Raspberry Pi

[Jeremy Harris]

Start a thread and see what happens.

There's already one here: http://www.picaxeforum.co.uk/showthread.php?18044-Maximite!

 

[Dippy]

Thanks Jeremy. So where does the "discouraged" bit come from?

 

[lbenson]

>"Belatedly discovered it needed to be CE certified before public sales" - oh dear oh dear. That sort of stuff should have been discussed while they were doing the Business Plan!

It’s not quite as simple as that. The business plan (as I understand it) was to sell 10,000 bare boards to hobbyists—CE certification no more needed than for the Beagleboard, which lacks it. Certification was to follow after a wringing-out period before mass production was started for the target educational users. Instead of 10,000, RS and Farnell, the distributors, got something on the order of 150,000-300,000 “expressions of interest” within the first few days.

The distributors were also to be the manufacturers going forward, and they decided that certification was needed before they started producing in that volume. Business plans can go awry if demand exceeds expectations by an order of magnitude--a happy problem to have.

 

[MFB]

Dippy, the "discouraged" bit comes from the following comment from Hippy regarding the Maximite thread mentioned above.

"It's an interesting product but ( to everyone ) let's not step over that fine line between can be used with to suggesting, even if not overtly stated, is better than or should be used instead of, or using the potential of integration as simply an enabler to advertise, promote or sell non Rev-Ed product which breaches forum policy".

And yet there seems no problem in mentioning the Pi because of its "potential integration" with the PICAXE. Case of double standards me thinks.

 

[manuka]

It all sounds very Sinclair

For those of a tender age, & thus perhaps ignorant of the "rush to market" implications,Dippy's comment from post #39 relates to Britain's Sir Clive Sinclair. FWIW I used to import his MAT121 transistors back in ~1964!

 

[mrburnette]

And yet there seems no problem in mentioning the Pi because of its "potential integration" with the PICAXE. Case of double standards me thinks.

Yes, a double standard, I agree. But, I understand the caution in Hippy"s note. I mentioned the other day that the Arduino forum was more granular and that there were some concepts that I wished RevEd would incorporate. "Community" consists of 7 sub-groups that cover the bulk of things we PICAXE forum members cannot overtly do here,

- Ray

 

[Dippy]

That's the fellah Stan.

If what Ibenson says is true then my eyebrows are up and my eyes are rolling. Even boy scouts know "Be Prepared".

MFB; Ah, I see. As that was one of Hippy's mega-sentences I probably fell asleep halfway through it

Personally, I can't see how either product should make any Rev-Ed sphincters twitch in fear....

 

[MFB]

No need for anyone to twitch because the PICAXE is great for its intended market. I have certainly not found anything better for quickly developing microcontroller based projects. However, should we feel constrained from mentioning more powerful alternatives when the application clearly exceeds its capabilities?

For example, posts often ask how to perform maths, graphics or mass storage functions and seem to receive far from convincing advice on how to do it with a PICAXE. At this point should we just suggest using a more powerful device like the Maximite or Pi microcontrollers with helpful details?

 

[hippy]

>"Belatedly discovered it needed to be CE certified before public sales" - oh dear oh dear. That sort of stuff should have been discussed while they were doing the Business Plan!

It’s not quite as simple as that. The business plan (as I understand it) was to sell 10,000 bare boards to hobbyists—CE certification no more needed than for the Beagleboard, which lacks it. Certification was to follow after a wringing-out period before mass production was started for the target educational users.

The way I see it is the Foundation thought it could get away with selling 10K without requiring CE certification by saying they were only selling to developers, but chose to sell them to the general public and then found they needed to be CE certified before such sales could go ahead.

 

[hippy]

Dippy, the "discouraged" bit comes from the following comment from Hippy regarding the Maximite thread mentioned above.

"It's an interesting product but ( to everyone ) let's not step over that fine line between can be used with to suggesting, even if not overtly stated, is better than or should be used instead of, or using the potential of integration as simply an enabler to advertise, promote or sell non Rev-Ed product which breaches forum policy".

And yet there seems no problem in mentioning the Pi because of its "potential integration" with the PICAXE. Case of double standards me thinks.

