Question about the hp35s

[Arne Halvorsen (Norway)]

Sorry about asking this question, but my excuses are: HP has for some reason choose not to put out pdf of manual, Norway not getting any hp-35s soon, I am curious!:

One feature (aside from the BIG ENTER KEY) that makes the machine interesting for my purposes are the 2D and 3D vector types. Working with computer graphics I could make use of this feature I hope (do a lot of checking result from vector calculations while debugging).

Question: What are the operations for addressing, unpacking (say into x,y,z) vector's components?

Kind of asking I guess: 'How practical are these types': They seems handy for input/presentation, but how are them to program linear algebra with.?.

Grateful for any info to one who is counting on his fingers to just order the thing from abroad :-)

Edited: 23 Aug 2007, 8:28 a.m.

[Valentin Albillo]

Hi, Arne:

Arne asked:

"Question: What are the operations for addressing, unpacking (say into x,y,z) vector's components?"
None. There's no built-in operation to disassemble a vector into its 1,2, or 3 components nor to extract a component from a vector. You need some programming to do it and it's quite an slow and cumbersome procedure.

'How practical are these types': They seems handy for input/presentation, but how are them to program linear algebra with.?.

I would need to work with them a lot more to be able to give a qualified answer but judging from what I've seen up to now, they seem to be greatly unpractical for most uses, including linear algebra. There are very few instructions that work with them, mostly basic arithmetic, and accessing their components is quite tricky and involved, as mentioned above.


Also, it's been mentioned that they might be somewhat buggy.

Best regards from V.

[Arne Halvorsen (Norway)]

Thanks a lot!

Had a feeling this could be the case...

What it sounds like is that a program that would work on a problem would access vector components stored as individual values in normal registers.

At least the results can be presented in vector form since it is easy to make a vector(s) (training stuff shows that) after computations.

[Valentin Albillo]

Hi again, Arne:

Arne posted:

"What it sounds like is that a program that would work on a problem would access vector components stored as individual values in normal registers."
For maximum speed, that's the way to go indeed until someone cames up with a SYSEVAL call or some hidden function to extract the individual components.

"Maximum speed" here means up to 100 times faster than resorting to some packing/unpacking program to get at the components.

Best regards from V.

[Arne Halvorsen (Norway)]

Well, the good thing about this is that I got pretty much enough information to design and simulate my programs before get the machine...

[bill platt]

I think the tried and true for vectors is the 48 series and its descendants--the 50G.

[Chris Dean]

Arne

From my experience if you are working in graphics then get a graphical calculator i.e. HP50G. It is so much easier to develop algorithms and see/visualise the results.

[Arne Halvorsen (Norway)]

Yea, well... I must say it has crossed my mind after been thinking about getting hp35s...

I had orginal NOT been planning to take up my youthfull passion for programming calculators (HP41CV, rip wherever you are now...), what
I been missing (but I borrow a hp15c a lot) is a simple rpn to add, sqrt to check program output.

A thing I am worried about by getting something as powerfull as the 50g is that I may be absorbed by it..., propably enjoy every second of it though...

Maybe....

Edited: 23 Aug 2007, 12:18 p.m.