Searching \ for '[PIC] Boarding the train of microcontrollers' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'Boarding the train of microcontrollers'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] Boarding the train of microcontrollers'
2007\02\17@165936 by Rikard Bosnjakovic

picon face
Greetings

I'm a newcomer to this list (which was referenced from the
usenet-group sci.electronics.basics) and after lurking around a bit
reading the posts to it, it looks like this list is a collection of
experts and experienced people.

My humble apologies if this post contains questions of too low level
for this list. The reason I'm writing to this mailing list is this:
During the past two years I've been thinking of - but never got the
grip  to - starting to learn how to use microcontrollers. For my next
hobby project, it seems to be a necessity to learn so I thought now -
once and for all - is The Time To Learn.

The next project is going to be a something that looks and acts like a
chess clock, with some few additions. While not completely impossible
to do with a bunch of 555-timers and a lot of logic gates feeding some
7-digit LEDs, the system would probably be a messy harness real quick
and blocks the ability for improvements. So I'm thinking a uC will
take care of all the timing / scoring, and use some external chips
(like a 7-LED decoder and the like) as helper chips.

Now, so far so good in the planning, here's the real reason that's
stopped me for learning microcontrollers: I have absolutely no idea
where to start. Some people say "Go Atmel!", others say "Go PIC!",
others say "Go Foo!".

I have googled around for tutorials, beginners' pages, introduction to
/ comparison between Atmel/PIC and simliar pages, I've seen plenty of
"starter kits" every here and there, some way more expensive than
others making me confused of which to buy, so I still don't know where
to put my foot.

I have 20+ years (software) programming experience. Around 1990-95 I
programmed system assembler for the MC68000, meaning I have atleast
some of the required skills. I think.

So, considering the next project of mine and my background, which kind
of uC might be good enough for me? Or atleast for starting out in the
area, somewhere to put my feet and get the LED flashing.


--
- Rikard.

2007\02\17@172114 by William Chops Westfield

face picon face

On Feb 17, 2007, at 1:59 PM, Rikard Bosnjakovic wrote:

> Now, so far so good in the planning, here's the real reason that's
> stopped me for learning microcontrollers: I have absolutely no idea
> where to start. Some people say "Go Atmel!", others say "Go PIC!",
> others say "Go Foo!".

I really don't think it matters much any more.  Everything is cheap,
everything has compilers and fancy development environments, almost
everything is buyable in small quantities from mail-order dealers.
The big motivator might be if a vendor has an "evaluation board"
that is very close to what you need for your first project.  For
instance, the Atmel "AVR Butterfly" is a phenomenal "value" IMO.

You might enjoy my "instructable" on "How to choose a microcontroller"
http://www.instructables.com/id/EKRQRQHQDIEWH1IMC9/

BillW

2007\02\17@173506 by Jake Anderson

flavicon
face
Rikard Bosnjakovic wrote:
{Quote hidden}

Probably the easiest in terms of toolchain and all that other stuff
would be a PIC and JAL.
A 16F877A or something for the micro and perhaps a pickit2 usb
programmer from microchip.
you can be flashing a LED inside an hour (especially with the "flash a
led" page from wouter (i think))


2007\02\17@175040 by Wouter van Ooijen

face picon face
> So, considering the next project of mine and my background, which kind
> of uC might be good enough for me? Or atleast for starting out in the
> area, somewhere to put my feet and get the LED flashing.

I'd say for a beginner support is more important than technical
features. You can't go wrong with PIC or AVR, both are pouplar and
technically OK. Choose on availability, support from fellow users, and
tools (in that order). Price per chip is much less important, 10 chips
will probably serve you for a year or so. Do you have programming
experience? In a particular language?

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu



2007\02\17@175523 by peter green

flavicon
face
> Probably the easiest in terms of toolchain and all that other stuff
> would be a PIC and JAL.
> A 16F877A or something for the micro and perhaps a pickit2 usb
> programmer from microchip.
> you can be flashing a LED inside an hour (especially with the "flash a
> led" page from wouter (i think))
if you are going to go for a 28 or 40 pin pic i think the PIC18 series (the PIC18F2620 and PIC18F4620 being the best choices from that series for the hobbyist imp) is a better bet, its a much nicer architecture and there is an only slighly crippled version of microchips C compiler for it availible free.


2007\02\17@182326 by Rikard Bosnjakovic

picon face
On 2/17/07, Wouter van Ooijen <spam_OUTwouterTakeThisOuTspamvoti.nl> wrote:

> I'd say for a beginner support is more important than technical
> features. You can't go wrong with PIC or AVR, both are pouplar and
> technically OK. Choose on availability, support from fellow users, and
> tools (in that order). Price per chip is much less important, 10 chips
> will probably serve you for a year or so. Do you have programming
> experience? In a particular language?

Except for the 'net (mailing lists, usenet-groups and the like), I'm
on my own when it comes to electronics. Noone (that I know of) in my
area has any interest of it at all. And regarding availability - since
I'm living in a small town there's no electronics store or similiar
here, so I have to fetch everything from online orders. I guess that
would be a "high availability" for everything.

This leaves me with the tools. I have just now read through the FAQ on
your website and I am currently browsing around for PIC-programmers.
The Microchip one would certainly be a nice thing, but alas - the
price was a bit too steep for me at this time. Instead, I found this
on eBay: http://tinyurl.com/2d2m4t but I'm not sure it's a ripoff or
something that might be usable for me.