My comments were mainly in response to Don's post #23. I don't believe there is a double standard; more an issue of context and intent ...

No need for anyone to twitch because the PICAXE is great for its intended market. I have certainly not found anything better for quickly developing microcontroller based projects. However, should we feel constrained from mentioning more powerful alternatives when the application clearly exceeds its capabilities?

For example, posts often ask how to perform maths, graphics or mass storage functions and seem to receive far from convincing advice on how to do it with a PICAXE. At this point should we just suggest using a more powerful device like the Maximite or Pi microcontrollers with helpful details?

There are fine lines between saying there are alternatives, noting what they are, saying why they may be a better choice, promoting a product as an alternative, and encouraging people to go elsewhere.

People do have to feel constrained in how they respond to some degree, and we all have to accept that as a fact of life, just as people have to constrain themselves in not telling their friends they should have purchased a different drink when they go to the bar or telling them they should have chosen a different partner, different car, lived somewhere else, voted for another party, supported another team, adopted a different religion, etc.

Whilst not easy to define, and without absolutes, I think we would all recognise that there comes a point where it can go beyond the line of being acceptable.

 

[lbenson]

I suspect that Beagleboard sellers don't do anything to assure that their boards don't get into the hands of "the general public". With Raspberry Pi, the combination of price point, media attention, and long-stated intention to sell first-come, first-served to all comers assured that there would be many end users among the initial buyers.

I think the Raspberry Pi foundation underestimated the degree to which the tinkerer community would want a piece of Pi, but I think they vastly underestimated the general interest which the media attention churned up (see XBMC compatibility).

Anyway, I don't see that the Raspberry Pi replaces the picaxe for the things which the picaxe does best, and as attractive a price as the Pi has for a small barebones computer, it doesn't compare with an 08M2 if an 08M2 will handle your needs.

 

[Grogster]

The Pi is a very interesting little thing, and I probably will tinker with one - one day, when I have time, and that might not be for a while!

Personally, I have settled on the PICAXE, but there is still more for me to learn about programming and using them, so the thought of starting all over AGAIN with another language is enough to stop me seriously looking at using anything else.

Ardunio is another one, but again - I would have to learn it's programming language, so unless they can offer something very specific, that I HAVE to have in the design, then I'll be sticking with the PICAXE.

Besides which, the PICAXE makes the heart of a fantastic modular system, using 3rd party modules for specific things such as the ALFAT SD card module for licenced FAT32(with long-filenames) file access.

Need TV out?
Add a TellyMate or uVGA-II module.

Need sound?
Add a TDB380 MP3 or SOMO module.

Need a RTC?
DS1307 or 3232.

Need EEPROM?
24LC256 or 24LC512

The list goes on with the possible exception of HDMI, but who really needs that for embedded systems? (rhetorical!)

 

[Dippy]

"..Raspberry Pi foundation underestimated the degree to which the tinkerer community would want a piece of Pi"
- this is precisely why comments on CE related delays make my eyebrows rotate.

Do you honestly think they were aiming at 10K hobbyists or kids?
Have you thought of the Apps for a low-cost SBC?
And we're saying that the RPiPeople were unware of that possibility?



Farnell and RS have been splattered with RPi Headline promotions for months?
And then they are suddenly surprised that non-hobbyists are interested?

Not having something like this as "CE ready" at the outset is naive, bordering on the incompetent.
Let's face it, CE ain't no UL.

PS. I'll still get one

 

[MFB]

All microcontrollers have their limitations, that's why I'm typing this on non-PICAXE PC, and Grogsters expansion list is therefore a good example of when it might be worth considering the use a more powerful device.

Without having to learn a new language the readily available Maximite chip can directly support SD card interfacing, composite video, VGA graphics (sorry, no HDI) and has a few I2C chips worth of EEPROM and Flash already on-board. This has got to be a more cost effective approach than trying to add these features to a microcontroller.

Yes, the Pi could of course be used for such an expanded systems but does not represent as linear a progression path from the PICAXE as the Basic language Maximite. Let's hope I have not crossed the line by saying this.

