jprodgers / lolshield Goto Github PK

Using a single-buffered grayscale display, I can't tell a difference between 
"uint8_t pixels[SHADES][24]" and "uint8_t pixels[SHADES][48]" - and the 
subsequent commentary makes me suspect that 48 is simply 24 bytes per frame * 2 
frames.  As that means the leds structure takes up 768 bytes, the amount of ram 
available for user projects is limited unnecessarily.  Especially since that 
structure is allocated twice, once for the work buffer and once for the display 
buffer, but I haven't poked at it enough to be certain.

imac:bin pfriedel$ ./avr-objdump -t -j .bss OneTenClock.cpp.elf 

OneTenClock.cpp.elf:     file format elf32-avr

SYMBOL TABLE:
[...snip...]
0080027a g     O .bss   00000300 leds
[...snip...]

Versus:
imac:bin pfriedel$ ./avr-objdump -t -j .bss OneTenClock.cpp.elf 

OneTenClock.cpp.elf:     file format elf32-avr

SYMBOL TABLE:
[...snip...]
0080029e g     O .bss   00000180 leds
[...snip...]

With a related decrease in the bss allocation:
imac:bin pfriedel$ ./avr-size OneTenClock.cpp.elf 
   text    data     bss     dec     hex filename
  15208     376    1371   16955    423b OneTenClock.cpp.elf
imac:bin pfriedel$ ./avr-size OneTenClock.cpp.elf 
   text    data     bss     dec     hex filename
  16626     376     987   17989    4645 OneTenClock.cpp.elf

I'm not quite proficient enough with C header programming to make this do the 
right thing when it's not being used in a double buffered application, but it 
seems like it should be reasonably straightforward.  Ultimately it seems like 
it should allocate 48 when double buffered grayscale, 24 when single buffered 
grayscale, and (???) when double-buffered B&W and single buffered B&W. To be 
honest, the interrupt code makes my eyes cross.

There's also a spurious Serial.end() in Charliplexing.cpp that causes the 
serial library to get pulled in unexpectedly.  It wasn't until I started 
looking at an object dump trying to figure out where all my memory was going 
that I saw that Serial was getting pulled in that I took a closer look at my 
libraries.

Or I'm just missing a subtlety that only comes up when you're using all 136 
LEDs on an official LoL Shield that isn't apparently on my stripboard 110-LED 
project, in which case you can ignore this.

What steps will reproduce the problem?
1. Run BasicTest with lolshield 0.2 release, watch LEDs.
2. Run BasicTest with latest lolshield svn, watch LEDs.

What is the expected output? What do you see instead?
Expected: Steady LED lights when they are switched on, like they did in 
lolshield-0.2 release.
Instead: LEDs flicker annoyingly. They do not seem to be at 100% brightness, 
but that's a guess.

What steps will reproduce the problem?
1. Figure won't display to bottom and right edge of display
2. Hardcoded x and y limits of 13 and 8 are incorrect for 14 x 9 shield
3. Easily changed - or have an Init routine with width and height

What is the expected output? What do you see instead?
Output near edge of display doesn't work

What version of the product are you using? On what operating system?
v1.0 arduino ide

Please provide any additional information below.
Font library also appears to have incorrect constants - maybe initially written 
for a smaller LOL shield.  Font library doesn't seem to work however (test 
program gives blank display)

I updated the Mega 1280/2560 support from Issue #16:
* to apply after the changes in Issue #32,
* to reduce code size (by internally remapping a few pins),
* to make it glitch-free.

Note that the original Mega (1280), like many boards, does not have a buffer on 
the pin 13 LED, which screws up the shield display pretty badly.

What steps will reproduce the problem?
1. Plug in Ardunio
2. Open Test sketch
3. Try Uploading

What is the expected output? What do you see instead?
Nothing happens because of error uploading

What version of the product are you using? On what operating system?
using Arduino Uno, with Arduino 0022 IDE. LoL SHield v 1.5 and Library v23

Please provide any additional information below.

Just installed LoL Shield SHD library, loaded Basic_Test and then ran into the 
following compilation error.
Does anyone have any pointers on what to do?

