Search results for:

There is no Wiki yet but I'm very happy to set one up. This is the first time anyone asked, but it
makes a lot of sense. I guess the question is: If I set one up, is anyone here willing to contribute?

@Gordon - think now would be a good time for the Wiki. I have never done a docs update and the lifecycle of maintaining a git repo and doing pulls /merge / push etc is too much overhead for what could be done in minutes on a wiki. I think write access would have to be granted on the basis of those you trust to do updates etc AND anyone could request write access if they want to contribute.

I think calling load() will do it.
That returns you to the launcher.

Hi,

I have an app that now uses 7 custom fonts. The font is selected from a settings menu.
So I only need one active font to be loaded when the App starts up.

At the moment I have 7 seperate font functions that really bloat the App in terms of memory footprint.
I guess I should really just load the one font I need from storage for the lifespan of the App.

Just trying to get my head round how I physically do this - what code goes where and how it is loaded.

Graphics.prototype.setFontArchitect = function() {
// Actual height 40 (41 - 2)
var widths = atob("CBolByEeJykkJCYhCg...==");
var font = atob("AAAAAAAAAAAAAAMAAAAAAAAAAAAA....=");
var scale = 1; // size multiplier for this font

g.setFontCustom(font, 46, widths, 58+(scale<<8)+(1<<16));
};

Would you recommend that I create a module for each of these fonts that basically does something similar to:

https://www.espruino.com/modules/FontDylex7x13.js

How do I tell the apps.json file that the code has 7 custom font files ?
I think the answer is the way that modules are loaded - example gpssetup
The module file drops the .js from the source file name.
Then the App does require("myfont");

  "storage": [
    {"name":"gpssetup","url":"gpssetup.js"},

Then I think I would just load settings at the top of the code and run this bit of one time code for the lifetime of the App.

 if (settings.font == "Architect")
    require("f_artitect").add(Graphics);
  else if (settings.font == "GochiHand")
    require("f_gochihand").add(Graphics);
  else if (settings.font == "CabinSketch")
    require("f_cabin").add(Graphics);
  else if (settings.font == "Orbitron")
    require("f_orbitron").add(Graphics);
...

x =864*dt.getHours()+14.4*dt.getMinutes()+0.24*dt.getSeconds()+0.00024*dt.getMilliseconds();

let wait = 4167*(1-(x%1));

This is getting the fractional part of x. So if x is 456.123
The statement becomes.

let wait = 4167*(1 - 0.123);

So its working out a value for the next timer call.

@Numerist - what an interesting clock. Be interesting to see how an analogue version looked.

I tried it on my Bangle 1. I have noticed that the screen sometimes goes black before the LCD times out.

Thought I would add some comments to help understanding the code.
I also added a couple of console.log() calls to print messages to the IDE.
If you connect to the IDE you should be able to see some messages when the code runs
This will tell you if it was able to extract the time from the GPS chip real time clock circuit.

setInterval(function() {

  // this is defines the GPS call back function
  Bangle.on("GPS",function cb(g) {
    Bangle.setGPSPower(0,"time");    // we got a response so turn off GPS power
    Bangle.removeListener("GPS",cb);  // unregister the callback function
    if (!g.time || (g.time.getFullYear()<2000) ||
       (g.time.getFullYear()>2200)) {
      // GPS receiver's time not set
      console.log("GPS real time clock is not set, use GPS info and get a fix");
    } else {
      // We have a GPS time. Set time
     var t = g.time.getTime()/1000;
      setTime(t);
      console.log("time set to " + t);
    }
  });  // this is the end of the GPS callback funtion

  Bangle.setGPSPower(1,"time");   //  power the GPS on, when GPS responds, callback is called
}, 60*60*1000);     //  run this once per hour, 3600 seconds in milli-seconds

But I can see that you added this code to your app ages ago.
Really interesting discussion regarding the accuracy of the RTC chip verses the one in the GPS.

Ok - had a few issues with that code, it did not work for a couple of icons. Issues now fixed. I have done a pull request.

