The Tibblog

We just had our Year-End party (we go by the Chinese calendar),...

Thu, 11 Feb 2010 04:07:00 -0500

We just had our Year-End party (we go by the Chinese calendar), and it was a blast! Here are just a few photos of the Taipei team having fun. :) HAPPY CHINESE NEW YEAR!

Our new DS10xx family of BASIC-programmable industrial...

Tue, 02 Feb 2010 06:50:00 -0500

Our new DS10xx family of BASIC-programmable industrial controllers is now out!

Our Designer, Leo, has recently visited us in Taipei. Here is a...

Tue, 02 Feb 2010 04:55:00 -0500

Our Designer, Leo, has recently visited us in Taipei. Here is a photo of him and his computers/monitors, one of which shows his top secret project E** (which, regrettably, we had to pixelate). Tomorrow Leo is off to get some much needed R&R in Thailand, so we all wish him a good trip.

AggreGate in Russian, Chinese, and More! And we don’t mean some...

Fri, 22 Jan 2010 05:22:00 -0500

Menu and System Tree in Russian

System Log

AggreGate in Russian, Chinese, and More!

And we don’t mean some “About AggreGate” document, either! It’s the actual software. The UI itself is being localized these days. Look at this sneak preview of our Russian AggreGate.

Security, Security, Security...

Mon, 28 Dec 2009 22:21:00 -0500

Hello there folks,

it is a holiday season all around the world but we are not slowing down. It looks like Q1 will bring our BASIC-programmable products several important security-related enhancements. Here is just some stuff we are working on: additional security modes for Wi-Fi (cause WEP is cr**), SSL support, and protection of customer’s Tibbo BASIC application loaded into the module.

Stay tuned…

SSI does MAXIM RTC

Fri, 11 Dec 2009 00:05:00 -0500

Yet another application for our SSI object. We have successfully used our SSI to interface with the DS3234 “extremely accurate” SPI RTC.

Here is the code:

'------------------------------------------------------------------
io.num=SPI_CS
io.state = HIGH
io.enabled=YES

io.num=SPI_CLK
io.state = HIGH
io.enabled=YES

io.num=SPI_MOSI
io.state = HIGH
io.enabled=YES

'setup SSI channel
ssi.channel=1
ssi.mode=PL_SSI_MODE_1
ssi.clkmap=SPI_CLK
ssi.dimap=SPI_MISO
ssi.domap=SPI_MOSI
ssi.zmode=PL_SSI_ZMODE_ALWAYS_ENABLED
ssi.direction=PL_SSI_DIRECTION_LEFT
ssi.baudrate=1
ssi.enabled=YES

'------------------------------------------------------------------
function rtc_read(address as byte)as byte
  io.lineset(SPI_CS,LOW)
  ssi.channel=SSI_CHANNEL_RTC
  rtc_read=ssi.value(address*256,16)
  io.lineset(SPI_CS,HIGH)
end function

SSI does Microchip Port Expanders!

Thu, 10 Dec 2009 23:50:00 -0500

G’day folks,

it hasn’t even been a week since we published our SSI object, and we already found an application for it. We used the SSI to control Microchip’s MCP23017 16-bit port expander.

For those of you who never tested this, Microchip port expanders are I2C or SPI ICs that give you 8 or 16 I/O lines. Add something like this to our (relatively low-pin count EM1206) and you get 16 additional I/O lines for <1US$!

Without further ado, here is the code that setups an I2C channel, configures all I/O lines of the MCP23017 as outputs, and writes &hAA55 to the output ports:

'------------------------------------------------------------------
  'configure I2C clock as output
  io.num=I2C_SCL 'you need to declare I2C_SCL, I2C_SDA
  io.state=HIGH
  io.enabled=YES

  'setup SSI channel
  ssi.channel=0
  ssi.mode=PL_SSI_MODE_3
  ssi.clkmap=I2C_SCL
  ssi.dimap=I2C_SDA
  ssi.domap=I2C_SDA
  ssi.zmode=PL_SSI_ZMODE_ENABLED_ON_ZERO
  ssi.direction=PL_SSI_DIRECTION_LEFT
  ssi.baudrate=40 'this I2C device is relatively slow
  ssi.enabled=YES

  'setup all I/O lines of the MCP23017 as outputs (see Microchip's documentation)
  i2c_start() 
  ssi.str(chr(&h40)+chr(&h00)+chr(&h00)+chr(&h00),1)
  i2c_stop() 

  'output &hAA55 (see Microchip's documentation)
  i2c_start() 
  ssi.str(chr(&h40)+chr(&h12)+chr(&hAA)+chr(&h55),1)
  i2c_stop() 

'------------------------------------------------------------------
sub i2c_start
  io.lineset(I2C_SCL,HIGH)
  io.num=I2C_SDA 'set SDA to HIGH first so we can have HIGH->LOW transition
  io.state=LOW 'we are manipulating data line through the OE register
  io.enabled=NO
  io.enabled=YES 'this will set the data output to LOW	
end sub

'----------------------------------------------------------------------------
sub i2c_stop
  io.lineset(I2C_SCL,LOW) 'this will remove the ack bit
  io.num=I2C_SDA 'set SDA to LOW first so we can have LOW->HIGH transition			
  io.state=LOW 'we are manipulating data line through the OE register
  io.enabled=YES
  io.lineset(I2C_SCL,HIGH)
  io.enabled=NO 'this will set the data output to HIGH
end sub

SSI Object is Ready!

