PICAXE vs Arduino - relative merits

[P Leiby]

Friends,
I know this is like discussing religion or politics, which one should never do with family or neighbors, but do any of you have experience with, and comparative views on, the PICAXE microcontroller series versus Arduinos?

We started with BasicStamps, and are now thrilled with Arduinos, like many of you.
But the PICAXE seems like an amazingly low cost ($3-$10/MCU), easy way to add microcontroller capability to a project.

Obviously, the choice could depend on exactly what one's project is, but that sort of specificity is not always available when one is looking to sink some time and a few $ into learning and hacking around with MCUs, and when many different projects may come up over time.

Here is the best case concise _against_ the PICAXE that I can summon:
1) If you want low cost, just program a $4 ATMega168 chip using the $28 Arduino host-board, and then pull it off and incorporate it bare-bones style in your project. 
2) The ATMega168 is a better chip.
3) BASIC - Ugh!

Is that your assessment?

Doing some homework, I Googled "PICAXE vs Arduino," and got one hit with particularly good content.
In it ,  Frits Lyneborg argues that PICAXE beats Arduino 6 points-to-3, for most modest project needs.
Arduino vs Picaxe http://letsmakerobots.com/node/80

For those unfamiliar, below I have attached a little background info on PICAXEs below.

Thanks very much.
Paul
(of Oak Ridge Robohacks)

Background:
A British product, the PICAXE is a PIC microcontroller with software that lets one program in Basic, very similar to BasicStamp basic.

Here's a third-party website that argues PICAXEs dominate on the three dimensions of cost/functionality/ease-of-

use:
http://dave.fraildream.net/picaxe/index.shtml
Here's the official website:
http://www.rev-ed.co.uk/picaxe/

A key difference between BasicStamp and PICAXE: the (free) PICAXE software is a BASIC _compiler_, converting BASIC to native assembly language, not P-code like the Basic stamp.  So after the program is uploaded to the PICAXE the entire solution is on a single, very low cost chip.  The chip can just be dropped into a solderless protoboard or production board.
I don't know yet what this also implies about program size and speed.  PICAXEs come with a wide range of memory, number of I/O lines and on-chip peripherals.

Cost: PICAXEs (the 8 or 14-pin svelte version, about $3) from SparkFun or this http://www.phanderson.com/picaxe/index.html. The only other things needed are a one-time investment in a $10 host board and $6 serial programming cable for programming.

[Robert L Cochran]

I've used both platforms and currently have development boards for both.
My interest is definitely more along the Freeduino lines. With that
said, let me tell you that I've learned one of the secrets of success.
That is to learn to work with a range of devices and platforms. For
instance, I should probably make April into Rabbit Wireless Modules month.

So much for openness -- let me return to plugging the Arduino and
Freeduino. I don't know why but the AVR, Arduino and Freeduino platform
attracts a lot of very very bright people including talented electrical
engineers willing to spend serious amounts of time and money on design
tasks. These people have been producing a list of interesting Arduino
and Freeduino devices such as the Sanguino and the Bare Bones Board.
Sparkfun has made the Lilypad into a household word. Not to mention its
designer. Arduino and Freeduino have really caught on in terms of
marketing glitz. There are definitely reputations and some amount of
money that can be made on the Arduino/Freeduino wave. It's catching on
the way the car replaced the horse and buggy.

The Picaxe platform doesn't have the appeal of the Arduino and Freeduino
platforms. There are no eye-catching, standout products like the
Sanguino. There are no shields, conductive thread, or authors like Tom
Igoe who can show you how to strap flex sensors to a stuffed monkey and
use an Arduino board to make it do things. For resellers like Brian
Riley and RepRap it is difficult to make money from Picaxe sales given
the investment they have to put in. Brian discontinued his Picaxe kits.
The few kits available don't seem to have a lot of appeal, and the
Picaxe does have two notable problem compared to the Arduino/Freeduino
platforms: first, they seem slow.

There are some support issues too. If you want Picaxe help, your best
source is the Revolution Education forums.

I know there are plenty of very good writers printing articles in Nuts &
Volts touting the Picaxe and putting the chips (mostly the 08M) to work
in some way. Just look at what Ron Hackett, L. Paul Verhage and Vern
Graner are doing with them. The Revolution Education forums have
resident geniuses. But, yawn, the Picaxe doesn't have the device/product
variety and the widespread acceptance that Arduino and Freeduino
captured. And you couldn't hack the Picaxe programming environment the
lat time I checked. With Arduino, numerous folks are out there
contributing code to the IDE because the source is there and you can
build it and change it pretty easily.