Arduino: 1.6.4 (Mac OS X), Board: "Arduino Uno"

In file included from Basic_Test.pde:31:0:
Basic_Test:49: error: variable 'BitMap' must be const in order to be put into 
read-only section by means of '__attribute__((progmem))'
variable 'BitMap' must be const in order to be put into read-only section by 
means of '__attribute__((progmem))'

  This report would have more information with
  "Show verbose output during compilation"
  enabled in File > Preferences.

I did some very minor grayscale enhancements to reap low hanging fruits:

- include Matt's "Fade" example
- remove BIT_DEPTH since it does not seem to be used anywhere
- rename COLORS to SHADES since there really is only one color ;)
- set SHADES to "5" (resulting in 4 shades plus black, i.e. OFF); this 
   seems to be the maximum bearable amount of shades the device can produce
   as of now.

Opinions? Rejections?

Library keywords don't appear highlighted in the Arduino IDE.  Adding a
keywords.txt file should correct this.

From: http://arduino.cc/en/Hacking/LibraryTutorial

If you tried the new sketch, you probably noticed that nothing from our
library was recognized by the environment and highlighted in color.
Unfortunately, the Arduino software can't automatically figure out what
you've define in your library (though it would be a nice feature to have),
so you have to give it a little help. To do this, create a file called
keywords.txt in the Morse directory. It should look like this:

Morse   KEYWORD1
dash    KEYWORD2
dot KEYWORD2


Each line has the name of the keyword, followed by a tab (not spaces),
followed by the kind of keyword. Classes should be KEYWORD1 and are colored
orange; functions should be KEYWORD2 and will be brown. You'll have to
restart the Arduino environment to get it to recognize the new keywords. 

What steps will reproduce the problem?

1. Pass an argument to Set() that is not zero and not one, like "Set(x, y, 
100)". 

What is the expected output? What do you see instead?

I'd expect a non-zero value to cause the LED to be lit. Standard boolean 
behavior is zero==false (or off) and non-zero==true (or on). The way it is now, 
a non-one value will cause the LED to be off. 

Please provide any additional information below.

You can "fix" this by changing the "if (c == 1)" in LedSign::Set() to "if (c != 
0)"

From this:
void LedSign::Set(uint8_t x, uint8_t y, uint8_t c)
{   
    uint8_t pin_low  = ledMap[x*2+y*28+1];
    uint8_t pin_high = ledMap[x*2+y*28+0];
    if (c == 1) {
        workBuffer[(pin_low-2)*2 + (pin_high / 8)] |=  _BV(pin_high & 0x07);   
    } 
    else {
        workBuffer[(pin_low-2)*2 + (pin_high / 8)] &= ~_BV(pin_high & 0x07);   
    }
}

To this: 

void LedSign::Set(uint8_t x, uint8_t y, uint8_t c)
{   
    uint8_t pin_low  = ledMap[x*2+y*28+1];
    uint8_t pin_high = ledMap[x*2+y*28+0];
    if (c != 0) {
        workBuffer[(pin_low-2)*2 + (pin_high / 8)] |=  _BV(pin_high & 0x07);   
    } 
    else {
        workBuffer[(pin_low-2)*2 + (pin_high / 8)] &= ~_BV(pin_high & 0x07);   
    }
}

Anyone know of a way o switch between "LedSign::Init();" and 
"LedSign::Init(DOUBLE_BUFFER | GRAYSCALE);" within a sketch?

I am building Christmas tree ornaments based on the LoL shield and need to to 
go from text to graphics. 

Build in object management would be great for game development. Some of the
existing memory issues will need to be resolved first.

Being able to simply move an object, and have it detect walls would be
nice. Also being able to scroll anything arbitrary like text or images
would be made much easier.

The game examples all seem to require an additional potentiometer and/or
button, but it isn't clear how they should be connected.  A set of
directions would be excellent!

What steps will reproduce the problem?
1. Using Arduino IDE 022
2. Load Tetris example
3. Try to compile it.

