Plus42 Equations, Preview Release

[Thomas Okken]

New update: UNITS catalog in the equation editor. Note that it only inserts the unit text; you'll still have to enter the _ and " manually.

[roberto_abraham]

(12-17-2021 04:49 PM)Thomas Okken Wrote:  You could write

C:300000_"km/s"

and then refer to it as C(), that is, a function call without passing arguments.

The only alternative, if you want to avoid the parentheses, would be to create a variable on the RPN side:

300000_km/s STO "C"

You could refer to that simply as C in your equations. It would be a variable, though, not a constant...

Thank you. Both approaches would work fine for my purposes.

Thanks again!

[Ren]

(12-16-2021 09:25 PM)Thomas Okken Wrote:  

(12-16-2021 09:00 PM)Ren Wrote:  What is the difference between Plus42 and Free42?

Everything you read in this thread!

Plus42 is an extended version of Free42, currently under development. It will feature all the capabilities of Free42, plus equations (modeled on those found in the HP-17B, 19B, 27S, etc., this is finished), attached units (modeled on those found in the HP-48/49/50 series; almost done), and directories for hierarchical storage of programs and variables, and support for larger display sizes, coming soon.

This will be a separate product, GPLv2 and free for Windows, MacOS, and Linux, and priced at something like €10 for Android and iOS. Price not yet decided, release probably in January or February.

Thanks for the quick answer, going through (now 8 pages) this thread to find the answer seemed pretty daunting to me!

[Thomas Okken]

New update:

1. UNITS catalog in equation editor is more helpful now; takes care of inserting _" and normalizing unit products etc.
2. UNITS catalog now working in PRGM mode; works essentially the same as in RUN mode.

[Thomas Okken]

New update:

1. Unit support in SOLVE
2. Unit support in INTEG

UPDATE:

Fixed two memory corruption bugs.

[Marco Polo]

(12-18-2021 05:04 PM)Thomas Okken Wrote:  New update:

1. Unit support in SOLVE
2. Unit support in INTEG

UPDATE:

Fixed two memory corruption bugs.

Outstanding work, indeed!
Is it possible to define custom UOM like in 48?

[Thomas Okken]

(12-18-2021 06:47 PM)Marco Polo Wrote:  Is it possible to define custom UOM like in 48?

No, not yet.

(Scampers off to read the 48G manual...)

[Thomas Okken]

Ah, I see. So a user-defined unit is simply a number with an attached unit, stored in a variable in a certain location? And the name of that variable then becomes the new unit?

That sounds straightforward enough. I'll add that functionality after I get directories working, it sounds like you'll want a separate directory to put those units and keep them from cluttering up your work areas.

[Marco Polo]

(12-19-2021 07:00 AM)Thomas Okken Wrote:  Ah, I see. So a user-defined unit is simply a number with an attached unit, stored in a variable in a certain location? And the name of that variable then becomes the new unit?

That sounds straightforward enough. I'll add that functionality after I get directories working, it sounds like you'll want a separate directory to put those units and keep them from cluttering up your work areas.

Yes.
Keep in mind in 48 the variables in upper directory are available to lower ones.
So the variables in HOME are available in the whole directory tree. Hence the user defidrn UOMs should be in Home Directory

[Thomas Okken]

My plan for name resolution is slightly more elaborate: search the current directory, and ancestors all the way to the root directory, but in addition, also a PATH directory where you can put symlinks to other directories, which will be searched from anywhere. So you don't have to clutter up the root directory with all the things that should be globally available.

[roberto_abraham]

(12-19-2021 11:11 AM)Thomas Okken Wrote:  My plan for name resolution is slightly more elaborate: search the current directory, and ancestors all the way to the root directory, but in addition, also a PATH directory where you can put symlinks to other directories, which will be searched from anywhere. So you don't have to clutter up the root directory with all the things that should be globally available.

That sounds very useful! The feature set of Plus42 sounds absolutely wonderful.

In case it helps to have an edge case for a user-defined unit to think about, a unit that I use all the time but which I think would be handy (but challenging) to incorporate as a user-defined unit is the Rayleigh (R). This is a unit of photon radiance defined as:

1 R = 1/(4 pi) 10^10 photon sec^−1 m^−2 Sr^−1

I bet this turns out to be a tricky edge case, since it involves two somewhat special situations. While 'photon' is part of the definition, it's just a dimensionless pure number that doesn't reduce to an SI base unit and instead should get carried along in calculations but stay symbolic. On the other hand, the solid angle is dimensionless but transforms to other dimensionless solid angle measures that have different exponents (e.g. arcsec^-2).

[Marco Polo]

(12-19-2021 11:11 AM)Thomas Okken Wrote:  My plan for name resolution is slightly more elaborate: search the current directory, and ancestors all the way to the root directory, but in addition, also a PATH directory where you can put symlinks to other directories, which will be searched from anywhere. So you don't have to clutter up the root directory with all the things that should be globally available.

It seems to me a very smart idea!