And for your final question regarding programming experience. Yes, I
started software programming back in 1989 or simliar. I've waded
through a lot of languages, like Pascal and C (both of which you
mention on your JAL-pages), but the one I use nowadays at a daily
basis is Python. I'm not sure Python is of any use for
microcontroller-coding, so I don't know if that would count as
(useful) experience.


--
- Rikard.

2007\02\17@183350 by Forrest Christian

flavicon
face
Rikard Bosnjakovic wrote:
> So, considering the next project of mine and my background, which kind
> of uC might be good enough for me? Or atleast for starting out in the
> area, somewhere to put my feet and get the LED flashing.
>
>  

My first impression was to point you towards the parallax Basic stamp
stuff.   This is a good place for the beginner to start sticking their
toe in the water so to speak.  For $150 you get everything you need to
start blinking LED's.   The downside is that it's all in BASIC and the
whole thing has been "sanitized" so that almost anyone can understand
it.   I can't state for certain whether or not this is a good thing or
not.  But it would get you started with the basic concepts of the whole
microcontroller thing.

However, as I was composing my email, I was thinking about what tools I
would recommend if you didn't want to go the basic stamp route.   There
is always the grab a breadboard and a pic and a handful of parts (don't
forget the ICD, etc) and go for it method, which is what I typically  
use in the early prototyping stage of my products, but this is a pain.

Then I got really thinking...   For a long time I've had my eye on the
Easypic4 development board ( http://www.mikroe.com/en/tools/easypic4/ ).
 I think this might be exactly what you are looking for.  You can buy
it domestically at http://www.circuit-ed.com/easypic4.htm .   For $169
you get a complete board with lots of buttons, a multiplexed LED
display, LCD, etc. etc. etc.  You should be able to do all of the "R&D"
on your project with this tool.   You can use the multiplexed LED to
test out the display stuff.  You can use the buttons for the inputs.  
And so on.

The reasons I've been looking at the easypic4 (again I don't have one,
so this is all based on other recommendations and reading) is that it's
one of those boards which you can do a lot of things without having to
resort to circuit building.  I get kinda tired of putting lots of
components on a breadboard just to be able to play with a particular
piece of code.  The manual looks awesome (easy to read/understand), and
it comes with examples in the micro-e tools.

One disadvantage I see is that the onboard programmer uses a proprietary
interface which means that programming and ICD will need to be through
the mikro tools.   All this means is that when programming, you'll need
to use their software for the programming itself (using a .hex file from
your chosen programming tool), and that the debugging part probably will
only work with their compilers.   It does look like the demo versions of
their compilers will build small programs without restriction, and that
they are priced reasonably.  I can't comment beyond that.

After you do the design with this board, then you can actually build
your project, using the software you've initially prototyped on the board.

I'm not 100% sure this is the direction I would take, but it might be
something to look at...

-forrest

2007\02\17@184352 by peter green

flavicon
face

> Except for the 'net (mailing lists, usenet-groups and the like), I'm
> on my own when it comes to electronics. Noone (that I know of) in my
> area has any interest of it at all. And regarding availability - since
> I'm living in a small town there's no electronics store or similiar
> here, so I have to fetch everything from online orders. I guess that
> would be a "high availability" for everything.
roughly where are you? if you tell us someone here can probablly point you in the direction of some decent mail order (or possiblly even local) suppliers

>
> This leaves me with the tools. I have just now read through the FAQ on
> your website and I am currently browsing around for PIC-programmers.
> The Microchip one would certainly be a nice thing, but alas - the
> price was a bit too steep for me at this time. Instead, I found this
> on eBay: http://tinyurl.com/2d2m4t but I'm not sure it's a ripoff
its a ripoff of microchips ICD2.

not that thats a bad thing, it seems to have proper USB in place too (some ICD2 clones use a USB to serial chip instead which is cheaper but slower).



2007\02\17@184550 by Victor Faria

picon face
Or you can try using a simulator.
http://www.virtualbreadboard.com  also http://www.oshonsoft.com  this one comes with a
compiler.
regards
victor
{Original Message removed}

2007\02\17@190918 by Rikard Bosnjakovic

picon face
On 2/18/07, peter green <.....plugwashKILLspamspam@spam@p10link.net> wrote:

> roughly where are you? if you tell us someone here can probablly point you in the direction of some decent mail order (or possiblly even local) suppliers

I'm located in Sweden / Europe, a small harbor town called Vänersborg.
The biggest electronics supplier in Sweden is Elfa (http://www.elfa.se, got
English pages as well), but they are like 250% overpriced on just
about everything. Their shipping time is usually <24 hrs (good for
companies, bad for hobbyists like me), hence their steep prices.


> > price was a bit too steep for me at this time. Instead, I found this
> > on eBay: http://tinyurl.com/2d2m4t but I'm not sure it's a ripoff
> its a ripoff of microchips ICD2.

Ripoff as in "worthless junk that will work only 5% of the times" or
ripoff as in "some pirated version with a nice front"?

> not that thats a bad thing, it seems to have proper USB in place too (some ICD2 clones use a USB to serial chip instead which is cheaper but slower).

I think I can live with a slower version at a start. The most
important thing is to get hold of a programmer to start with, then -
when maturing in the area of microcontrollers - increase the levels of
tools and knowledge.

--
- Rikard.