 

[manuka]

MFB: No lines crossed here -very well said old son!

 

[lbenson]

I don’t know anything but what I’ve read in the forums or in the media, but I have followed the Raspberry Pi story since February a year ago, when they were mentioned on the defunct LinuxDevices site.

The Raspberry Pi Foundation is a very small charitable organization and they were very constrained financially. The principals had to mortgage their home to be able to finish the design and build the initial 10,000 units. It was a shoestring operation with volunteer assistance at every point.

RS and Farnell didn’t come into the picture publicly until the actual launch date—February 29 when those who were up awaiting a chance to purchase one of the advertised first 10,000 (6am Greenwich time, 1am my time) learned that we would have to go to those sites to “express interest”. Farnell crashed immediately, RS not until sometime after 6:05, when I completed my entry (and managed, it turns out, 2+ months later, to be within the first 10,000).

Then came the Ethernet jack snafu (part without magnetics substituted by the manufacturer after the first few with the right part were tested), and the CE certification hangup (only a software change required to bring a marginal condition squarely into bounds).

Now RS and Farnell say they are manufacturing another 75,000 each, which will not clear the backlog, and they will try to figure out what the ongoing demand will be.

It hasn’t been pretty, but I think it will have been worth the wait, and I think the principals deserve to be commended.

 

[MFB]

Ibenson, I think you will also find this story deserve to be commended...http://geoffg.net/Maximite_Story.html

 

[techElder]

Which is all well, interesting and good, but PICAXE related?

 

[lbenson]

@Texasclodhopper: All of the "little computers" which I have used (that includes NSLU2, WL-520gU, WL-500gpV2, Dockstar, WR703N) have been picaxe-related for me. I use them for home automation and monitoring. They are the communication devices and servers--the picaxe handles the actual monitoring or servo control.

None of them were designed to be "little computers"--they were NAS devices or routers. All except the Dockstar are pretty constrained with regard to flash and ram. A couple of the more useful ones, with multiple USB ports, are not in production (and were fairly expensive when they were--the NSLU2 and the Dockstar).

The Raspberry Pi was designed from the start to be a "little computer"--a fairly powerful one with sufficient expandability and a good price. I also expect that it (and successors) will be available for a long time and that it will be supported by a robust community. Because I've been working with Linux on these devices (openWrt and Debian), that won't be a stumbling block for me.

@MFB: I don't know much about the Maximite--only what was posted in the thread here--and certainly have nothing against it. It looks capable, but I personally wouldn't go in that direction if I had a small Linux system available. Others may certainly feel differently.

 

[techElder]

Isn't your explanation talking about Italian food when I asked about German food? Interesting and good, but the heading does say, "PICAXE Forum."

 

[lbenson]

Isn't your explanation talking about Italian food when I asked about German food? Interesting and good, but the heading does say, "PICAXE Forum."

Definitely and consistently German food since 2007:

http://www.picaxeforum.co.uk/showthread.php?7567-SimpleLan-and-PicAxe

http://www.picaxeforum.co.uk/showthread.php?9425-Pan-and-tilt-control-of-webcam-with-PICAXE-08M

http://www.picaxeforum.co.uk/showthread.php?11908-House-Monitor-System-with-Wireless-PICAXE14M-and-PICAXE08M-Sensor-Modules

http://www.picaxeforum.co.uk/showthread.php?13705-Rivercam-with-PICAXE-pan-tilt-control

http://www.picaxeforum.co.uk/showthread.php?14252-Net-access-for-picaxe-with-openWrt-router

http://www.picaxeforum.co.uk/showthread.php?15907-OpenBridge-development

Or maybe the cooking style would now be labelled "fusion".

 

[Grogster]

Without having to learn a new language the readily available Maximite chip can directly support SD card interfacing, composite video, VGA graphics (sorry, no HDI) and has a few I2C chips worth of EEPROM and Flash already on-board. This has got to be a more cost effective approach than trying to add these features to a microcontroller.

You make a good point there.

I have a Maximite unit(the all SMD assembled one), and did have a play with it, and you are quite right - there are several things bundled natively with it, that the PICAXE does not have.

Still...