What is the expected output? What do you see instead?
I expect it to compile and indicate how much ram it'll use. Instead I get 
warnings about C++ incompatibilities.

What version of the product are you using? On what operating system?
LoL shield 1.5, Library version .2beta, Arduino IDE 022, Mac OS X 10.6.6

Please provide any additional information below.
Comment your source code.

This version of charlieplexing.cpp uses analog pins instead of digital 8-13 to 
allow the use of ethernet shield.

At all times two LEDs in the upper right corner are lit at all times. It
seems to go across all versions of Arduino that I've tried. It does not
happen if the refresh rate is slowed down to light only individual LEDs.
Only when it is refreshing at a decent rate.

This patch reduces the code size and improves the CPU performance of 
Charliplexing.cpp by:
* removing zeroed data in videoPage
* precalculating information for LedSign::Set()
* removing unneeded register writes in TIMER2_OVF_vect()
* using a roving pointer to access the display buffer rather than recalculating 
the array index each time
* eliminating the extra blank interval (1/8 display cycles)
* eliminating an extra display “cycle” (which could also display garbage)

On my test platform, this:
* reduced code size by 178 bytes
* increased the FadeTest max FPS from 55 to 71
* increased brightness (not sure how much offhand)

The current text sketch works, but it is very limited due to memory 
constraints. Current letters are stored as individual points that are moved 
around. A single letter can take up 8-34 bytes each! That eats up the available 
2k of RAM very quickly.

Just putting all the fonts in progmem would help. The fonts will need to be 
stored in bit arrays. They really shouldn't be any larger than 5-9 bytes per 
letter, one for each vertical line. A full ASCI character set shouldn't be more 
than 2k stored in ROM, and that's a full board height font. Though adding a 
single byte at the beginning of each character could allow for each letter to 
have it's own width, and would allow for some kerning.

It would probably be helpful to do the object tracking library first, as that 
would make the font one much easier. Storing and reading to ROM is fairly 
straightforward.

It would be cool to add grayscale support to the display.

It would be neat to be able to be able to programatically adjust the
brightness of the display, to be able to fade images in and out. This might
be possible to do by varying the duty cycle of the display, perhaps by
using a second timer to turn the LED pins off early.

Have you thought about the future of this project?
Thanks!

What steps will reproduce the problem?
1. set lolshield with "LedSign::Init(GRAYSCALE);"
2. Send text to LoLshield via Myfont::Banner(tLen,Text);
3. LedSign::SetBrightness(); Is ignored due to draw code in Myfont.cpp

What is the expected output? What do you see instead?
Expected: Variable led shades due to LedSign::SetBrightness(); initialization 
within code. Instead: Myfont:Banner uses the below to draw the text. "tmps" 
sets the brightness when the (GRAYSCALE) option is used.
[code] LedSign::Set(xval+i, j+1, tmps); /*write it*/ [/code]

What version of the product are you using? On what operating system?
LolShield v1.5, v0.23 Library

The Lol-Shield library is not compatible with Arduino Atmega 2560 out of the 
box. All basic sketches fail, no LED lights. 
I think this fails because there is no timer enabling routine in 
"Charliplexing.cpp" for the Atmega 2560. The timer codes start at line 154 but 
it lacks one for the 2560.

The library currently uses a namespace-based approach to segmenting itself,
where as the standard for Arduino libraries appears to be a class-based
approach.  The current implementation works perfectly fine, however it
might be preferable to encapsulate the library into a class for consistency
with the broader Arduino community.

... and along with it a lot of the documentation for this project.

I am trying to get one of these working with a new Uno board.

Is there another source for all the files and documents? The LoLShield 
V0.2.beta.zip readme refers to -libs and -font folders(?) which are not in the 
ZIP file. SparkFun's links at http://www.sparkfun.com/products/9796 just go 
here and to the dead web site.

TIA
Bruce

LoLcomm is an example application which enables you to set individual LEDS, or 
set the whole shield via the serial port. You can also pulse the shield, or 
pulse individual LEDs (using grayscale).

I also provide a Linux command line tool for remote-controlling the LED shield.

This enables you to play around with LED patters and even to script your shield.

