The Museum of HP Calculators

HP Forum Archive 21

[ Return to Index | Top of Index ]

HP-67 internal function timings
Message #1 Posted by Dieter on 10 Feb 2013, 10:37 a.m.

I admit I am amazed at the speed of the 34s Matrix functions - even a 10x10 random matrix is inverted in fractions of a second (approx. 0,7Śn3 ms on my hardware). That's roughly as fast as some transcendental functions on earlier HPs like the 67/97.

I remember a table with the execution times of most 67/97 functions, given in milliseconds, somewhere in a HP related magazine or journal. I am sure I must have stored it somewhere on some harddrive, but I cannot seem to find it. Maybe someone can help and tell me where to find this info?!

Dieter

      
Re: HP-67 internal function timings
Message #2 Posted by Jake Schwartz on 10 Feb 2013, 4:57 p.m.,
in response to message #1 by Dieter

Quote:
I remember a table with the execution times of most 67/97 functions, given in milliseconds, somewhere in a HP related magazine or journal. I am sure I must have stored it somewhere on some harddrive, but I cannot seem to find it. Maybe someone can help and tell me where to find this info?!

Hi,

One of these appeared in PPC's "Better Programming On The HP67/97" publication, which was sent to members roughly in June of 1978, around the time of Volume 5, Number 5 of the PPC Journal. The table appears in pages 27-29.

Jake

            
Re: HP-67 internal function timings
Message #3 Posted by Dieter on 11 Feb 2013, 2:50 p.m.,
in response to message #2 by Jake Schwartz

Ah, thank you - now I know what to look for. I think I have a PDF of this. It must be somewhere deep down in some sub-sub-sub-folder... :-)

Dieter

      
Re: HP-67 internal function timings
Message #4 Posted by Paul Dale on 10 Feb 2013, 5:18 p.m.,
in response to message #1 by Dieter

Quote:
I admit I am amazed at the speed of the 34s Matrix functions - even a 10x10 random matrix is inverted in fractions of a second (approx. 0,7Śn3 ms on my hardware). That's roughly as fast as some transcendental functions on earlier HPs like the 67/97.

The surprising thing is I'm not using a particularly efficient algorithm for matrix inversion. A Doolittle LU decomposition with partial pivoting followed by solving lots of equations.

- Pauli

[ Return to Index | Top of Index ]

Go back to the main exhibit hall