I actually quite like modular systems, as if any one bit is not working, USUALLY it is that specific module, so rip it out and replace it, and you are away again. You can't do that with all-in-one systems, although, you can replace the entire system board if you run into a problem.

The one major thing that stopped me playing with the Maximite was it's lack of serial - no 8N1, no I2C, no SPI.

This has been fixed with the latest firmware updates, so I understand, which make the device more appealing, but prior to that, it was really just an interesting toy.

 

[manuka]

Geoff's original Maximite has lead to diverse spin offs, & I've been particularly taken with the US$30 CGMMSTICK1. However a so called DuinoMite version offers an onboard ******* shield footprint (*= in the recent style of London's Boris).

Not to detract from the RPi of course, but perhaps there's scope for PICAXEd equivalents (or hybrids)? I'm reminded of the early (late '70s-early '80s) home PC era, when yelling & screaming your darling's features was the norm. "If you can't beat them,then join them" may be best overall?

 

[Grogster]

I'm reminded of the early (late '70s-early '80s) home PC era, when yelling & screaming your darling's features was the norm. "If you can't beat them,then join them" may be best overall?

OH yes.

I was a teenager when the 8-bit home computer war was on - Remember that?
The main battle was between the Commodore 64 and the Atari 800XL.
I did all my initial programming on the Atari using BASIC.

Not that I am giving the Maximite a hard time - it IS a very clever idea, and had I not already become so obsessively in love with the PICAXE range...

 

[MFB]

Although Grogster's, concluding remark was light hearted, it does risk giving the impression that the PICAXE and Maximite could somehow be competing products. I would certainly not have mentioned the latter on this forum if that were the case.

The PICAXE is a remarkably easy to use microcomputer 'product'. In that it combines a refined development environment with extensive layers of support. Whilst the Maximite is an open source project that offers microcomputer features. As mentioned above, it is now available in a range of hardware configurations and the structured Basic has developed to support the full range of commonly used interfaces. Plus floating point maths and graphics.

The reason that I mentioned the Maximite in the first place was that I thought it was a much better progression path [from the PICAXE] than the Pi, and could not therefore see why the Pi was getting such good exposure on this forum.

 

[Jeremy Harris]

Surely any sort of comparison between these boards and the Picaxe is a bit like comparing apples and oranges, isn't it?

I don't know about others, but all the Picaxe chips I've used have been embedded in projects, ranging from a simple 08M based ebike battery Ah used meter to something a bit more complex like a data logger with µSd card interface. None of them would have been well suited to using something like the R Pi, or any of the other board-based solutions, because price, physical size and current consumption would rule them out.

There's only one thing these board-based systems have that I'd like on the Picaxe, and that's floating point math, but it's usually possibly to work around that limitation without having to incur the expense of adding a µM-FPU.

 

[srnet]

I dont see the likes as Maximite and Arduino as competitors to the core PICAXE market at all.

If you have a good niche market, it makes sense to stick to that market and not try and compete by a major rework of your core product.

I can also see why some PICAXE users perhaps want the benefits of an easy to use product but with the power and flexibility of more complex competitors. Maybe that will happen one day, as there are PICs available with enough core memory to host the expanded firmware.

 

[Grogster]

Although Grogster's, concluding remark was light hearted, it does risk giving the impression that the PICAXE and Maximite could somehow be competing products. I would certainly not have mentioned the latter on this forum if that were the case.

That was not my intention at all. To qualify my remarks, Maximite clever thing, but PICAXE is better, and I prefer it - which should be obvious from my previous comparison comments.

 

[westaust55]

And we're saying that the RPiPeople were unware of that possibility?

Fiendishly flailing fingers, small screen and auto correct are my excuses


I see spelling errors (mine and from many others here even to the most senior level) but just don't feel the need to highlight them all.

 

[boriz]

Also the online documentation could do with some proof reading.

 

[g6ejd]

By way of a rough comparison, lately I have been reading air pressure from a Bosch BMP085 sensor using the I2C bus. I don't have my Rasp Pi yet, but I do have an Ardunio (somewhat similar in controller respects IMO) and to read the sensor takes some 8296 bytes of the 32K, conversely I can do the same on the PICAXE in 921 bytes, which shows the inherent efficiency of using the inbuilt interpreter of the PICAXE and running a p-code rather than compiled code.