2007\02\17@194646 by peter green

flavicon
face

> I'm located in Sweden / Europe, a small harbor town called Vänersborg.
> The biggest electronics supplier in Sweden is Elfa (http://www.elfa.se, got
> English pages as well), but they are like 250% overpriced on just
> about everything. Their shipping time is usually <24 hrs (good for
> companies, bad for hobbyists like me), hence their steep prices.
yeah, thier prices do look rather steep, surely there must be other suppliers arround.

farnell appear to have a branch in your country, thier prices seem about the same for one off pics but they seem to drop off more at the 10+ break.

microchip direct are a reasonable source for pics if you can wait a while and plan to order quite a few at once (for one offs the headline price looks good but the delivery charges are horrible). Microchip also give away free samples pretty readily.

> > > price was a bit too steep for me at this time. Instead, I found this
> > > on eBay: http://tinyurl.com/2d2m4t but I'm not sure it's a ripoff
> > its a ripoff of microchips ICD2.
>
> Ripoff as in "worthless junk that will work only 5% of the times" or
> ripoff as in "some pirated version with a nice front"?
its almost certainly a clone based on reverese engineered schematics that are floating arround on the web

the sellers feedback seems pretty damn good, one of the things about ebay is if you wan't to keep selling something you have to make it at least reasonable quality.

2007\02\17@195436 by Bob Blick

face
flavicon
face
On Sun, 18 Feb 2007, Rikard Bosnjakovic wrote:
{Quote hidden}

I know this is the wrong list to say this in, but since I've been on the
list and contributing for 9 years I feel OK doing it. Things have changed
considerably in the last 9 years. Back then, the PIC really was the best
in class. But now I'd have to say, as familiar and comfortable with the
PIC as I am, I rarely choose it for any new designs. There is no
application where the PIC is superior. I don't even trust that part
numbers will be available in the future. And don't you love how a data
sheet is preliminary until the day the part is obsoleted?

If you are starting out and want a simple microcontroller that is
versatile and can run at a variety of speeds and supply voltages, use an
AVR. It is even with or beats the PIC in most areas, but is way ahead when
it comes to writing code. AVR is so much easier to program than the PIC12,
PIC16 and PIC18. And if you use a C compiler, I have been using PIC C
compilers for 7-plus years and I have to say that compiler bugs are
prevalent plus harder to track and reproduce because of the nasty PIC
architecture.

For more complex designs, the higher-end PICs are too expensive, full of
bugs, and not really well thought out.

OK, flame on, but I'm speaking from the heart here, I am not trolling. I
have a sentimental attachment to the PIC but it's not the right choice for
me anymore.

Cheerful regards,

Bob

2007\02\17@201347 by Jinx

face picon face
> farnell appear to have a branch in your country

Radiospares too http://www.rsonline.se

If it were me I think I'd get

16F877A part# 4671690 DIP40 56,45kr for 1
18F452 part# 4672126 DIP40 68,70kr for 1

Unfortunately RS are a little behind the times, understandably,
so they don't have more recent devices

Considering the complexity of those two, you could spent
the rest of year playing with them, and still not be done, which
isn't bad value for money if you look at it that way

Then maybe you could look at someone like Arrow

http://www.arrowne.com/arrow_worldwide/sweden/main.html

for a tube or, as Peter said, Microchip (buy / sample), especially
if you want anything in the nanoWatt range

There are at least a couple of PIClisters in Sweden, maybe
they can help

2007\02\17@201456 by Marcel Duchamp

picon face
Bob Blick wrote:
{Quote hidden}

Oh, come on now, Bob; tell us how you really feel!

I'm just kidding of course. You made it very clear how you feel.  I'm
interested in feedback here.  I have on my desk in front of me an ARM
development board along with the $20 TI MSP430 USB development stick.
There is lots out there to choose from these days.

Although I have also been playing with PIC's since the mid nineties or
so, it does look like it's time to "have a look around".

What do others have to say?




2007\02\17@204908 by Byron A Jeff

face picon face
On Sat, Feb 17, 2007 at 10:59:36PM +0100, Rikard Bosnjakovic wrote:
> Greetings

Welcome!

> I'm a newcomer to this list (which was referenced from the
> usenet-group sci.electronics.basics) and after lurking around a bit
> reading the posts to it, it looks like this list is a collection of
> experts and experienced people.

Some. Lots of new people too.

> My humble apologies if this post contains questions of too low level
> for this list.

No apologies necessary. In general as long as you make an attempt to
gather some information yourself, all questions are welcome.

> The reason I'm writing to this mailing list is this:
> During the past two years I've been thinking of - but never got the
> grip  to - starting to learn how to use microcontrollers.

Good! I'm a firm believer that microcontrollers are the 555 of the new
millenium. A basic programmable building block to a whole lot of
electronics projects.

> For my next
> hobby project, it seems to be a necessity to learn so I thought now -
> once and for all - is The Time To Learn.

Do it and do it now! The hardest part of getting started on any project.

> The next project is going to be a something that looks and acts like a
> chess clock, with some few additions. While not completely impossible
> to do with a bunch of 555-timers and a lot of logic gates feeding some
> 7-digit LEDs, the system would probably be a messy harness real quick
> and blocks the ability for improvements.

Which is exactly why microcontrollers are so useful.

> So I'm thinking a uC will
> take care of all the timing / scoring, and use some external chips
> (like a 7-LED decoder and the like) as helper chips.

No need for that either. A 7-LED decoding routine is a trivial piece of
software. Little more than a jump table that converts digits/characters
into display values.

