WP 34S --> WP 31S

[Thomas Klemm]

(04-03-2014 08:42 AM)fhub Wrote:  What has happened to the WP34s/WP31s code section on Sourceforge?
http://sourceforge.net/p/wp34s/code/

All files have gone ...

Sikuq, you should really stop hitting that delete button.

[walter b]

(04-03-2014 04:08 PM)Thomas Klemm Wrote:  

(04-03-2014 08:42 AM)fhub Wrote:  What has happened to the WP34s/WP31s code section on Sourceforge?
http://sourceforge.net/p/wp34s/code/

All files have gone ...

Sikuq, you should really stop hitting that delete button.

Sourceforge is obviously S...proof.

[rprosperi]

(04-03-2014 04:08 PM)Thomas Klemm Wrote:  

(04-03-2014 08:42 AM)fhub Wrote:  What has happened to the WP34s/WP31s code section on Sourceforge?
http://sourceforge.net/p/wp34s/code/

All files have gone ...

Sikuq, you should really stop hitting that delete button.

(04-03-2014 06:04 PM)walter b Wrote:  

(04-03-2014 04:08 PM)Thomas Klemm Wrote:  Sikuq, you should really stop hitting that delete button.

Sourceforge is obviously S...proof.

While my wife just doesn't see it, I keep telling her that the humor in MoHPC forum postings is only matched by it's brilliance... Thank you Thomas and Walter!

[Paul Dale]

(04-03-2014 09:34 AM)walter b Wrote:  What about <- ? Shall be no command unless executed as CLx.
Sigma+ ? Shall be reversible by UNDO though there's no pressing need since Sigma- exists.
ENTER, FILL, CLALL, RESET ? Shall be commands.

These are almost all correct as currently implemented.

The ones that aren't are CLALL & RESET and two that aren't mentioned CLRPROG & CLPALL. These take a confirmation and aren't command because of that -- UNDO could still be supported for them if desirable. What shall UNDO do if the confirmation answer is no?

Sigma+ and Sigma- are reversible by UNDO. So are CLREG and CLSTK.


- Pauli

[BarryMead]

I just downloaded the wp31s emulator from: http://sourceforge.net/p/wp34s/code/HEAD...hes/wp31s/
and discovered a small error in the "sin" function and the "To Rectangular" function (because it uses the sin function).

With the calculator in degrees mode if you take the sin(90) or sin(270) the sign of the returned value
is inverted. sin(90) returns -1 (not +1), and sin(270) returns +1 (not -1). This also affects the "to Rectangular"
function which uses the sin function.

The sign of the returned result is incorrect for all angles between 90 degrees and 135 degrees, and
for angles between 270 degrees and 315 degrees.

All other angles return the correct result, and it seems that in radians mode all angles return the correct result.

Also I noticed that when you press "f mode" and try to select ssize8 the stack size stays at size 4.

Thanks for building another calculator, I realize that there will be plenty of tiny mistakes initially.
Keep up the excellent work.

[Paul Dale]

An oops for me to fix.

Thanks for finding this.


- Pauli

[Paul Dale]

I think I've fixed the trig sign problem, no build yet.

- Pauli

[Sanjeev Visvanatha]

(04-03-2014 10:11 PM)BarryMead Wrote:  Also I noticed that when you press "f mode" and try to select ssize8 the stack size stays at size 4.

Thanks for building another calculator, I realize that there will be plenty of tiny mistakes initially.
Keep up the excellent work.

I checked the catalog entry in the code, and it seems correct. The mode switch to SSIZE8 was working previously. I did not test it after some recent modifications. I wonder if this is related to the recent modifications required to limit the number of registers on 31S?

[Jonathan Cameron]

(04-04-2014 01:14 AM)Sanjeev Visvanatha Wrote:  

(04-03-2014 10:11 PM)BarryMead Wrote:  Also I noticed that when you press "f mode" and try to select ssize8 the stack size stays at size 4.

Thanks for building another calculator, I realize that there will be plenty of tiny mistakes initially.
Keep up the excellent work.

I checked the catalog entry in the code, and it seems correct. The mode switch to SSIZE8 was working previously. I did not test it after some recent modifications. I wonder if this is related to the recent modifications required to limit the number of registers on 31S?