Just updated to fw 2.10.114 and can see that @Gordon has changed the system font for the menu's. Here is what the settings app now looks like. Much easier to read. My preference would be still be the 6x8 x2, but the new solution also works for me.

The standard 6x8 scaled at 2 is actually quite an acceptable font for the menus in settings.

1. thick enough to read
   
2. thin enough not to look blotchy
   
3. produces 6 lines instead of 8 per screen
   
4. the font is already available as a system font
   
5. nicer than the other samples that were demo'd above (in my opinion)
   
6. its pretty close to some of the X11 fonts.
   
   

If this is what had been used on the B2 when I had first seen it I dont think it I would have noticed any usuability issues.

To make some of the menu text shorter,

App/Widget Settings => App Settings

I forgot Sinclair ZX80 was monochrome.
I bitterly could never afford such a machine, and feel loss that I never had the chance of a mispent youth. Kudos to any brave kid who saved up spent the money and soldered their machine. The first real computer I owned was a 1990 Dell 286, could at a pinch run Windows 3.

https://en.wikipedia.org/wiki/ZX80?wprov=sfla1

@Robin - you did have it bad.
Never mind, I remember cross compiling for a 6801 board on an IBM 370, and a v52 terminal.

I've managed to get toucher working on a Bangle 2.

Made a few tweaks and changes and thought I would run them past people in this forum before doing a pull request. The main thing is I have removed the version information on a 2nd line. Basically this makes the UI less cluttered.

On B2 the icons look really nice now in toucher.
Let me know what you think.

To give it a try install toucher through the IDE. Then upload the file below as toucher.app.js.
Then use File Manager to load toucher.app.js

Interesting site. I remember messing on with X11 font.
Is there a way of trying these in your image converter ?

Making the Menu Size / Font configurable is probaly the way to go. Each to his own poison.

Black, White, Red, Green, Blue, Cyan, Yellow, Magenta
https://en.wikipedia.org/wiki/List_of_monochrome_and_RGB_color_formats#3-bit_RGB

This is what we used to use in the late 1970s when those first computers first came out.
Try telling that to the kids of today and they wont beleive ya.

No 3 from the top - but with a font height of 20 so that you get 6 options per screen. As shown it looks like you would still get 8 options which is too small. However it looks more readable than the existing font for showMenu();

Its not easy finding a font that works and looks nice. I have just spent a couple of hours trying out various ones on fonts.google.com. The problem is that 300 fonts come out too spidery and weedy, where as regular 400 fonts come out thick and blotchy.

I have tried out most of these but they all look very similiar at 400.
https://www.justinmind.com/blog/best-font-mobile-app-design/

Attached is my hackemnu.js code.
test code I am using once that is installed is:

var hackmenu = eval(require("Storage").read("hackmenu.js"));
var m = {
  "" : { "title" : "Settings" },
  "Make Connectable" : function() { LED1.set(); },
  "App/Widget Settings" : function() { LED1.reset(); },
  "BLE" : function() { LED1.set(); },
  "Debug Info" : function() { LED1.reset(); },
  "Beep" : function() { LED1.set(); },
  "Vibration" : function() { LED1.reset(); },
  "Quiet Mode" : function() { LED1.reset(); },
  "Locale" : function() { LED1.reset(); },
  "Exit" : function() { }, // nothing for now
};
hackmenu(m);

The hope would be that the app codebase improves to support both devices. I think a lot of the discussioons apply to both B1 and B2. I would not want to have to flip between a B1 , B2 and a Bangle both forum. And then what about when Bangle 3 comes along :).

Definitely - like it. Maybe a configurable grid with images that would be scaled to fit a 2x2, 3x3 or a 4x4 grid. 4x4 might be pushing it a bit.

BTW - I have a modifued toucher working on a B2 ! its very nice. Need to test on a B1 which I dont have with me right now as I am on a train.

The above idea works for me.

I like the new settings version .30 and the new Customise Theme.
I have my highlight set to white on black and its much easier to read - though I still find the font excruciatingly small and difficult to read.

There are various Apps that are very slow when working out the number of records etc (eg GPS Logger, Accellog) as they have to read through the entire file.