> Now, so far so good in the planning, here's the real reason that's
> stopped me for learning microcontrollers: I have absolutely no idea
> where to start. Some people say "Go Atmel!", others say "Go PIC!",
> others say "Go Foo!".

There are a bunch of specs that you need to think about. A few include:

1) Price.
2) Availability
3) Ease/Cost of development environment.
4) Peripheral features (timers, USB, PWM, and the like)
5) The available user community.

But you don't necessarily need to be married to one particular architecture.
I think that most hobbyist fall into the "Love what you Learn" black hole.
Once they learn a particular part, they treat it loke a hammer where every
problem is a nail. It's understandable because you can leverage what you learn
over a ton of projects. But having an open mind can get you into better choices
especially a few years down the road. I mean there are folks who still start
projects with PIC 16F84 parts, though they are virtually obsolete.

> I have googled around for tutorials, beginners' pages, introduction to
> / comparison between Atmel/PIC and simliar pages, I've seen plenty of
> "starter kits" every here and there, some way more expensive than
> others making me confused of which to buy, so I still don't know where
> to put my foot.

OK. Let's eliminate 4 and 5 above since your two contenders have extremely
active communities and similar periperals. For one off projects, price is
only a moderate concern for raw materials. So that leaves availability and
ease/cost of the development environment from my initial list.

There was a time that Atmel availabilty was suspect. However these days
pretty much any part is pretty easy to get into. In fact Microchip for example
has a simple program where you can get up to 12 parts delivered to your door
in a week or so. I just took advantage of it to check out the 40 pin 16F887
parts, which seems to be the 40 pin version of a 16F88. I like all of the
following features:

- Self programmable, so a bootloader can be installed.
- internal 8 Mhz osciallator, so no external hardware needed for time
 insensitive applications
- Wide array of periperals including ADC, PWM, serial, I2C, SPI, and
 multiple timers
- Nice DIP packages.
- Still has 5V operation available.

I'm sure that Atmel has similar offerings.

> I have 20+ years (software) programming experience. Around 1990-95 I
> programmed system assembler for the MC68000, meaning I have atleast
> some of the required skills. I think.

A lot of the required skills.

> So, considering the next project of mine and my background, which kind
> of uC might be good enough for me? Or atleast for starting out in the
> area, somewhere to put my feet and get the LED flashing.

My suggestion for hobbyists is to start at the top of heap and then start to
work your way downward to a comfort level. By getting the biggest, more
feature packed part, when you find you need a feature or some more I/O,
it's there.

The real question isn't the hardware, but the software environment. You really
do not need to program in assembler long term. I generally advise that one
should learn assembler at a good reading level simply because many ideas are
communicated in assembler. Also it's handy for those time/performance
critical routines that every project ends up having. But a high level language
can really simplify the development and maintenance process.

I haven't yet gotten to where I'm suggesting that you go. I'm still functioning
in the 16F family of PICs and still writing in assembler. But I've been at
it for awhile. On the PIC side if you're just starting out I think that a
40 pin USB based 18F part and a HLL like C may be a good fit. A part like the
18F4550 would sit at the top of the heap.

Just some thoughts.

BAJ

2007\02\17@210716 by peter green

flavicon
face

> I haven't yet gotten to where I'm suggesting that you go. I'm
> still functioning
> in the 16F family of PICs and still writing in assembler. But I've been at
> it for awhile. On the PIC side if you're just starting out I think that a
> 40 pin USB based 18F part and a HLL like C may be a good fit. A
> part like the
> 18F4550 would sit at the top of the heap.
if you aren't actually going to use the USB perhiperal (which isn't really a good place for newbies to start) then the 4550 is somewhat poor, it can't combine the internal oscilator with the PLL, one IO line is permanently lost to Vusb and its not the most readilly availible part either (on the plus side however its clock structure is more powerfull but also more complex)

the 18F2620/18F4620 are the best of the normal (e.g. not unusual perhipherals like usb and can) 18 series pics that come in a nice dil package.

one nice thing is that all the 18F pics are very similar, once you've programmed one programming the rest is easy.


2007\02\17@211102 by Rikard Bosnjakovic

picon face
On 2/18/07, Jinx <joecolquittspamKILLspamclear.net.nz> wrote:
> > farnell appear to have a branch in your country
> Radiospares too http://www.rsonline.se
[...]
> http://www.arrowne.com/arrow_worldwide/sweden/main.html

Thanks for finding the companies.

Unfortunately, all of the three above listed requires you to be a
company in order to buy items from them. As a hobbyist not running a
company, I cannot purchase anything from them.

This is usually the standard behaviour for all electronic components
retailers here in Sweden, except for Elfa (their overpricings do
however make you really depressed). I see a lot of web stores in the
US that's got fair pricings and gazillions of components, but they
usually don't ship abroad and if they should do this, the shipping
costs is around 300% of the actual order value. And this is why it
sucks to be an EE-student (or mechatronics, like me) in Sweden who is
in need for components/tools; there are no fair priced providers.

Anyway. I found another PIC-programmer on ebay:
<URL:http://tinyurl.com/2uv92u>. I guess this one is also a pirated
version (due to the fact that it's shipped from China), but since the
seller has got 15000+ feedback, it cannot be entirely crappish. I was
about to buy it, but then Bob Blick's post arrived to the list, making
me hesitate yet once more whether i really *should* go PIC or Atmel,
which is the reason that has stopped me earlier from beginning with
microcontrollers.

But instead of bailing out completely, I kept the sanity; so what if
PIC sucks and Atmel rocks, this will be something I will find out on
my own when I get more experienced. And this choice is really not like
painting yourself into a corner (except for the cost of tools), just
because I might chose PIC at this time - I must start SOMEWHERE! -
there is nothing saying I cannot chose AVR later on. Plus, Wouter's
pages and JAL looks pretty user friendly for my level.

So, I'll go for PIC this time. Can you say anything about that chinese
programmer in the URL above? I mean compared to the other one I listed
earlier (http://tinyurl.com/2d2m4t).

It's 3 o'clock in the morning here. It's usually a problem for me,
because when I get a new design for a project like the one I'm having
now, I'm having a hard time getting tired until all details are done
for the day. The details for today were to make a draft for the
block-diagram and to peek some more into microcontrollers. The draft
got done 6 hours ago, and thanks to Wouter's PIC-pages, I think I can
finally rest my case (regarding which branch I will chose as a start)
for today.


--
- Rikard.

2007\02\17@212509 by peter green

flavicon
face

> Unfortunately, all of the three above listed requires you to be a
> company in order to buy items from them. As a hobbyist not running a
> company, I cannot purchase anything from them.
are you sure? have you actually tried to order?

the british branch of farnell seems to have no problem with private orders.


2007\02\17@213233 by William Chops Westfield

face picon face

On Feb 17, 2007, at 5:14 PM, Marcel Duchamp wrote:

> it does look like it's time to "have a look around".
> What do others have to say?
>
I say you should always be looking around.

Outside of pic and AVR, you start to run into chips with
odd voltage requirements, and low current pins.  The Luminary
"$1 ARM" chip blows 7 of 28 pins for power and features "up to"
8mA drive capability, for instance.  Since it uses JTAG for
programming access and has a feature (or maybe bug) that you can't
get at the JTAG functionality once the pins are reprogrammed for
regular IO, you throw away another 5 pins (usually) for JTAG, and
there's no internal clock, so you wind up with 13 anemic IO in a
28pin package...

The problems microchip has are pretty endemic to the industry as
a whole, I think.  I agree that I don't think microchip understands
the "larger" chips as well as they should, but I think they understand
the low end better than most.

BillW

2007\02\17@214021 by Rikard Bosnjakovic

picon face
On 2/18/07, peter green <.....plugwashKILLspamspam.....p10link.net> wrote:

> are you sure? have you actually tried to order?

Yes. As soon as you mentioned the companies (which I had never heard
of before) I went there trying to create a user account. Required
details was among others company name, tax number etc.

> the british branch of farnell seems to have no problem with private orders.

Really? The swedish one does not allow private orders.

I'm like always completely out of luck.


--
- Rikard.

2007\02\17@220628 by Jinx

face picon face

> > Radiospares too http://www.rsonline.se
> [...]
> > www.arrowne.com/arrow_worldwide/sweden/main.html
>
> Thanks for finding the companies.
>
> Unfortunately, all of the three above listed requires you to be a
> company

That would be most unusual for RS and Farnell. And I can
have cash sales with Arrow, as long as it's over NZ$50. I
think most companies will do cash sales to anyone. In fact
in the case of RS and Farnell, the higher prices are part of
the cost of cash sale and one-off convenience. Trade firms
like Arrow will probably not establish a line of credit (ie pay
for goods later) unless you're a regular big buyer or company
or both

Are you sure the information RS require is not just for their
records ? That's my experience with RS in New Zealand.
Whether I to register or order under a company name is
irrelevant. However, I have been a customer of theirs for
20 years, maybe new customer criteria is different now,
but I find that hard to believe. Looking at the equivalent
page on http://www.rsnewzealand.com I see nothing that would
stop me opening an account as a sole trader. I simply would
not fill in the company section. They are going to run credit
checks of course, but if you've no history of debts and
don't have a problem with that sort of background enquiry,
then go for it. Even if you hardly use them, the account is
there when you need it

2007\02\17@222922 by Howard Winter

face
flavicon
picon face
On Sun, 18 Feb 2007 02:24:10 -0000, peter green wrote:

> > Unfortunately, all of the three above listed requires you to be a
> > company in order to buy items from them. As a hobbyist not running a
> > company, I cannot purchase anything from them.

> are you sure? have you actually tried to order?
>
> the british branch of farnell seems to have no problem with private orders.

Yes, RS likewise (you have to be a company to use their "Trade Counters" but they'll accept mail order from anyone), but there may be some reason
why Sweden is different - perhaps consumer protection laws make it too much hassle to deal with individuals there?

Rikard, you've been talking about Wouter's web site - have you considered buying components from him?  I believe his shipping charges are very
reasonable, so it may be worth trying him!

Cheers,


Howard Winter
St.Albans, England


2007\02\17@223032 by Aaron

picon face


Bob Blick wrote:

>If you are starting out and want a simple microcontroller that is
>versatile and can run at a variety of speeds and supply voltages, use an
>AVR. It is even with or beats the PIC in most areas, but is way ahead when
>it comes to writing code. AVR is so much easier to program than the PIC12,
>PIC16 and PIC18. And if you use a C compiler, I have been using PIC C
>compilers for 7-plus years and I have to say that compiler bugs are
>prevalent plus harder to track and reproduce because of the nasty PIC
>architecture.
>

If the original poster is considering AVRs, do be sure to check out
Atmel's US$49 Dragon.  It is a programmer which can also emulate devices
with less than 32KB of flash.  I love mine!
http://www.atmel.com/dyn/products/tools_card.asp?tool_id=3891

Aaron

2007\02\18@025707 by Bob Axtell

face picon face
Marcel Duchamp wrote:
{Quote hidden}

Well, it is true that Atmel has made a strong effort to support us
recently, but in times past, Atmel
had a bad habit of dropping parts just as soon as I consider using them.
I'm still a little bit skeptical,
whereas MC is still supporting the PIC16C54 after 15 years.

However, lately Microchip has taken several left turns. They are
releasing too many parts too fast,
and the onboard peripherals seem to be buggy; on one PIC I never got the
I2C engine to
work properly at all so had to bit-bang it.

I no longer trust the PIC EEPROM unit in the nanowatt series- I have to
store variables in triplet for
reliability. This is the most serious problem MC has, and it had better
be addressed quick or I will
drop the nanowatt PIC in future designs. This is VERY annoying; MC knows
they have a serious problem, and
their workarounds are really brain-dead. If you make careful tests, you
will be surprised at the poor
reliability of the EEPROM unit. BTW, don't make too much fun of the
PIC16C stuff; its EEPROM _works_.

I don't like the self-programming algorithms; they vary so much from
chip to chip that each chip must have
a bootloader designed from scratch.

I also think that dropping the old tried and true serial programming
scheme that was a totally separate
serial path (instead of access thru an external serial port that really
makes the chip to program itself) was
a really bad idea.

They should release the "ICD2 command set" to the public, so that a
decent ICD2 can be designed and
sold by someone competent, such as Olin's company.

They should anticipate the end of Microsoft dominance (it will all be
over by 2015) by providing a solid Linux
version of MPLAB and ICD2.

There's more..

--Bob
>
>
>
>  

2007\02\18@030818 by Wouter van Ooijen

face picon face
> making
> me hesitate yet once more whether i really *should* go PIC or Atmel,
> which is the reason that has stopped me earlier from beginning with
> microcontrollers.

If you can't decide just flip a coin, you can't go wrong with either. It
would be a pity to cacel a nice hobby just because there are too many
good options! And especially when you programm in C: experience with one
is almost 1:1 transferrable to the other (and to MSP430, ARM, etc).

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\18@030818 by Wouter van Ooijen

face picon face
> I'm just kidding of course. You made it very clear how you feel.  I'm
> interested in feedback here.  I have on my desk in front of me an ARM
> development board along with the $20 TI MSP430 USB development stick.
> There is lots out there to choose from these days.
>
> Although I have also been playing with PIC's since the mid
> nineties or
> so, it does look like it's time to "have a look around".
>
> What do others have to say?

I'd say PICs still have an advantage in the wide range of available
chips, the 14-bit core is still nicely priced, and I havn't seen
anything like the 10F's elsewhere. Climbing up the food chain things get
less nice, 18F's might still be OK, but I much prefer ARMs over 30F's.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\18@030818 by Wouter van Ooijen

face picon face
> I'm located in Sweden / Europe

That makes my Wisp628 reseller Jan-Erik Soderholm 'near' to you (at
least, from a world-wide perspective).

If you even want to do a 'big' order, check http://www.reichelt.de

For smaller orders you might check my webshop.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\18@030818 by Wouter van Ooijen

face picon face
> Except for the 'net (mailing lists, usenet-groups and the like), I'm
> on my own when it comes to electronics.

That still gives you anywhere from 1k to 1M people to give you (often
conflicting!) advice.

> The Microchip one would certainly be a nice thing, but alas - the
> price was a bit too steep for me at this time.

too late now, but did you consider for instance pickit2?

> And for your final question regarding programming experience. Yes, I
> started software programming back in 1989 or simliar. I've waded
> through a lot of languages, like Pascal and C (both of which you
> mention on your JAL-pages), but the one I use nowadays at a daily
> basis is Python. I'm not sure Python is of any use for
> microcontroller-coding, so I don't know if that would count as
> (useful) experience.

It's my favourite too (at least for certain PC applications).

If you have C exeprience I suggest you start with 18F PICs and the
Microchip C18 compiler. Get a few 18F252 (or 18F2520), 18F452 (or
18F4520), or maybe jump straight to one of the USB-capable chips for
more fun later.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\18@031022 by Wouter van Ooijen

face picon face
> The Luminary
> "$1 ARM" chip blows 7 of 28 pins for power and features "up to"
> 8mA drive capability, for instance.  Since it uses JTAG for
> programming access and has a feature (or maybe bug) that you can't
> get at the JTAG functionality once the pins are reprogrammed for
> regular IO, you throw away another 5 pins (usually) for JTAG, and
> there's no internal clock, so you wind up with 13 anemic IO in a
> 28pin package...

Juk! I was going to read up on these chips, but this saves me the
trouble. No wonder they require a sign-up before you can get the
datasheets, you don't want this details known to the public...

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\18@035453 by William Chops Westfield

face picon face

On Feb 18, 2007, at 12:10 AM, Wouter van Ooijen wrote:

>> "$1 ARM" chip blows 7 of 28 pins for power and features "up to"
>> 8mA drive capability, for instance.  Since it uses JTAG for
>
> Juk! I was going to read up on these chips, but this saves me the
> trouble. No wonder they require a sign-up before you can get the
> datasheets, you don't want this details known to the public...
>
Oh, it's not THAT awful.  Maybe.  I certainly don't see a reason to
not look at their literature; I haven't seen any negative impact from
giving them (or ARM, where you'll need to sign up again to see the
ARM manuals) my information.

It's one more set of power pins than the usual 28pin PIC, plus
a Vcpu (from an onboard regulator) that needs an external bypass
cap.  They have a workaround for the JTAG issue (consists of having
your software have a way of turning the pins back into JTAG) that
isn't TOO awful.  (Just don't do it wrong.)

The biggest issues I'm having doing anything with it are more
philosophical.  Since the "cortex" is a thumb-only implementation
of ARM7 with is a subset of ARM9, there doesn't seem to be any
"bottom-up" documentation (since you already do ARMs, you'll have
less problem with this.)  And while encouraging programming in C
is nice, I could do with more documentation of the actual hardware
instead of "oh just use our C library; here's the 400page document
that tells you how to use it."  Their 48 pin chips look more
promising (but less hobbyist-friendly.)

It's a good example of what I meant when I said Microchip understands
the "low end" of microcontrollers.  It reminds me of when Atmel copied
the 8pin PICs but didn't implement an internal oscillator or reset.
"It's got three I/Os!  What did you expect from an 8pin chip?"  Heh.
I expected what Microchip managed to do!

(Hmm.  If I had to pick a vendor for "cluefullness" as shown by the
products they're coming out with now, I think I'd go with TI.  The
MSP430 group keeps doing things that cause me to go "that was a good
idea!  Clever!")

BillW


2007\02\18@052638 by Wouter van Ooijen

face picon face
> Oh, it's not THAT awful.  Maybe.  I certainly don't see a reason to
> not look at their literature;

main reason: time. must read up on SAM7 too.

> I haven't seen any negative impact from
> giving them (or ARM, where you'll need to sign up again to see the
> ARM manuals) my information.

I want to sell chips, many to hobbyists/small-scale-professionals.
Havaning to sign up will not encourage potential customers.

> It's one more set of power pins than the usual 28pin PIC, plus
> a Vcpu (from an onboard regulator) that needs an external bypass
> cap.  

ARMs all seem to have a lot of power pins. That does not hurt much on
the higher pincount packages, but it is a waste on 28 an lower.

> They have a workaround for the JTAG issue (consists of having
> your software have a way of turning the pins back into JTAG) that
> isn't TOO awful.  (Just don't do it wrong.)

Anything that can be done wrong in software will be (at least by
someone..)

> The biggest issues I'm having doing anything with it are more
> philosophical.  Since the "cortex" is a thumb-only implementation
> of ARM7 with is a subset of ARM9, there doesn't seem to be any
> "bottom-up" documentation (since you already do ARMs, you'll have
> less problem with this.)  And while encouraging programming in C
> is nice, I could do with more documentation of the actual hardware
> instead of "oh just use our C library; here's the 400page document
> that tells you how to use it."  Their 48 pin chips look more
> promising (but less hobbyist-friendly.)

There are plenty good 48 pin SMD ARMs. A hobbyist friendly DIP ARM would
be a breakthrough.

> It's a good example of what I meant when I said Microchip understands
> the "low end" of microcontrollers.  It reminds me of when Atmel copied
> the 8pin PICs but didn't implement an internal oscillator or reset.
> "It's got three I/Os!  What did you expect from an 8pin chip?"  Heh.
> I expected what Microchip managed to do!

Agreed.

> (Hmm.  If I had to pick a vendor for "cluefullness" as shown by the
> products they're coming out with now, I think I'd go with TI.  The
> MSP430 group keeps doing things that cause me to go "that was a good
> idea!  Clever!")

Maybe I'll have to look into that line too. Anyone got some 9-day weeks
idling in their dustbins?

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\18@075102 by Mattias Hedenskog

flavicon
face
Hi!

> I'm located in Sweden / Europe, a small harbor town called Vänersborg.
> The biggest electronics supplier in Sweden is Elfa (http://www.elfa.se, got
> English pages as well), but they are like 250% overpriced on just
> about everything. Their shipping time is usually <24 hrs (good for
> companies, bad for hobbyists like me), hence their steep prices.

I'm from Sweden as well and regarding the overall price range in
electronic components over here is what I'd do to keep the expenses
down..

1. Get a fairly complete setup of electronic components from Belganets
Dataelektronik (http://www.bde.se). Make sure to check out the surplus
stock, it's pretty hilarious.

2. I recommend Wouter's Wisp628 (PIC) for programming purposes.

3. Start to stock up free PIC samples from http://www.microchip.com

Regards
Mattias Hedenskog

2007\02\18@075830 by Byron A Jeff

face picon face
On Sun, Feb 18, 2007 at 12:23:19AM +0100, Rikard Bosnjakovic wrote:
> And for your final question regarding programming experience. Yes, I
> started software programming back in 1989 or simliar. I've waded
> through a lot of languages, like Pascal and C (both of which you
> mention on your JAL-pages), but the one I use nowadays at a daily
> basis is Python. I'm not sure Python is of any use for
> microcontroller-coding, so I don't know if that would count as
> (useful) experience.

Actually a Python compiler for the PIC exists. It's called Pyastra and you
can find it here:

http://pyastra.sourceforge.net

They give this program as a sample of the blinky LED...

while(1):
   PORTA[0] = 1
 
   for i in xrange(0,255):
       pass
 
   PORTA[0] = 0

   for i in xrange(0,255):
       pass

Now in the interest of full disclosure, the author makes the following
disclaimer:

"Current versions are very unstable. Some code may be generated incorrectly.
So you shouldn't use it in critical tasks!"