Bob

> *Arduino vs Picaxe* http://letsmakerobots.com/node/80

[Matt Joyce]

I think both of these posts are correct and informative.  I would say that arduino currently has the market locked on being an awesome environment to be introduced to electronics projects with.  And I don't think anyone would not agree.  PIC used to dominate this market, but arduino is a juggernaught now.  That being said, it is important to realize at some point that sourcing the right part for the project will become a matter of cost to performance ratio.  PIC is the right tool for the job at times, and there's no denying that. 

Getting back to the glitz of the arduino, let's be honest the arduinos fame isn't due to it's technical specs or it's performance.  It's the people that use it and the projects that are made with it.  It has a charismatic and talented community backing it, and that inspires others to get their hands dirty.  There's immense value there that's not easily quantifiable.  Linux wasn't the best, and it certainly hasn't cemented itself as the best yet either... but the community surrounding linux is by far one of the most innovative groups of people in software today.  And that community, and their love for their scrappy little unix clone is what has made linux a success more than anything else.

All that moderation aside, I would like to throw my flag into the fray and stand fast as a devout hater of the 8051.  Death to the intelfidel!  Non-vonneuman designs are an afront to the universal difference engine.

-Matt

- Show quoted text -

[chf...@columbia.edu]

As an avid Picaxe-user (and Electrical Engineer!), I would like to
point out the following things:

1) From the initial e-mail, they are wrong about the Picaxe running
"native assembly" code. The "BASIC compiler" software is actually a
tokenizer, and it stores its program code on the PIC's EEPROM (Not in
it's "program code" space), I believe. I know that especially on the
08M line, it actually uses 5-bit tokens that are bit-packed across
byte boundaries to increase storage space. This is also why they are
so slow (2-4000 instructions/sec)

2) They are really slow. MUCH slower than an arduino. But an awful lot
of microcontroller projects are limited by the speed of human
interaction, and on that scale, they are blazingly fast. If you need
to do real-time DSP, both the Arduino and the PICAXE probably suck.

3) Picaxe chips are primarily aimed at the educational market, as the
price of a single-chip is incredibly low, and since no dedicated
programmer is required, you can setup everyone in a 30-person lab with
their own microcontroller. If it's for personal use, the cost of a
programmer is often negligible, but if you have to buy in quantity,
it's a big deal. Additionally, you can buy a pretty awesome Picaxe
chip for $10-12, whereas most arduino boards cost ~$20+. If all you
need is to blink some LEDs, an arduino is often overkill.

4) People *do* implement awesome projects with Picaxes! Whereas your
Arduino laptop? Yeah, that's what I thought =)
Example A: http://chrisfenton.com/diy-laptop-v2/

-chris

[Ryan Micallef]

I agree with Matt: thanks for the enlightening posts.  I learned about PICs in undergrad and when I looked into learning about uC's last year, I decided to learn Arduino because of the broad support.  I've wondered what I've been missing.  Apparently, it's a laptop that plays pong.  It would seem that Chris has thrown down the gauntlet.  Arduinophiles, mount up!

Ryan

[Ron Hackett]

I'm also an avid PICAXE user, and I definitely agree with the points
that Chris made.

I think the main thing to keep in mind is that PICAXE was designed
for an educational setting and, as a result, it's cheap and easy to
learn. There are plenty of grade-school children (especially in
England) successfully working with them. That having been said,
there's also no question that the PICAXE chips are SLOW. I did a
speed test on a simple LED-blink program (with no delays) and the
Arduino wass about 200 times faster than the PICAXE. But then, if all
you want to do is blink LEDs, you need to slow down the program anyway.

Personally, I don't think it has to be an either-or situation. My
favorite PICAXE chip is the 08M - with only 8 pins, it makes an
excellent peripheral processor for either a PICAXE (e.g., the 28X1)
or an Arduino master processor. The PICAXE-08M can accomplish a
surprising variety of processing tasks, and it can communicate
serially with its master processor. I'm currently working on a robot
design that uses either an Arduino or a PICAXE-40X1 (or 2) as its
master processor and as many as eight 08M or 14M "slave processors"
which handle various I/O functions. Together, they produce a super-
flexible robotics platform.

Anyway, that's my 2-cents worth. I enjoy working with both PICAXEs
and Arduinos, and I appreciate all I have learned from lurking around
the NYCR group!

-Ron

[Mycroft2152]

I would pretty much have to agree with Ron's thoughts. As a fan and
user of both the PICAXE and the Arduino, I tend to use the PICAXE for
the smaller / slower projects. Once the pin count / requirements get
up there, the ARDUINO beats the PICAXE hands down and is my first
choice.

Re; the comments about the lack of boards for the PICAXE, it is
important to remember that the PICAXE was developed for the secondary
school market in England. As such the marketing by Rev_Ed is directly
positioned toward this market. the hobbyists are really a second
thought, unlike the Arduino.

For the longest time, Rev_Ed tightly controlled the supply of chips
and boards to its distributors and did not allow them to manufacture
and sell boards in direct competition. At the same time Rev_Ed was
focused on the English market and did very little marketing outside,
especially in the US. Even now the inventory requirements by Rev_Ed
preclude all but the major players.

It is only because of the interest by hobbyists and a few non-British
vendors that the availibitly expanded.

Of course, there is always evolution. It does appear that the new
BASIC ATOM NANO on a chip will give both the PICAXE and the ARDUINO
serious competition.

TANSTAAFL!

Myc Holmes

[P Leiby]

I'm grateful to all respondents on this thread over the last day.
Especially the humor of Matt's impassioned "Death to the intelfidel!  Non-vonneuman designs are an afront...." and Chris's amazing PICAXE-based DIY laptop!  (BTW, Matt, aren't both PICs and AVRs/Arduinos also Non-von Neumann, i.e. Harvard or modified Harvard architecture, with separate data and program storage?  http://en.wikipedia.org/wiki/Harvard_architecture)

Thanks also for correcting my mis-apprehension that the PICAXE development software compiled the BASIC to fast assembly code rather than interpreted P-code.  I was confused by this posted description of the PICAXE Version 5 Programmers Editor (http://www.rev-ed.co.uk/picaxe/progedit.htm) that says
"Software Features: ... Has inbuilt BASIC-assembler interpreter so that BASIC programs can be automatically converted into sequential assembler code (requires Serial PIC Programmer)."  It clearly shows BASIC code with generated assembler code side-by-side: http://www.rev-ed.co.uk/picaxe/progedit.htm
Apparently, if you use a different programming cable/device you can alter the PICAXE _program_ memory with new assembled code, overwriting the BASIC interpreter and treating the PICAXE as an ordinary PIC.
There is obviously some key point about how this system works I do not yet fully understand.

In summary, the messages from the discussion seems to be that :
- The PICAXE is inexpensive, compact and convenient for small control tasks that do not require extreme speed; 
- Running an on-board interpreter, the PICAXE's notable slowness was repeatedly observed.
- The Arduino open-source user community is a major strength for that platform, showing a high degree of skill, enthusiasm and vitality;  This is promising for the future;
- While the PICAXE has an active user community (http://www.picaxeforum.co.uk/), it was originally targeted to the UK grade-school educational market, and still focuses in and excels in those type of learning applications
- As always, the correct answer to questions like "Arduino vs PICAXE?" is not only "It depends," but "Why not both?"
- At ~$4 for a self-standing system with low set up costs, the PICAXE makes a great secondary, peripheral processor (thanks Ron).
In fact, I was attracted to the PICAXE 08M in the first place for offloading the task of steadily generating PWM pulses to control servos or ESCs.  The Primary processor will be an Arduino.
(http://www.phanderson.com/picaxe/pwm_stamp.html)

Thanks, and best wishes,
Paul

[e c kern]

On Wed, Mar 4, 2009 at 12:40 AM, P Leiby <ple...@gmail.com> wrote:

are an afront...." and Chris's amazing PICAXE-based DIY laptop!  (BTW, Matt, aren't both PICs and AVRs/Arduinos also Non-von Neumann, i.e. Harvard or modified Harvard architecture, with separate data and program storage?  http://en.wikipedia.org/wiki/Harvard_architecture)

Actually he's got it backwards.  Nominally they're all harvard, but an 8051 can be turned into a von-neumann architecture chip with just one and gate.  I don't think anything like that is possible with either AVR or PIC.

-chris.

--
http://jormungand.net/

[chf...@columbia.edu]

If you want something *really* non von neumann:

http://chrisfenton.com/non-von-1/

Quoting P Leiby <ple...@gmail.com>:

> I'm grateful to all respondents on this thread over the last day.
> Especially the humor of Matt's impassioned "Death to the intelfidel!
> Non-vonneuman designs are an afront...." and Chris's amazing PICAXE-based
> DIY laptop! (BTW, Matt, aren't both PICs and AVRs/Arduinos also Non-von
> Neumann, i.e. Harvard or modified Harvard architecture, with separate data
> and program storage? http://en.wikipedia.org/wiki/Harvard_architecture)
>
> Thanks also for correcting my mis-apprehension that the PICAXE development
> software compiled the BASIC to fast assembly code rather than interpreted
> P-code. I was confused by this posted description of the PICAXE Version 5
> Programmers Editor (http://www.rev-ed.co.uk/picaxe/progedit.htm) that says

> "*Software Features: ... *Has inbuilt BASIC-assembler interpreter so that