T800 floating pt. results

Thread View: comp.sys.transputer

3 messages

[T] Back to Threads [V] Tree View [G] Group Articles

3 total messages Started by Andy Raine Thu, 25 Apr 1991 12:05

#1616

Author: Andy Raine
Date: Thu, 25 Apr 1991 12:05

32 lines
1333 bytes

Hello all

I just tried the test program on a T800 in our Meiko in-sun computing surface, 
as well as a SUN 4/110 and a microVAX 3100. the results I got were:

MEIKO C (CStools/SUN) compiled for T800:

The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
The Calculated Values:> free_l[1]=0.000031 and free_l[2]=-0.000118
Done...

MEIKO C (CStools/SUN) compiled for T414:

The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
The Calculated Values:> free_l[1]=0.000031 and free_l[2]=-0.000118
Done...

SUN C (SunOS 4.0.3):

The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
The Calculated Values:> free_l[1]=-0.000002 and free_l[2]=0.000007
Done...

VAX C (VMS 5.4-1):

The Expected Results:> free_l[1]=0.000000 and free_l[2]=0.000000
The Calculated Values:> free_l[1]=0.000000 and free_l[2]=0.000000
Done...

Interesting huh?

Andy

Re: T800 floating pt. results

#1622

Author: Reinhard Kirchner
Date: Fri, 26 Apr 1991 10:23

28 lines
1501 bytes

From article <7983.9104251204@prg.oxford.ac.uk>, by ARCR1@biology.cambridge.ac.uk (Andy Raine):
> Hello all
> 
> 
> The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
> The Calculated Values:> free_l[1]=0.000031 and free_l[2]=-0.000118
> 
> The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
> The Calculated Values:> free_l[1]=0.000031 and free_l[2]=-0.000118
> 
> The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
> The Calculated Values:> free_l[1]=-0.000002 and free_l[2]=0.000007
> 
> The Expected Results:> free_l[1]=0.000000 and free_l[2]=0.000000
> The Calculated Values:> free_l[1]=0.000000 and free_l[2]=0.000000

This is why mathmaticians at Karlsruhe University think about optimal
arithmetic since about 20 years and why they have developed products
which help to produce results you can trust.

For details send mail to

Michael Neaga         ae18@dkauni2.bitnet

Reinhard Kirchner
Univ. Kaiserslautern, Germany

( formerly working on these things and still having connections )

Re: T800 floating pt. results

#1620

Author: Tony
Date: Fri, 26 Apr 1991 10:31

52 lines
2393 bytes

In article <7983.9104251204@prg.oxford.ac.uk> ARCR1@biology.cambridge.ac.uk (Andy Raine) writes:

Andy> I just tried the test program on a T800 in our Meiko
Andy> in-sun computing surface, as well as a SUN 4/110 and a
Andy> microVAX 3100. the results I got were:

Andy> blah, blah, blah, his results.

I just ran the same program here on various things (sparc,
meiko sparc-hosted, mips).  Here are the results of the Kent
jury...

(t800) mcc txxx.c ; mrun m.out

The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
The Calculated Values:> free_l[1]=0.000031 and free_l[2]=-0.000118
Done...

---------------------------------------------------------------------------------
(sparc) cc txxx.c ; a.out

the Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
The Calculated Values:> free_l[1]=-0.000002 and free_l[2]=0.000007
Done...

---------------------------------------------------------------------------------
(t414) mcc -t4 txxx.c ; mrun m.out

[node`m] register dump at iptr 0x8000ad88 (0x4be0) wptr 0x8001b625 (0xec5) (program error)

---------------------------------------------------------------------------------
(sparc) gcc txxx.c ; a.out

The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
The Calculated Values:> free_l[1]=0.000000 and free_l[2]=-0.000000
Done...

---------------------------------------------------------------------------------
(mips) cc txxx.c ; a.out

The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
The Calculated Values:> free_l[1]=-0.000002 and free_l[2]=0.000007
Done...

---------------------------------------------------------------------------------
(mips) gcc txxx.c ; a.out

The Expected Results:> free_l[1]=-0.000002 and free_l[2]=0.000007
The Calculated Values:> free_l[1]=0.000000 and free_l[2]=0.000000
Done...

---------------------------------------------------------------------------------

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads