Hide soft menus?

[Lithium]

I wish you could hide soft menu on a 17BII

[Software49g]

Hello,

here is a 50g where the soft menus are hidden and the free space is used for the stack.

The menus are active, so pressing the corresponding soft menu key will execute it, they are simply not shown.

Additionally errors can be shown as in the HP 48 using the status line for showing error messages.

Both features are controlled through flags, so you can choose whether you want to use it or not.

The code is already quite stable, some minor quirks/bugs needs to be sorted out.

Enjoy,
Andreas
http://www.software49g.gmxhome.de

[JDW]

Andreas!

I must say, your code appears to be the best solution. I love the idea of setting it via flag. When will your code become available for us to beta test?

Also, I like the battery level indicator in your screenshots. What software is that?

Thanks,

James W.

[DavidM]

Initially I was a bit skeptical of the value provided by removing the menus, but decided to try it out anyway. My first attempt was a prototype app that altered the system outer loop just enough to skip menu drawing and extend the stack downward:


The more I used the calculator without the menus, the more it grew on me. I also tried removing the stack line numbers from the display:

(a feature I refer to as "Joe Horn Mode" because he was once willing to go to great lengths to get rid of them

The result is a very clean and simple display, and there's an aesthetic to it that (for me) makes it easier to focus on the numbers as I'm working. It didn't take long to realize, though, that there were still plenty of times that a function or app was needed that temporarily required menu access. Quitting and restarting the app was easy, but a nuisance. This caused me to refine the approach somewhat.

Attached to this post is a "proof of concept" library for the 50g that attempts to hide the menu and status areas in a semi-intelligent way. You can think of it like this: the default display mode has no menus and 0-2 lines of header (user-selectable). If you perform some action that otherwise would have changed the displayed menus (such as pressing VAR, TOOL, or any other menu-changing feature), the menus and header will temporarily return. Executing a system or user function will work as usual, then the menus/header will be removed again when the function completes. There's a bit more to it, but that's essentially the paradigm this is based on.

I'm not suggesting this is a significantly better approach than the standard display. Rather, it's a simpler alternative that works best when you aren't needing all the status info and functions that normally surround the stack. For me, that situation is actually quite common.

If you're curious, I've made it fairly simple to try this out. The attached file is a zip containing an installer and README info. The installer stores library 1420 (NoMnu) into port 0 and attaches it to your HOME directory. In its current form, the library works best in port 0. I'll probably make some changes later to support storing it in ports 1 or 2 along with some other planned tweaks. Removal is a simple one-button selection from the library's menu.

I encourage you to at least give it a try, and please share your thoughts. I've found it surprisingly nice to be able to remove some of the clutter from the display. When needed, it's very easy to temporarily show the "normal" info just by pressing the ON/CANCEL key (which toggles the display mode). Otherwise you can simply exit the app to resume normal menu operations. If you're like me, there are plenty of times when you just need to bang through some calculations without using a lot of menu-bound functions. That's when this type of interface is most useful.

[Software49g]

Hello,

some remarks:

1. The EditLine aka CommandLine is at its original position. I would expect that the EditLine uses the space of the menu line if the menu is turned off.





Achieving this for Font8/Font7/Font6/MiniFont and all headers ist quite challenging...

2. The RAM area that is used for the display while an EditLine is active allows a max. of 12 display lines (Font6/MiniFont with Header0 and the menu activated).
Getting this to work with the menu off (max. 13 display lines) is even more challenging...

Regards,
Andreas

[DavidM]

Hey Andreas -

Glad to see that you're still working on this. I wasn't sure if you were still pursuing the idea.

(11-17-2016 11:46 PM)Software49g Wrote:  1. The EditLine aka CommandLine is at its original position. I would expect that the EditLine uses the space of the menu line if the menu is turned off.

Achieving this for Font8/Font7/Font6/MiniFont and all headers ist quite challenging...

As you know, the command line position is one of the first obstacles that comes up when altering the SOL to extend the stack. I looked into this just long enough to realize that it was going to take a lot more time and effort to move it than I was prepared to put in for this experiment. That's why I currently just bump up the stack contents above the existing command line during data entry in the current version. Good to see that you've already done that part!

(11-17-2016 11:46 PM)Software49g Wrote:  2. The RAM area that is used for the display while an EditLine is active allows a max. of 12 display lines (Font6/MiniFont with Header0 and the menu activated).
Getting this to work with the menu off (max. 13 display lines) is even more challenging...

Which makes me even more glad that I didn't continue to pursue moving the EditLine down.

What are your thoughts about the concept of a menu that temporarily returns?

[Software49g]

Hello,

> What are your thoughts about the concept of a menu that temporarily returns?

That would be trival (once everything is working without the menuline).

Regards,
Andreas

[DavidM]

(11-20-2016 07:33 AM)Software49g Wrote:  > What are your thoughts about the concept of a menu that temporarily returns?

That would be trival (once everything is working without the menuline).

I guess I should have been more clear with my question. What I meant was more along the lines of whether you perceived the concept of menus reappearing temporarily to have merit, and if so, how would you envision the user experience for that type of functionality?

The current "model" for an off-the-shelf 50g uses a static display configuration which only changes if you manually alter specific settings. I suppose what I am proposing is that there could be value in having a display configuration which changes automatically as a result of contextual differences as well.

I've already put forth one example of how it might work with menus, but I don't presume that it's the best way. It may not even be a desirable feature. Other considerations include things like the mode indicators (angle, number base, etc.). Should those be taking up space in the header all the time? Are there other indicators that could go in that limited space if they changed contextually? My hope is that others might have thoughts to share on this subject that might make a more spartan interface more informative and easier to use.

Thanks

[JDW]

While Software49g/Andreas takes his time to answer David's question, I feel compelled to chime in and say that David's code is an absolutely fabulous piece of work.

If any of you reading this have not made time to download it and give it a try, you are really missing out on something truly fabulous. This is the best piece of 50g software that I've seen come in the seen in a long time, if ever. Stop making excuses and install it today!

I've attached the latest version WITHOUT *.hp filename extension, just in case some of you had problems installing the version with filename extension. But it should not matter either way.

Once you've installed it and given it a try, post your feedback here.

Thanks,

James W.

[Software49g]

Hello David,

I have already expanded the status area with a battery symbol and the currently free memory available which is adjusted at every key press. Whether this is meaningful for anybody else is out of my scope. I wanted it for myself - so I programmed it into the MLP while developing it.
The battery indicator is probably useful, as most other portable electronic devices have one.
About the memory indicator I am not so sure. It is nice to have, but basically it will only tell you when the next garbage collection will occur.

About your proposal about context sensual menus I believe that they should be static (or user definable but still static). I find it most annoying having a software that sorts it functions by its use or some other mechanism. I rather prefer having functions at the same place where I know where they are instead of having a "ghost" in the machine that resorts them in a way.
If the menus are structure well, than there is absolutely no use for it or in other words: if there is need for it something is wrong with the user interface IMHO.

Regards,
Andreas

[Software49g]

Hello,

this is the English message library from the MLP that supports toggling the menuline.
The cursor behaves correctly towards the additional space that is gained for the command line.
This was/is more complicated than I thought as there are some significant differences between the ARM adjusted ROM used in a real 50g and the pure Saturn ROM used in Emu48 and also there are differences between the Saturnator and Emu48 - especially if it comes to low level routines.

UserFlag 79 toggles the menuline and is mapped to the [TOOL] key. Pressing the [TOOL] key for approximately one second will toggle the menuline.

SystemFlag 4 toggles the battery indicator and the currently free memory available shown in status line one.

SystemFlag 127 (warmstart required after setting or clearing it) allows HP 48SX / 48GX error style in the status line of the display through the reserved variable STARTERR. Note: this will overwrite or delete an existing STARTERR.

Required is a ROM > 2.00 - ROM 2.15 is recommended.


Known Issues:
- This is BETA sw, so back up your memory. You have been warned.
- Not compatible (due to internal code call changes) with MLP/OSE/help for all commands (this is beta sw - of course this will be adjusted if there is an official release).
- The system outer loop (SOL) is replaced with a custom version that runs from precious RAM - so this requires a little bit more memory as the built-in SOL.
- The menus are active, so pressing the corresponding soft menu key will execute it, they are simply not shown.
- The RAM area that is used for the display while an EditLine is active is hardcoded and allows a max. of 12 display lines (Font6/MiniFont with Header0 and the menu activated) - calling this twice with different parameters might allow 13 lines with Font6/MiniFont but at present it is limited to 12 display lines.
- Minor display issue if scrolling down in the edit line and Font6/MiniFont is used.
- Needed ROM routines are copied to RAM, modified and run from there - the RAM area used is not reviewed by the Garbage Collector, so I believe this is garbage collection safe.
- Tested only in Port0 (but if there is an official release it will run from Port0/1/2).
- Tested only in RPN mode.
- Currently there is only this English version, the goal is to make it available for all languages.
- However, I am running out of space - especially in the German version, as Germans love long words like "Donaudampfschiffahrtskapitän".


The attached software runs in a real 50g and in Emu48 emulating a 50g (although there are slightly different timing issues).
Please report any bugs that you might find, thank you.


Enjoy,
Andreas

[DavidM]

Hey Andreas -

This is a very nice implementation of what James was originally looking for! You've obviously put a lot of work into this and it shows.

I particularly like the "one second press" of the TOOL key to toggle activation. It's a nice implementation touch that shouldn't interfere with any other key assignments.