I got similar results when getting readings from my GPS module and the PICAXE wins every time on storage needs. I therefore conclude the PICAXE has it every time for ease and speed of programme development. THe moment I incrfease the complxity of the C+ code the byte counts rises quickly.

BTW the Ardunio at 16MHz is comparable with the PICAXE at 4MHz for these tasks, that is there is sufficient processor capacity to undertake the roles in either!

 

[manuka]

The old "horses for courses" saga. You wouldn't take a Ferrari down to the shops !
By way of a further comparison perhaps pnder the Arduino DS18B20 thermostat code below.

// LCD Thermostat
// www.hacktronics.com

#include <OneWire.h>
#include <LiquidCrystal.h>

// Connections:
// rs (LCD pin 4) to Arduino pin 12
// rw (LCD pin 5) to Arduino pin 11
// enable (LCD pin 6) to Arduino pin 10
// LCD pin 15 to Arduino pin 13
// LCD pins d4, d5, d6, d7 to Arduino pins 5, 4, 3, 2
LiquidCrystal lcd(12, 11, 10, 5, 4, 3, 2);
int backLight = 13;    // pin 13 will control the backlight

OneWire  ds(8);        // ds18b20 pin #2 (middle pin) to Arduino pin 8

byte i;
byte present = 0;
byte data[12];
byte addr[8];
  
int HighByte, LowByte, SignBit, Whole, Fract, TReading, Tc_100, FWhole;

void setup(void) {
  pinMode(backLight, OUTPUT);
  digitalWrite(backLight, HIGH); // turn backlight on. Replace 'HIGH' with 'LOW' to turn it off.
  lcd.begin(2,16);              // rows, columns.  use 2,16 for a 2x16 LCD, etc.
  lcd.clear();                  // start with a blank screen
  lcd.setCursor(0,0);           // set cursor to column 0, row 0
  
    if ( !ds.search(addr)) {
      lcd.clear(); lcd.print("No more addrs");
      delay(1000);
      ds.reset_search();
      return;
  }

  if ( OneWire::crc8( addr, 7) != addr[7]) {
      lcd.clear(); lcd.print("CRC not valid!");
      delay(1000);
      return;
  }
}

void getTemp() {
  int foo, bar;
  
  ds.reset();
  ds.select(addr);
  ds.write(0x44,1);
  
  present = ds.reset();
  ds.select(addr);    
  ds.write(0xBE);

  for ( i = 0; i < 9; i++) {
    data[i] = ds.read();
  }
  
  LowByte = data[0];
  HighByte = data[1];
  TReading = (HighByte << 8) + LowByte;
  SignBit = TReading & 0x8000;  // test most sig bit
  
  if (SignBit) {
    TReading = -TReading;
  }
  Tc_100 = (6 * TReading) + TReading / 4;    // multiply by (100 * 0.0625) or 6.25
  Whole = Tc_100 / 100;          // separate off the whole and fractional portions
  Fract = Tc_100 % 100;
  if (Fract > 49) {
    if (SignBit) {
      --Whole;
    } else {
      ++Whole;
    }
  }

  if (SignBit) {
    bar = -1;
  } else {
    bar = 1;
  }
  foo = ((Whole * bar) * 18);      // celsius to fahrenheit conversion section
  FWhole = (((Whole * bar) * 18) / 10) + 32;
  if ((foo % 10) > 4) {            // round up if needed
       ++FWhole;
  }
}

void printTemp(void) {
  lcd.clear();
  lcd.setCursor(0,0);
  lcd.print("Temp is: ");
  lcd.setCursor(0,1);   
  
  if (SignBit) {  
     lcd.print("-");
  }
  lcd.print(Whole);
  lcd.print(" C / ");
  lcd.print(FWhole);
  lcd.print(" F");
}

void loop(void) {
  getTemp();
  printTemp();
  delay(1000);
}

 

[Grogster]

Thats one thing among many that I do love about the PICAXE - the code is really quite simple, even for complicated tasks. I can't remember how many threads I have read where someone is wanting to do something relatively complicated, and the experts here come back with code snippets which allow that to happen within a surprisingly little amount of code.