Have fun!

Available since SVN revision 25 or later.

When running in 'black and white' (i.e. non-GRAYSCALE) mode, there's no need 
for Set to run through the for loops or increment pointers.

The attached patch provides a noticeable improvement when Set() is being called 
as rapidly as possible.

What steps will reproduce the problem?
1. Just run the fonntest example
2.
3.

What is the expected output? What do you see instead?
I'd expect to see the fonts displayed as per a youtube video I saw
What I actually see is nothing

What version of the product are you using? On what operating system?
Charliplexing.cpp V0.9
Font.cpp V 0.0
fonttest.pde V1.0

Please provide any additional information below.

I had to change WProgram.h to Arduino.h
Basic_Test passes OK, so I guess the shield is wired up correctly.
Charli_heart appears to start OK but soon degenerates into a mess of LEDs
Grayscale produces wavy diagonal lines
invader, life, LoLShield_breath, pong2 and myfont all work OK.
tetris won't compile

What steps will reproduce the problem?
1. arduino 022
2. Mega 2560
3. LOL shield

What is the expected output? What do you see instead?
Nothing the charlieplex library is incompatible with Mega 1280/2560


What version of the product are you using? On what operating system?
see above

Please provide any additional information below.
http://arduino.cc/forum/index.php/topic,76626.0.html thread on this.

What steps will reproduce the problem?
1. Build LOLshield.
2. Put it on a Leonardo.
3. Install LOLshield library code.
4. Connect LOLshield+Leonardo combi to computer.
5. Load Basic_Test example sketch.

What is the expected output? What do you see instead?
All dark LEDs. Some random LEDs light when you upload other,
non-LOLshield sketches.

What version of the product are you using? On what operating system?
LOLshield library v0.23
LOLshield 1.5
Arduino IDE 1.0.1
Arduino Leonardo
Mac OS X 1.6 (Snow Leopard)

Please provide any additional information below.
My LOLshield is not bad. I performed the above steps, substituting an Uno for 
the Leonardo and everything worked as expected.

When using the Font::Draw command to draw text that should be drawn through the 
14th row, the row is blank instead of lighting up. I would expect the text to 
be drawn through this row.

This is an experiment strobing the display in 12 cycles rather than 24, which 
(in theory) doubles the light output.  I haven't studied it scientifically, but 
the output looks about as even as before.

As a bonus, it greatly simplifies the logic in the ISR, and even makes the 
compiled code smaller.

Currently the library allows for up to 11 LEDs to be lit at once.
Unfortunately, the Arduino doesn't supply enough current to light all of
them brightly, so you get a dimming across the bus. However, decreasing the
number of LEDs lit at once decreases the frame rate.

It needs to be determined at what point the number of LEDs lit, and the
frame rate is both uniform and acceptable for the LoL Shield.

What steps will reproduce the problem?

When I load Charlieplexing.h and Wire.h, and then call the req'd 
"LedSign::Init()" in the void.setup(), it seems the sketch crashes.

In the code example below, when the LedSign::Init is commented out, The Serial 
Monitor shows:

init           (command before the fouling Init)
made it        (command following the fouling - but commented out- Init)
get time       (the proper operations in the loop())
...
...  and works properly (it won;t show every second, as there are a number of 
debugging delay() statements)



When LedSign::Init() is active,  all I get in the serial monitor is:

init                        command before the fouling Init
                              ...
                              (chirping crickets)
                              ...



1. any arduino I have (Mega, Duemillablahblahblah, and ProMini5/16
2. Macetech Chronodot
3. charlieplex network w/ 72 LEDS


What is the expected output? What do you see instead?
i am expecting to see lights.  the sketch seems to freeze once either: 
Charliplexing or Wire gets init'ed/started.
see below..

What version of the product are you using? On what operating system?
Win7
arduino IDE 1.0 & 1.01


Please provide any additional information below.
Snippet of code:

/*
DS3231_test.pde
Eric Ayars
4/11

Test/demo of read routines for a DS3231 RTC.

Turn on the serial monitor after loading this to check if things are
working as they should.