Wondering if one of the kickstarter improvements could be a ability to read segments of a StorageFile by offset. Currently to get the last line or the record count of a large file you have to read the whole file back. The StorageFile system would have to keep track of file length somehow to do this.

Thanks Gordon, no massive rush.

The previous version of findphone printed a newline but it didn't make any difference.

When I encountered this issue I tested against the settings app and found the same issue and that the presence of the "\n" made no difference in findphone and was not present in the settings app version.

I only have 3 clocks installed, wave s7clck and pastel. Non have console.log in them.

I wonder if anyone else has noticed this ?

@user107850

OK, I have renamed time_t to sc_time_t so it doesn't conflict with linux time_t. Try pulling.

Thanks - can confirm compiles without error now.

Thanks to work on the harness - its great to have 2 algorithms to compare.
Are you going to write a brief for your students ?
Pointers I would give them would be:

1. Use 1 .c file and one .h file and a simple make file.
   
2. Do not over engineer the stages using fancy queues, buffers, message passing etc
   
3. Use the int_sqrt() function as that is what the embedded devices will mostly have to use.
   
4. avoid large arrays.
   
5. Add lots of comments to explain how the code works.
   
6. If you use some statistical symbol / concept make sure you reference to a site that explains it clearly - ie don't just say /* using black's algorithm */
   
7. Avoid fancy Computer Science terminology like 'single directed graph - instead use plain english like - tree structure'
   
8. Testing should prove good accuracy when walking for long periods (not just short bursts) and non stepping use cases such as sleeping. driving, sitting at a desk, watching TV - for an hour of such activities we want sub 10 steps to be logged.
   
   

As a general aim I think we should try to have something as good as an average fitness tracker.

Agree. I would like to get to a consistant 2% within control on walking and 200 steps across the day on non stepping activities. 200/10000 steps is 2%. Still a long way from the later.

I have a 1 hour train journey tomorrow so going to try an capture a log just sitting on a train - 0 steps.

In the end it depends on the use case.

I've had a few fitness bands / watches over the years. The most common use case for me and most people I know who use them is just monitoring that you are getting your 10,000 steps per day in. For training / running and anything that needs split second and distance accuracy you have to use GPS - I would see the Bangle Run app fulfilling that niche - if its made to work well.

3hz is within slow running pace, approx 12 mins per mile.
See the table I produced for this. http://forum.espruino.com/conversations/359542/?offset=75

There is a fundemental limit on what we can do with the Bangle with a 12.5hz sample time and 3hz low pass filter. The effectiveness of the LPF is constrained by the sampling rate.

@johan_m_o - thats brilliant. I've found them to be quite accurate for walking and non walking
activities. EG - my Fitbit Charge HR records about 20 steps overnight during a sleep period, non while driving and about 2000 steps per mile which is spot on for my stride length. So I trust them as something to measure against. At this stage a few steps over/under on the control is neither here nore there in terms of testing. We are sill a long way off that kind of accuracy.

The method I use is:

1. note START step count on control device.
   
2. not START step count on Bangle
   
3. start the logging
   
4. start walking - I usually walk the dog for a couple of miles
   
5. stop walking, stop the logger
   
6. when stopped not END counts of bangle / control device
   
   

When you upload the file looks like we have agree'd a filename format like:

1234_Name_Activity_bYYYY.csv
Where 1234 is the control step count.
bYYYY is the Bangle step count (2.10.27 firmware or greater)
Name - your name
Activity - walking, sitting, sleeping, driving, tv-sitting, typing

I install the active Pedometer Widget and set the display to large digits so I can see what the step count is.

I'd be happy to checkin the logs into the various repo's if you are not used to git etc.

@Gordon - any idea's on this one. When you go into settings / app / gadgetbridge / find phone - you have to select ON twice before the phone starts ringing. Its as if the first event is somehow lost.

Did you try checking what Gadgetbridge sends for an actual call received on the phone?

I did. I get 2 call events logged, but both have a blank cmd string.
I assume the first is the incoming call notification and the 2nd is the missed call ?

http://forum.espruino.com/comments/16201584/

Could it be a combination of the phone and Gadgetbrige ?