Public Beta Availiable - Win/Mac/Firmware [UPDATED]

[Anders]

(11-12-2017 04:50 AM)AlexFekken Wrote:  

(11-11-2017 04:33 PM)Anders Wrote:  It also does not matter if you change the Num X Step and NUM Y Step to something smaller say 0.0000001 because that is in the Graph 3D Num Set Up screens.

Like I mentioned in the "How to specify +/- in a function definition." thread where this was first discussed, you can get a better result by increasing the Grid size in the Setup. Unfortunately this is currently limited to 100x100, so at this stage it may be more productive to request an increase in the max grid size, rather than the implementation of what would be an entrirely different algorithm.

Just my 2c.

Thanks for pointing out the grid size. It could work to allow higher numbers than 100x100 but it would also need more CPU time (the reason why I think HP limited to 100 in the first place). If you allowed 1000x1000, I suspect the system would need to evaluate 2 orders of magnitudes approx: 10x10=100 more points (I am guessing how the algorithm work here). Basically, it will take 100x more time to plot the surface...

So, I do think they need to modify the algorithm to handle cases like this, not to bog down the CPU unnecessarily.

I just did a few others cases like this (not hard) that will surely generate a negative number inside a square root here and there on the surface:

for instance try:
FZ3 = √(1-SIN(X)^2-COS(Y)^2) or
FZ4 = 1/√(4-X^2-Y^2)

As you can see, there are holes/anomalies here and there.

So net, net: I think the plot algorithm need to:
- NOT imminently terminate when Z(X+dx,Y+dy) fails to evaluate (like in the case of negative square roots)
- instead: if Z(X,Y) worked but Z(X+dx,Y+dy) failed, use a smaller step size (e.g. try smaller steps: dx=dx/2, dy=dy/2) until Z(X+dx,Y+dy) evaluates withour failure, then plot and repeat stepping with smaller step until there is no visible hole given the screen setting. Basically when Z(X+dx,Y+dy) and Z(X,Y) generats the same pixel or x+dx and x is the same point (in the grid) you are done. You do need to floor dx,dy to not go beyond an absolute minimum step size to guaranteed this corner case loop terminates.

[Tim Wessman]

(11-09-2017 04:29 PM)komame Wrote:  I think the values of transparency parameters in graphical commands should be consistent.

RECT_P - 0 = opaque, 255 = transparent
PIXON_P - 0 = opaque, 255 = transparent
BLIT_P - 255 = opaque, 0 = transparent

Yes, it should be consistent. We may have missed this and may have to break some older programs to make the consistency happen. Please send an email with that bug report (and anyone else who feels they have some feedback). It is very difficult to track bugs in a forum posts as they get lost easily.

[Tim Wessman]

(11-10-2017 04:26 AM)webmasterpdx Wrote:  You probably already know this, but not all the new commands are in the help (e.g. cloc2, clop2)....not that the help for these commands even in XCAS is very good :-)

Did not notice that was missed. Please send an email feedback.

Quote:Also, I'm not sure that putting the user functions in the catalog is of any use. Keeping the catalog as brief as possible is a good thing, as often you might want to view functions in the catalog as you are trying to find a particular function.

The issue is that if you DO NOT have them in there, you need to remember where and in which category it is to find the commands. Having at least one central location is quite helpful in this regard. All the original subdivision is still there, you just have a new option now.

[Tim Wessman]

(11-10-2017 09:53 AM)salvomic Wrote:  Some programs of mine are still useful. However now the new Finance app does 90% of things that HP 12C can do, and I find this a great goal.

90%??????

More like the 12C does FAR less (at least if you exclude falling back to base formulas and manually doing everything).

[Tim Wessman]

(11-10-2017 09:56 PM)dg1969 Wrote:  In function app go to plot view and then press help key... the text concerns the stat2var app; ( in french language).
-----
off topic on
-----
I would love to use the theme "dark colors" unfortunately the light colors proposed in page 2/4 of the home / setting menu does not allow good contrast...(for example in the spreadsheet application it is impossible to distinguish the selected cells) except to place the calculator on the desktop upside down so that the direction of vision is slightly oriented from top to down of the screen...