Tue, 08 Dec 2009 05:04:00 -0500

Busy day here at Tibbo folks. We’ve just assembled TIDE V2.20.16 (see experimental build at the bottom of the page). This includes the EM1000 firmware and platform files supporting our new Synchronous Serial Interface object.

Do SPI, I2C, as well as custom clock/data interfaces using this new object. You get four independent channels, programmable bitrate, endianness, clock phases, etc. you have a total flexibility needed to work with RTCs, EEPROMs, DACs, ADCs, and other serial ICs.

With speed reaching 1.2MBit/sec, the SSI object gives you performance that far exceeds anything you were able to do with direct bit-banging using IO object. Any I/O pins of your module can be used as CLOCK, DO (MOSI), or DI (MISO) lines. Moreover, the DO and DI lines can be mapped onto a single physical I/O line, which is necessary for I2C.

Stay tuned for actual code examples. We’ve got some serial ICs we are going to try…

Note: this excellent Wikipedia article explains SPI modes 0-3.

TIDE = Speed (re: Modbus)

Tue, 08 Dec 2009 05:01:00 -0500

Case in question: our own in-house test of an energy meter with RS485 interface running MODBUS protocol.

One of our potential Customers wanted to us to create a small Tibbo BASIC app that would query the meter and show the data on a simple HTML page. Two hours later (spent mostly on connecting and arranging things) we had a working application prototype. The source code is published here.

We didn’t go as far as to format the data (in KWatts, etc.). This demo displays raw contents of MODBUS registers.

Now that we’ve got Wi-Fi working (and ZigBee coming) it is possible to use our modules in power metering applications — a huge market worth gazillions*.

* Based on numerous reports by multiple totally trustworthy sources, or so we remember.

Two members of our Russian team — Dmitry (left) and Victor...

Wed, 02 Dec 2009 21:32:00 -0500

Two members of our Russian team — Dmitry (left) and Victor — have recently visited us. Victor is our AggreGate Project Manager, and Dmitry is our “go to” guy for all IT and web- related technologies. Hope you liked it here, guys. Many happy returns.

Note: The Christmas tree is not visiting, we keep it here in Taipei.

These are the screenshots of AggreGate status panels for our...

Wed, 02 Dec 2009 21:11:00 -0500

These are the screenshots of AggreGate status panels for our upcoming DS1004 and DS1005 industrial I/O devices (not yet officially released).

The DS1004 is our new industrial controller with analog I/O. The DS1005 is the same, but with digital I/O (opto-couplers and relays). Both devices are BASIC-programmable and can be tested using free Tibbo BASIC applications posted on our website (here: the one for the DS1004, and another one for the DS1005).

These applications connect to the AggreGate server and let you control the DS1004 and DS1005 through these status panels. I know you all want to ask this, and here is the answer: yes, everything is customizable. Take it and change it in any way you want.

We are about to start handing out CDs containing the trial...

Wed, 02 Dec 2009 20:55:00 -0500

We are about to start handing out CDs containing the trial version of our AggreGate (TM) software.

For those of you who never heard of AggreGate: this is a complete device management platform that can be used for access control, time/attendance, industrial automation, and many other applications.

For those of you who tested AggreGate before: a lot have been added since the last time you’ve checked. Take another peek.

Tibbo was recently visited by our good Indian partner, Raja (left), the head of Theta Sales and...

Fri, 27 Nov 2009 03:12:00 -0500

Tibbo was recently visited by our good Indian partner, Raja (left), the head of Theta Sales and Service. He spent some time with us, getting the latest news and hands-on experience with our new Tibbo BASIC products and AggreGate. Always nice to have such visitors!

Check out our new packaging. This box is about to start...

Sun, 22 Nov 2009 13:22:00 -0500

Check out our new packaging. This box is about to start delivering our development kits to interested parties around the world. Yam!

Project E** update

Sun, 22 Nov 2009 13:17:00 -0500

Heya Folks,

just wanted to let all you know that we have achieved a major milestone in the development of the new embedded module code-named E**.

Those of you in the know already, err, know that E** is the ******** and the ******** module to ever come out of our lab.

Stay tuned for more bits and pieces on this important development.

Bit-banging to get easier/faster

Sun, 22 Nov 2009 11:20:00 -0500

Here is a quick update re what’s been going on at our lab.

We are working (and will soon publish) a new object called SSI. That stands for “Synchronous Serial Interface”.

Pretty soon you will be able to do speedy data exchange with externally connected I2C, SPI, etc. devices such as EEPROMS, RTCs, ADC/DACs, etc.etc. Manyfold increase in tx/rx speeds compared with bit banging based on T-BASIC code. We have this working at 1.5MBit/second*.

You will get 4 independent channels with multiple programmable options. Each channel can use any I/O lines you wish to use. It is all configurable.

This is coming to your EM1000, EM1202, and EM1206 devices pretty soon. Stay tuned for updates.

*not a sustained throughput metric

A smarter way to format your flash disk (for debugging)

Sun, 22 Nov 2009 11:04:00 -0500

How much flash space do you allocate for your flash disk? Normal answer is “the rest of it”. That is, take all space not occupied by your code:

fd.format(fd.availableflashspace,32)

This works, except when you are actively developing/debugging and your project keeps getting bigger, thus encroaching onto the disk space and forcing you to format your disk all the time.

Here is an elegant solution: leave some spare space for the project to grow:

 fd.format(fd.availableflashspace/2,32)

This way, formatting will happen a lot less. You can also improve it this way:

 if sys.runmode=DEBUG then fd.format(fd.availableflashspace/2,32) 

  else fd.format(fd.availableflashspace,32) 

end if

Smart, huh?