 

[MFB]

After playing with the Arduino I have also come to the conclusion that [for typical microcontroller tasks] the PICAXE is a more efficient development environment. The Arduino also uses a 'write only' language that someone like me -who only undertakes a project ever few months- finds difficult to understand when revisiting. In contrast, the very alike PICAXE and Maximite versions of Basic are inherently more readable.

 

[srnet]

to read the sensor takes some 8296 bytes of the 32K, conversely I can do the same on the PICAXE in 921 bytes, which shows the inherent efficiency of using the inbuilt interpreter of the PICAXE and running a p-code rather than compiled code

That comparison is ignoring the firmware code in the PICAXE and a tad biased.

For instance, the 28X2 has available for use 4 x 2048 byte slots for code. Thats 8K. The native PIC is a 18F25K22, which is a 32K device. Thus we assume the firmware takes up 32k - 8K = 24K.

So your 921byte PICAXE program takes up 921 + 24K (firmware) = 25K.

So 8K for Arduino, 25K for PICAXE.

 

[nick12ab]

That comparison is ignoring the firmware code in the PICAXE and a tad biased.

For instance, the 28X2 has available for use 4 x 2048 byte slots for code. Thats 8K. The native PIC is a 18F25K22, which is a 32K device. Thus we assume the firmware takes up 32k - 8K = 24K.

So your 921byte PICAXE program takes up 921 + 24K (firmware) = 25K.

So 8K for Arduino, 25K for PICAXE.

Your comparison is also a "tad" biased. The 28X2 actually has 4 x 4096 byte slots for code so that's 16K, leaving 16K for the firmware. And on Arduino, with every simple library you add that is provided as standard on the PICAXE, you add lots more bytes to the used program memory total. An Arduino with 16K memory wouldn't go as far as a PICAXE with 16K memory.

Another problem with Arduino is that it defaults to keeping all serout etc. strings in RAM!!! This instantly decimates available RAM unless you use the fiddly progmem thing.

 

[srnet]

Your comparison is also a "tad" biased. The 28X2 actually has 4 x 4096 byte slots for code so that's 16K, leaving 16K for the firmware

Fair enough, a typo on my part.

Although what I was trying to illustrate is that code size comparisons of this type are fairly meaningless. What you perhaps need to do, is take a program that occupies most of the memory of one type of system and replicate that program on another, then you might have a better idea of which is actually more efficient code space wise.

I have done the GPS locator program in both compiled Mikrobasic and PICXAXE basic, the amount of free memory was about the same on a 18F25k22\28X2, 11K.

And to take a silly example, a simple program of var1 = var2 + var 3, has 32,724 bytes free on the compiler and 16,376 bytes free on the PICAXE.

 

[g6ejd]

I should stated that I was comparing the Arduino to an 18M2 part, although my code would run on an 08M part too.

 

[Marcwolf]

I think the biggest thing going for the RPi is that it runs Unix and that really appeals to a lot of Uni students looking for small (micro) systems that they can develop on in C/C++
Plus it's a standard environment that many of them are used to.

 

[crowland]

The thing I really like about the Picaxe is that it's low power; this makes it really good for things that are used ocasionally. It means that I can run things off batteries and don't need to bother with adding a power switch; just run at 32Khz and use Pause when it's inactive.

It's low price and the lack of a need for support hardware makes it good as a way of doing complex logic operations, the microcontroller equivalent of a 555.

Its disadvantages are that it's not fast and, for educational purposes, the slightly strange language (the order of evaluation of expessions in particular).

If I want more speed, multiple inputs, or more complex mathematics then I would look at an Arduino.

Even for hobby use I see ease and speed of development important, so will throw more memory and processor power at a problem rather than struggle to force a system to do more than it can do easily.

As for the RPi, I think it's a great concept but not for me. I don't mind Linux but can't get on with the Linux philosophy.

This is all personal of course, I'm not suggesting it's what everyone should do.