000001E23
>     Math.pow(10,24)=1.0E24
> 
> Because invoking Math.pow() is not a basic operator +, *, /, etc..

     Have you tried using StrictMath.pow() instead of Math.pow()?
It is certainly possible that the Math implementations are different
on different platforms -- at least, the Javadoc seems to allow such
freedom:

	"[...] Code generators are encouraged to use platform-specific
	native libraries or microprocessor instructions, where
	available, to provide higher-performance implementations of Math
	methods. [...]"

It's my understanding, though, that StrictMath implementations on all
platforms are supposed to agree exactly.

-- 
esosman@comcast-dot-net.invalid
Nine hundred forty-one days to go.

[toc] | [prev] | [next] | [standalone]

#38406

Frombursejan@gmail.com
Date2018-06-24 16:08 -0700
Message-ID<7bf8cf29-d6fc-4e18-b31a-81d9a8d66ba6@googlegroups.com>
In reply to#38405
Yes, this seems to be the explanation, on the
windows platform, Math.pow() seems to get replaced
by something else than StrictMath.pow().

With StrictMath.pow() (JDK 1.8, Windows):

   Math.pow(10,22)=1.0E22
   Math.pow(10,23)=9.999999999999999E22
   Math.pow(10,24)=1.0E24

With Math.pow() (JDK 1.8, Windows):

   Math.pow(10,22)=1.0E22
   Math.pow(10,23)=1.0000000000000001E23
   Math.pow(10,24)=1.0E24

What is Math.pow() doing on Windows, some Wintel nonsense?

Am Sonntag, 24. Juni 2018 23:29:11 UTC+2 schrieb Eric Sosman:
> On 6/24/2018 4:35 PM, bursejan@gmail.com wrote:
> > The test code would be:
> > 
> >          for (int i=22; i<25; i++) {
> >             System.out.println("Math.pow(10,"+i+")="+Math.pow(10,i));
> >          }
> > 
> > The result (without strictfp modifier to main()) is:
> >     
> >     Math.pow(10,22)=1.0E22
> >     Math.pow(10,23)=1.0000000000000001E23
> >     Math.pow(10,24)=1.0E24
> > 
> > The result (with strictfp modifier to main()) is:
> > 
> >     Math.pow(10,22)=1.0E22
> >     Math.pow(10,23)=1.0000000000000001E23
> >     Math.pow(10,24)=1.0E24
> > 
> > Because invoking Math.pow() is not a basic operator +, *, /, etc..
> 
>      Have you tried using StrictMath.pow() instead of Math.pow()?
> It is certainly possible that the Math implementations are different
> on different platforms -- at least, the Javadoc seems to allow such
> freedom:
> 
> 	"[...] Code generators are encouraged to use platform-specific
> 	native libraries or microprocessor instructions, where
> 	available, to provide higher-performance implementations of Math
> 	methods. [...]"
> 
> It's my understanding, though, that StrictMath implementations on all
> platforms are supposed to agree exactly.
> 
> -- 
> esosman@comcast-dot-net.invalid
> Nine hundred forty-one days to go.

[toc] | [prev] | [next] | [standalone]

#38418

Frombursejan@gmail.com
Date2018-06-26 10:29 -0700
Message-ID<eecccf4c-862b-4a1b-9777-6c39fa357538@googlegroups.com>
In reply to#38406
I didn't send or was involved in the last
7 emails. Maybe some robot was going astray.

LoL

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.java.programmer

csiph-web