Please send an email with the feedback. Thank you!

[Tim Wessman]

(11-11-2017 11:19 AM)webmasterpdx Wrote:  The menu key only shows me zoom options....nothing related to shading.
I didn't see anything related to surface shading in the settings. There is the drawing method (checkerboard, etc..).. but I wanted to keep the drawing method and change the extent of shading, as certain surfaces like a wide conic (where the curve is very gradual) is often difficult to see (looks like a triangle instead of a cone, as there isn't enough contrast between darkest and lightest spot).
??

I see what you are saying. The appropriate way to deal with this is to change you Z min/max I think. That will change your shading.

[AlexFekken]

(11-12-2017 06:04 PM)Anders Wrote:  Thanks for pointing out the grid size. It could work to allow higher numbers than 100x100 but it would also need more CPU time

Obviously, and more internal memory as well.

But that impact would be predictable and expected (if you have any idea what you are doing). The impact of an adaptive algorithm on the other hand will be much harder to predict and might be even worse if any singularities are present (and this is not even taking into account the effort to create such a new algorithm).

The fact that the current Function app has the same issue (try Y = sqrt(X-pi)) is also an indication that singularities are just difficult to deal with. Which is why I think it would be better to divert the development efforts (for a future upgrade) to supporting implicitly defined and/or parametrised surfaces. Not that that would be trivial either of course, but it would probably be a better ROI than trying to patch up an algorithm for situations for which it is not well suited by design. After all: this 3D stuff works really well for smooth functions already.

[Anders]

(11-13-2017 01:05 AM)AlexFekken Wrote:  

(11-12-2017 06:04 PM)Anders Wrote:  Thanks for pointing out the grid size. It could work to allow higher numbers than 100x100 but it would also need more CPU time

Obviously, and more internal memory as well.

But that impact would be predictable and expected (if you have any idea what you are doing). The impact of an adaptive algorithm on the other hand will be much harder to predict and might be even worse if any singularities are present (and this is not even taking into account the effort to create such a new algorithm).

The fact that the current Function app has the same issue (try Y = sqrt(X-pi)) is also an indication that singularities are just difficult to deal with. Which is why I think it would be better to divert the development efforts (for a future upgrade) to supporting implicitly defined and/or parametrised surfaces. Not that that would be trivial either of course, but it would probably be a better ROI than trying to patch up an algorithm for situations for which it is not well suited by design. After all: this 3D stuff works really well for smooth functions already.

Y = sqrt(X-pi) looks fine to me in the Function app (no hole in the curve when x -> pi from positive side). In fact it actually proves the point that it is possible and that useful functioning source code already exists in Function app!

To bring home the point directly:
Try plotting the half circle function Y = sqrt(1-X^2) in the Function app
and you get a smooth half circle without gaps as x->1 or x->-1. HP can simply reuse parts of the working code from the Function app and put it in 3D app.

Also, I disagree that this is a hard problem to solve even if they cannot reuse the Function App code in part. I suggest that fixing the algorithm is in the 10s of lines of code in C++, C, Java, C# or similar language (not sure what language the 3D app is written in). Any decent 2nd year Computer Science college student who have taken some basic Numerical Analysis class can do this if given access to the source code - we are talking hours not days.

[AlexFekken]

(11-13-2017 02:16 AM)Anders Wrote:  Y = sqrt(X-pi) looks fine to me in the Function app (no hole in the curve when x -> pi from positive side).

Ah, you are probably looking at the public beta, which does seem to have this fixed. The current production version does have the gap, and presumably it has had it for a while because it wasn't that easy to fix.

I am not so sure that going from a fixed, i.e. equidistant, X-Y grid to a completely variable one (depending on the function(s) plotted) can be done in "10s of lines of code". You'd probably need a totally different data structure just to hold the data to begin with. But if you say so...

[Carlos295pz]

Graphic error? when printing empty line or single line. Sent by mail

[webmasterpdx]

Tim, you've asked people to send a feedback email. To you directly?
I presume you just want more details right?
Thx

[Dirk.nl]

Read the first post!
calcbeta@hp.com

Dirk

[salvomic]

(11-12-2017 11:33 PM)Tim Wessman Wrote:  90%??????

More like the 12C does FAR less (at least if you exclude falling back to base formulas and manually doing everything).

:-D ;-)

[salvomic]

(11-13-2017 05:51 AM)webmasterpdx Wrote:  Tim, you've asked people to send a feedback email. To you directly?
I presume you just want more details right?
Thx

hi Donald,
you can send a feedback via the Connection Kit program, see the image below.

Salvo

[Eddie W. Shore]

Anyone else has this problem:

When creating a program through the Program Catalog, the program is created but the calculator resets. Same thing when I try to delete programs.

This happens to only affect C hardware.

Eddie

I have notified calcbeta@hp.com of this

[JohnBest]

(11-14-2017 02:36 AM)Eddie W. Shore Wrote:  Anyone else has this problem:

When creating a program through the Program Catalog, the program is created but the calculator resets. Same thing when I try to delete programs.

This happens to only affect C hardware.

Eddie

I have not had this problem on either my rev. A or C hardware.

(Long time reader, first time poster).
John

[Eddie W. Shore]

(11-14-2017 04:44 AM)JohnBest Wrote:  

(11-14-2017 02:36 AM)Eddie W. Shore Wrote:  Anyone else has this problem:

When creating a program through the Program Catalog, the program is created but the calculator resets. Same thing when I try to delete programs.

This happens to only affect C hardware.

Eddie

I have not had this problem on either my rev. A or C hardware.

(Long time reader, first time poster).
John

Welcome John! Glad to have you here with us!

[DrD]

With the public beta firmware, I'm not seeing this problem, either. On two C versions, I can create or delete programs using [Shift][Program]. [Shift][Program][New] to create a program, and [Shift][Program][More][Delete], to delete the program.

When strange things happen, re-formatting sometimes works miracles!

-Dale-

[toml_12953]

(11-14-2017 02:36 AM)Eddie W. Shore Wrote:  Anyone else has this problem:

When creating a program through the Program Catalog, the program is created but the calculator resets. Same thing when I try to delete programs.

This happens to only affect C hardware.

Eddie

I have notified calcbeta@hp.com of this

Yes, I reported this quite a while ago but it happens on my HW A as well as my HW C version. I get rid of it by copying all my programs to a PC (NOT backup!) reformatting the storage and copying programs back to the calculator one at a time testing the new command after each until I find the program that causes the error. I then reformat again, copy the programs one at a time omitting the program that caused the error and test again. If I find more programs that cause the error, I go through the whole process again each time omitting the error-causing program. Eventually none of the programs in memory cause a reboot. I then take the omitted program(s) and reenter them by hand into the CK and from there copy them to the calculator. This whole thing is easier than it sounds and is the only sure cure I've found. The problem is a corrupted storage area.

[Anderson Costa]

Hi for all!!! Thanks for the new Prime Update. I will test to see any bugs and report here.
EDIT 1: The new 3D graphing app is better than Han's classic Graph 3D app... Tested f(x,y)= sin xy+ cos xy as an example. The plot is better in the new 3D graphing app than Han's app. It is a greatest start for these new HP Prime application, since the 3D graph is better plotted than Casio graphing calculators actually (Does TI color graphing calculators having a great detail in 3D Graphing plotting???)
2- The new skin remembers me of the classic HP graphing calculators such as HP 50G. If this skin were available before, it give me some nostalgia...
3- The CAS operations, such as calculating integrals for example, only give me exact results when I have in home mode. The elementary calculus, like calculating limits and derivatives is now a CAS only command. Is there any way to find approximate results of a CAS command???
4- No problem with changing languages so far. I have seen help comments in Portuguese now. You can enter in Apps menu or any other app to change calculator language.
I am anxious for the new Prime official update after this.