I'll try to spend some more time with this when I can and give some more feedback at a later date. This time of year gets to be a bit busy for me, so it probably won't be in the next few days. But please don't think that a lack of response indicates lack of interest.

Thanks!
- David

[JDW]

(12-06-2016 02:28 PM)Software49g Wrote:  SystemFlag 4 toggles the battery indicator and the currently free memory available shown in status line one.

SystemFlag 127 (warmstart required after setting or clearing it) allows HP 48SX / 48GX error style in the status line of the display through the reserved variable STARTERR. Note: this will overwrite or delete an existing STARTERR.

I just installed the Library. Oh my! This library is intriguing! I must say I am shocked and dismayed by the apparent lack of interest though. There are no posts from people other than myself, David and Andreas. Have all the 50g users left earth? :-) Seriously, folks, these two Libraries are fabulous software for the 50g.

Since no detailed instructions were given for beginners, I shall provide such as follows...

INSTALLATION (Make sure you Uninstall David's Library first):
1. Copy "MessagesEnglish.hp" to your SD card.
2. Put the SD in the 50g and press LeftShift (WHT) and APPS to bring up the File Manager.
3. Select "3: SD" with your silver arrow keys and press the OK softkey.
4. Select "MESSAG~1.HP" (the LIB, not the STRNG) and press the COPY softkey.
5.Copy it to Port 0.
6. Press ON and F3 for a Warmstart. You should briefly see some text your normally don't see, confirming the Library has been installed.
7. Verify operation by pressing and holding the TOOL key. The menubar should vanish. Repeat and it should come back.


Most of you know FLAGS are accessible via the MODE button, but if you go there you will not see 04 or 127 that Andreas mentions. Although there are faster ways of doing it, here is one way to SET & CLEAR the FLAGS Andreas mentions...

CHANGE FLAGS:
1. LeftShift (WHT) & EVAL to bring up the PRG menu.
2. Press the NXT button.
3. Press the F4 to select MODES.
4. Press F3 to select FLAG.
5. Type 4 and ENTER.
6. Now to make it a SYSTEM flag, you need to press the "+/-" key to make it Negative.
7. Press the softkey associated with "SF" to set it. Notice the battery indicator appears in the upper right corner of the screen, along with free memory.
8. To toggle OFF the battery indicator, repeat above steps 1 thru 6, then press the softkey associated with "CF" to clear the flag. Note the battery indicator then vanishes.
9. To change USER FLAGS, repeat steps 1 thru 5 and skip Step 6 because USER FLAGS must be Positive. Only SYSTEM FLAGS are Negative.


For good measure I repeated a test that David had given me for use when testing his Library. There is a slow-down when using David's library that occurs when you have many things on the stack (way more than 50, in my experience). I repeated the test with Andreas' Library installed and the menu bar toggled OFF. No speed slowdown that I can see. Here is the test:

1) Enter 12345 on the stack
2) Enter 800 on the stack
3) Enter NDUPN (this loads the stack)
4) Press any function key a couple of times ('+' or '*' is fine).
5 ) Now start typing in some more numbers.

Again, I did not notice any performance hit when using Andreas' Library, but you should still do the test for yourselves.


Now for my Bug Report:

When I first pressed RightShift (ORG) and SYMB to bring up CAT, I got an insufficient memory error. But that was after I had run the above test (duplicating things on the stack 800 times), so I pressed ON and F3 to Warmstart and tried again. But that next time I tried I got an Hourglass forever -- the 50g locked up. I had to do another Warm start to break out of it. The Memory indicator in the upper right says 13004. But no matter how many times I try, pressing RightShift (ORG) and SYMB (CAT) produces the Hourglass lockup. This is a showstopper bug.

Best wishes,

James Wages

[Joe Horn]

(12-09-2016 01:12 AM)JDW Wrote:  I must say I am shocked and dismayed by the apparent lack of interest though. There are no posts from people other than myself, David and Andreas. Have all the 50g users left earth?

Many of us lurkers are simply struck speechless by the sophistication of the proposed solutions. We feel like when Wayne and Garth told Alice Cooper, "We're not worthy!" in Wayne's World.

[Software49g]

Hello,

> The Memory indicator in the upper right says 13004. But no matter how many times I try,
> pressing RightShift (ORG) and SYMB (CAT) produces the Hourglass lockup. This is a
> showstopper bug.
Can you reproduce this? As it is working fine here.
Please provide all needed information, so that I can reproduce this and thus be able to fix it.

> Again, I did not notice any performance hit
There is a performance hit – as the SOL has to do a lot more things now than the one in ROM.
However, it should not be notable under normal circumstances (less than 3.000 objects on the stack).

> Since no detailed instructions were given for beginners, I shall provide such as follows...
The final product comes with an installation program.
https://www.youtube.com/watch?v=fiTFUXBgSTs

HTH + BR,
Andreas

[JDW]

(12-09-2016 08:36 PM)Software49g Wrote:  > The Memory indicator in the upper right says 13004. But no matter how many times I try,
> pressing RightShift (ORG) and SYMB (CAT) produces the Hourglass lockup. This is a
> showstopper bug.
Can you reproduce this? As it is working fine here.
Please provide all needed information, so that I can reproduce this and thus be able to fix it.

Yes, it is reproducible. What information do you need me to provide?

Of course, we need other 50g users to try it too. Wayne and Garth... Please give it a try!

[rprosperi]

(12-09-2016 01:12 AM)JDW Wrote:  I must say I am shocked and dismayed by the apparent lack of interest though. There are no posts from people other than myself, David and Andreas. Have all the 50g users left earth? :-) Seriously, folks, these two Libraries are fabulous software for the 50g.

Since no detailed instructions were given for beginners, I shall provide such as follows...

The BETA and compatibility warnings (fully appropriate to do so and appreciated) are bit of an impediment; I suspect many folks are waiting for the more intrepid explorers like you to test drive and comment. And it's only been a couple days after all.

Thanks to you James for taking the time to document the procedures (though they are surprisingly simple for such a powerful and invasive library) and your comments/feedback.

Can this library be uninstalled using normal library removal procedures, and is any cleanup needed?

[JDW]

Well, after much fiddling, and with the menu bar turned off, I was able to get CAT to display, but the content is garbled and the 50g locks up with the hourglass:



Pressing ON and F3 to Warmstart results in Andreas Library getting uninstalled and now every time I warm start, even without an SD card, I still briefly see the following message:

Waning:
Invalid Card Data

When I now press Rightshift (ORG) and 2 to get LIB, I see only 0: 1: 2: and nothing else, showing how all previously install libraries are now uninstalled. However, if I press 0: I still see 261 and 772 saved there.

At this point, I pressed LeftShift (WHT) to open File Manager. I then selected 0: IRAM and pressed the OK softkey. With "English Messag L261 94136" selected, I pressed the NXT button and then the PURGE softkey, followed by the YES softkey to confirm the delete. But I get an Error...

Error:
Object In Use

Not sure what to do at this point.

My other library stored in IRAM is 130271 in size, so maybe all these problems happened due to insufficient memory in IRAM?

I then used a narrow tool to press the recessed RESET button on the back of the 50g. I then used File Manager and this time I could PURGE Andreas' Library. 0: IRAM now shows 233KB free.

Next, I used File Manager to COPY Andreas Library back to IRAM and then did a Warmstart. Free RAM in the upper right corner says 143241. I can now get CAT to display and bring up the Periodic Table just fine. No lockups. So clearly the presence of my other big Library before caused Andreas' Library to choke and freeze my poor 50g.

I then performed another Warmstart so I could see 143241 of free RAM in the upper right again. I pushed-and-held the TOOL button for more than 1 sec. and released, and my menu bar vanished (as it should). Free RAM down to 121650. I long-pressed TOOL again to toggle the menu bar back ON. Free RAM down to 105986. I pressed TOOL to toggle again and now RAM reports 89299. Repeated toggle again and now it's 74147. Again and now it's 59796. Then 45196. Then 30621, 18808, 5098, and then it's back to 141597 again. No lockup, so I guess that's normal?

I look forward to hearing advice on what I should do next.

Thanks.

[DavidM]

(12-10-2016 02:00 AM)JDW Wrote:  Error:
Object In Use

Not sure what to do at this point. ... I look forward to hearing advice on what I should do next.

James,

Try holding down the backspace key immediately after doing ON-C to warmstart. Keep it down until you see the normal stack display. That will block the configuration programs for libraries from running the way they normally do on a warmstart.

At that point you should hopefully be able to do the usual steps to uninstall.

[JDW]

(12-10-2016 02:14 AM)DavidM Wrote:  Try holding down the backspace key immediately after doing ON-C to warmstart. Keep it down until you see the normal stack display. That will block the configuration programs for libraries from running the way they normally do on a warmstart.

Thanks for the tip, David. Please note that I edited/updated my previous post several times before spotting your reply, so refer to that post to see all I tried. But I just reinstalled my big 722 library and then Andreas library, which prevents CAT from displaying and which locks up the 50g with the hourglass. Using your tip allows me to PURGE Andreas' Library without resorting to a RESET.

So the lockup problem I am seeing is either an incompatibility with my large 772 library, or it is locking up simply due to insufficient RAM in IRAM. For those of you testing along with me, try adding a big Library like 132KB in size or an equivalent number of Libraries in IRAM, then install Andreas' Library, then try to open CAT and see if you get a lockup. Andreas, you should try that too.

Thanks,

James