For that matter, do we really need SSIZE4? Couldn't we make the 8-level stack the default an only mode? (Getting out my asbestos suit...)

-Jonathan

[Sanjeev Visvanatha]

(04-04-2014 01:20 AM)Jonathan Cameron Wrote:  For that matter, do we really need SSIZE4? Couldn't we make the 8-level stack the default an only mode? (Getting out my asbestos suit...)

No flame thrower here. However, I think some sane defaults need to apply. SSIZE4 should be default. SSIZE8 is appealing, but would not promote inside-out methods of calculation, IMO.

The traditional X display should also be default, rather than the YREG display. That one is confusing (to me), and I now see it is on by default.

[Jonathan Cameron]

(04-04-2014 01:27 AM)Sanjeev Visvanatha Wrote:  

(04-04-2014 01:20 AM)Jonathan Cameron Wrote:  For that matter, do we really need SSIZE4? Couldn't we make the 8-level stack the default an only mode? (Getting out my asbestos suit...)

No flame thrower here. However, I think some sane defaults need to apply. SSIZE4 should be default. SSIZE8 is appealing, but would not promote inside-out methods of calculation, IMO.

The traditional X display should also be default, rather than the YREG display. That one is confusing (to me), and I now see it is on by default.

The current build is only for testing purposes and does not indicate what the defaults should be for any of these new options.

-Jonathan

[BarryMead]

After some additional testing, I confirmed that the sign inversion also occurs in the cos function
in two angle regions. between (90.0001 and 135) degrees, and between (270.0001 and 315) degrees.

You probably already caught and fixed both cases, but I thought you should know.

Thanks again, Barry

[Paul Dale]

Yeah, part of the same problem. Sin and cos are calculated simultaneously and I'd stuff up the final sign correction step for some angles.


- Pauli

[walter b]

(04-04-2014 01:20 AM)Jonathan Cameron Wrote:  For that matter, do we really need SSIZE4? Couldn't we make the 8-level stack the default an only mode?

Congratulations! You're the first - after three years ...

d:-)

[walter b]

(04-04-2014 01:27 AM)Sanjeev Visvanatha Wrote:  SSIZE4 should be default. SSIZE8 is appealing, but would not promote inside-out methods of calculation, IMO.

The traditional X display should also be default, rather than the YREG display. That one is confusing (to me), and I now see it is on by default.

I concur for the traditional X display.

Regarding SSIZE4: 98% of all RPN calculators feature a 4-level stack - hence it's reasonable to continue supporting it. I agree on SSIZE8 being superior as explained in the manual. BTW, inside-out shall be applied with SSIZE8 as well.

d:-)

[Paul Dale]

(04-04-2014 05:13 AM)walter b Wrote:  Congratulations! You're the first - after three years ...

I'd prefer an unlimited stack


- Pauli

[walter b]

(04-04-2014 08:00 AM)Paul Dale Wrote:  I'd prefer an unlimited stack

An unlimited stack has a significant drawback: there's no top level which can be repeated. IMHO, 8 levels are a nice practical compromise connecting the best of two worlds.

d:-)

[Jeff O.]

(04-04-2014 01:27 AM)Sanjeev Visvanatha Wrote:  The traditional X display should also be default, rather than the YREG display. That one is confusing (to me), and I now see it is on by default.

(04-04-2014 01:55 AM)Jonathan Cameron Wrote:  The current build is only for testing purposes and does not indicate what the defaults should be for any of these new options.

If the yReg display is not the default, please make it either available in all builds as an option that can be enabled or disabled (as an option in the MODE menu along with stack size), or as separate bin files. I have no ability to build my own bin files with certain things enabled or disabled.

Thanks.

[walter b]

(04-04-2014 12:09 PM)Jeff O. Wrote:  If the yReg display is not the default, please make it either available in all builds as an option that can be enabled or disabled (as an option in the MODE menu along with stack size), or as separate bin files.

The yReg display will be toggled by YDOFF / YDON. Probable default will be YDOFF on both the WP 34S and the WP 31S.

d:-)

[Marcus von Cube]

(04-04-2014 12:37 PM)walter b Wrote:  The yReg display will be toggled by YDOFF / YDON. Probable default will be YDOFF on both the WP 34S and the WP 31S.

The commands need to be implemented as will be the necessary code to toggle the display at run time. So "will be" implies some undetermined future...