*/

#include <DS3231.h>
#include <Wire.h>
#include <Charliplexing.h>

DS3231 Clock;
bool Century=false;
bool h12;
bool PM;
byte ADay, AHour, AMinute, ASecond, ABits;
bool ADy, A12h, Apm;

byte year, month, date, DoW, hour, minute, second;

void setup() {

    // Start the serial interface
    Serial.begin(9600);

delay(1000);

    Serial.println(" init ");            //  <-- This prints, ++
delay(1000);

 //  LedSign::Init();                   //  <-- when this is enabled

delay(1000);
    Serial.println("made it");      //  <-- this doesn't print ++
delay(1000);                             //  <-- Nothing after the LedSign 
command executes

// Start the I2C interface
    Wire.begin();

}

void loop() {
    // send what's going on to the serial monitor.
    // Start with the year



    Serial.print("get time"); delay(1000);
    Serial.print('\n');
    delay(1000);

    // Display the time once more as a test of the getTime() function
    Clock.getTime(year, month, date, DoW, hour, minute, second);
    Serial.print(year, DEC);
    Serial.print(month, DEC);
    Serial.print(date, DEC);
    Serial.print(DoW, DEC);
    Serial.print(hour, DEC);
    Serial.print(minute, DEC);
    Serial.println(second, DEC);
}



please help!

uncommented #define MEASURE_ISR_TIME since per default this does not make sense 
and might even cause erratic errors
deleted Serial.end() in SetBrightness() which is plain wrong

What steps will reproduce the problem?
1. Start up Arduino dev GUI

2. In the GUI load the "LoLShield BasicTest" example:
File -> Examples -> LoLShield -> LoLShield_BasicTest

3. Press the compile button


What is the expected output? What do you see instead?

Expected: compilation works.
Instead:
libraries/LOLshield/Charliplexing.cpp:32:22: fatal error: WProgram.h: No such 
file or directory


What version of the product are you using? On what operating system?

- Latest Arduino GIT (git://github.com/arduino/Arduino.git), rev. 
11281ec77ece61aeab64d047ed899d47221dd4a7, 2012-01-05
- latest lolshield SVN, rev 19


Please provide any additional information below.

This is a known bug w/ lolshield code and Arduino 1.0 or newer. Seemingly 
Jimmie keeps fixing this every time he encounters the problem in one of his 
workshops so I thought patching the actual source in the repository would be 
more convenient ;)

I attached a patch file which sets things straight for Arduino 1.0 while still 
maintaining compatibility to older releases. 

This patch has been tested on ARDUINO 1.0 ONLY!

How to apply the patch:
- check out the lolshield SVN trunk
- put the patch file arduino-1.0-include.patch into the checkout directory
- run 
    patch -p0 --dry-run < arduino-1.0-include.patch

The patch in Issue #32 isn't quite as tolerant of buggy code sending too large 
a value to Set().  This exposed a bug in the Breathe example.  Patch included.

Ahm, this is actually an *Enhancement*, not a *Defect*. 

Unfortunately I'm unable to change the issue type, guess it's because I'm not a 
committer. Or maybe I just cannot find the button that does this.

-------

Add grayscale test to LoLShield_BasicTest.

    This patch adds grayscale to the basic LoLShield test.
    The test will run three times, increasing the brightness
    of the LEDs in each run, then start all over again.
    The frame delay has been reduced to a third of
    its initial value so the overall runtime of the test
    stays the same.

How to apply the patch:

- check out the lolshield SVN trunk
- put the patch file add-grayscale-to-basic-test.patch into the checkout 
directory
- run 
    patch -p0 < add-grayscale-to-basic-test.patch

I just tried enabling DOUBLE_BUFFER for the examples. The fonttest example 
doesn't show anything at all and the basic-example only lights up the top-left 
LED.

Is this intentional? Does DOUBLE_BUFFER needs special handling?

Arduino 1.0 will introduce some changes. Many won't interference with lolshield 
in any way, but one thing will:
WProgram.h got renamed to Arduino.h. In beta3 of Arduino 1.0, this makes 
Sketches (and one example) using your library fail to compile.