I think Plus42 is becoming something very near my "killer Calc" (the ideal would be a DM48, based on 48gx but without graphic)

[Thomas Okken]

(12-19-2021 12:56 PM)roberto_abraham Wrote:  1 R = 1/(4 pi) 10^10 photon sec^−1 m^−2 Sr^−1

I bet this turns out to be a tricky edge case, since it involves two somewhat special situations. While 'photon' is part of the definition, it's just a dimensionless pure number that doesn't reduce to an SI base unit and instead should get carried along in calculations but stay symbolic. On the other hand, the solid angle is dimensionless but transforms to other dimensionless solid angle measures that have different exponents (e.g. arcsec^-2).

It sounds like the 'photon' part could be dealt with by making it possible to define new base units, not just derived units. That's simple enough; even in the existing code, the only special treatment that base units get is that UBASE and friends leave them alone.

I don't know about the solid angle, though. The 48G has only the Steradian, which is a base unit and doesn't convert to anything else: 1_sr 1_r^2 CONVERT => Inconsistent Units.

[Vincent Weber]

(12-19-2021 01:36 PM)Marco Polo Wrote:  

(12-19-2021 11:11 AM)Thomas Okken Wrote:  My plan for name resolution is slightly more elaborate: search the current directory, and ancestors all the way to the root directory, but in addition, also a PATH directory where you can put symlinks to other directories, which will be searched from anywhere. So you don't have to clutter up the root directory with all the things that should be globally available.

It seems to me a very smart idea!

I think Plus42 is becoming something very near my "killer Calc" (the ideal would be a DM48, based on 48gx but without graphic)

Plus42 already is my killer calc, I always wanted a pocket programmable (not a monster like the 48) merging the powers of the 42S, some 41cx extensions (like X?NN), the 17B/27S solver, and the 35S extensions. For this last one, Plus42 already have the ability the use equations in programs, and vector/matrices litterals in equations. It will soon have constants. The only missing stuff I could think of would be fraction mode and one-line complex numbers entry, but I'm not sure there is a lot of demand for that. Also an ALG mode, but I think nobody is interested, this is a RPN kingdom

[Marco Polo]

(12-19-2021 03:46 PM)Vincent Weber Wrote:  

(12-19-2021 01:36 PM)Marco Polo Wrote:  It seems to me a very smart idea!

I think Plus42 is becoming something very near my "killer Calc" (the ideal would be a DM48, based on 48gx but without graphic)

Plus42 already is my killer calc, I always wanted a pocket programmable (not a monster like the 48) merging the powers of the 42S, some 41cx extensions (like X?NN), the 17B/27S solver, and the 35S extensions. For this last one, Plus42 already have the ability the use equations in programs, and vector/matrices litterals in equations. It will soon have constants. The only missing stuff I could think of would be fraction mode and one-line complex numbers entry, but I'm not sure there is a lot of demand for that. Also an ALG mode, but I think nobody is interested, this is a RPN kingdom

To become a full replacement of my 50g (considering the way and the features I use it) only two things are missing: constants and a multiple equation solver (like MES or Solvesys).
And, of course, the port to Swissmicros DM42 platform :-)

[roberto_abraham]

(12-19-2021 03:38 PM)Thomas Okken Wrote:  I don't know about the solid angle, though. The 48G has only the Steradian, which is a base unit and doesn't convert to anything else: 1_sr 1_r^2 CONVERT => Inconsistent Units.

I'm not too surprised. Being dimensionless, solid angles seem to pose significant challenges for unit conversion routines. In any case, I think if the only thing that Plus42 can't handle is solid angle conversions you'll be doing very well... it's definitely not a show-stopper!

Bob

P.S. According to Wikipedia the SI committee themselves changed their view on whether or not to consider Steradians as derived or more fundamental: "The steradian was formerly an SI supplementary unit, but this category was abolished in 1995 and the steradian is now considered an SI derived unit."

[Thomas Okken]

(12-19-2021 08:04 PM)roberto_abraham Wrote:  Being dimensionless, solid angles seem to pose significant challenges for unit conversion routines.

No, dimensionlessness is not a problem. It can convert between degrees, arc minutes, radians, etc. But steradians are defined as their own thing, not based on the square of a flat angle, and that's why it won't convert between, say, sr and r^2.

[Thomas Okken]

Not that it couldn't be. Mathematically speaking, one steradian is one square radian... But if there are undesirable side effects to changing the definition of sr from base unit to r^2, another option would be to simply add appropriate definitions for square degree, square arcminute, etc., based on the steradian.

[roberto_abraham]

(12-19-2021 09:18 PM)Thomas Okken Wrote:  Not that it couldn't be. Mathematically speaking, one steradian is one square radian... But if there are undesirable side effects to changing the definition of sr from base unit to r^2, another option would be to simply add appropriate definitions for square degree, square arcminute, etc., based on the steradian.

That sounds to me like the most straightforward way to handle it.

I'm really excited by all this functionality... Plus42 is going to be such a handy tool for quick calculations!

Bob

[tommi60]

Yes, outstanding work! I don't see the use of my beloved HP-17 BII or HP-27S anymore, plus42 will cover all my needs.