Just passing on some information.

BAJ

2007\02\18@080532 by Gerhard Fiedler

picon face
Jinx wrote:

> Are you sure the information RS require is not just for their records ?
> That's my experience with RS in New Zealand. Whether I to register or
> order under a company name is irrelevant. However, I have been a
> customer of theirs for 20 years, maybe new customer criteria is
> different now, but I find that hard to believe. Looking at the
> equivalent page on http://www.rsnewzealand.com I see nothing that would stop me
> opening an account as a sole trader. I simply would not fill in the
> company section.

I've ordered from companies that needed something filled into the company
field -- but anything at all (like my name) was fine. (But that wasn't in
Sweden :)

I don't know about Sweden, but there are in many places "individual
companies" that are individuals with a license to work somewhat like a
company -- but usually without a proper company name. I've worked in this
way in Germany ("Selbständiger") and in the USA ("sole proprietor"), and
never had a problem ordering from such companies, without having a
registered company. Just try it...

Gerhard

2007\02\18@081149 by Jan-Erik Söderholm

face picon face
Rikard Bosnjakovic skrev/wrote:

> I'm located in Sweden / Europe, a small harbor town called Vänersborg.

Nice place... :-)

First, make sure you're a member of :
"elektronikforumet.com/forum/index.php".
A lot of nice (incl me) people there... :-)

For PIC related tools and chips, see :

http://www.jescab.se/Wisp628.html
http://www.jescab.se/Prod_PIC.html
http://www.jescab.se/Info_PIC.html

(Yes, that's my site...)

I've shipped over 150 Wisp628's in about a year and a half
localy i Sweden, and as far as I know, most are happy.

Best Regards,
Jan-Erik.

2007\02\18@090230 by Andrew Kieran

picon face

Forrest;
I use the EasyPIC4 extensively, so please allow me to comment on
your post.

> You can buy it domestically at
> http://www.circuit-ed.com/easypic4.htm  For $169

I really like the board, the programming software, and the
development tools.  And I my experience with Circuit-Ed was
fantastic!

> ... it's one of those boards which you can do a lot of
> things without having to resort to circuit building.

This is all true.  The board is extremely flexible. Using the
DIP switches, you can enable or disable all of the board's many
peripherals.

> You should be able to do all of the "R&D" on your
> project with this tool.   You can use the multiplexed
> LED to test out the display stuff.  You can use the
> buttons for the inputs.  And so on.

Correct again.  You can do an awful lot with everything that's
on the board.  And if necessary, you can run a small cable from
the connectors to a breadboard containing any additional
circuitry.

> One disadvantage I see is that the onboard programmer uses
> a proprietary interface which means that programming and ICD
> will need to be through the mikro tools.   All this means is
> that when programming, you'll need to use their software for
> the programming itself (using a .hex file from your chosen
> programming tool), and that the debugging part probably will
> only work with their compilers.  

You nailed it.  Fortunately, their PICFLASH programming tool is
simple, flexible, and reliable.  It, and the board, accomodate
just about all of the Flash-based PICs.  

But, as you noted, the ICD debugger only works with their
compilers.  You can program chips with code from Microchip's
MPLAB but there is no ICD capability when programming in
assembly language.

> It does look like the demo versions of their compilers will
> build small programs without restriction, and that they are
> priced reasonably.

The 2K code limit is pretty generous.  I bought the full C
Compiler about nine month ago, but I haven't installed it yet
since the demo is doing everything I need.

If you have any further questions about the board, I'd be happy
to (try to) answer them.

Andrew

________________________________________________
Get your own "800" number
Voicemail, fax, email, and a lot more
http://www.ureach.com/reg/tag

2007\02\18@155622 by Wouter van Ooijen

face picon face
> Oh, it's not THAT awful.  Maybe.  I certainly don't see a reason to
> not look at their literature; I haven't seen any negative impact from
> giving them (or ARM, where you'll need to sign up again to see the
> ARM manuals) my information.

I quickly read the datasheet, but I am not impressed. The LM3S101 costs
about the same as an LPC2101, which has much more pins and peripherals,
and has a full ARM cpu. For selling to my cutomsers that's too much
difference to justify the DIP package :(

List admins: maybe add an [ARM] topic?

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\18@163600 by William Chops Westfield

face picon face

On Feb 18, 2007, at 12:56 PM, Wouter van Ooijen wrote:

> but I am not impressed. The LM3S101 costs about the same as
> an LPC2101, which has much more pins and peripherals, and
> has a full ARM cpu. For selling to my cutomsers that's too
> much difference to justify the DIP package :(

It's not even DIP; just SOIC (though that's still more hobbyist
friendly than most.)  I think I gotta agree than the difference
between a $3 LPC2101 and a $1 3S101 isn't huge.  Although it's a
pain that the LPCs need multiple supplies.  The Cortex M3 core in
the Luminary part is supposed to outperform ARM7 "under many
circumstances"; it's "real" enough, it's just the explanations
and documentation that I find badly organized if this is your
"intro to ARMs."

BillW

2007\02\18@193835 by Wouter van Ooijen

face picon face
> It's not even DIP; just SOIC (though that's still more hobbyist
> friendly than most.)

I keep forgetteing that, but indeed the SOIC pitch is much more friendly
than the TQFP 48.

> I think I gotta agree than the difference
> between a $3 LPC2101 and a $1 3S101 isn't huge.

I don't care what the manufacturer says: I checked the prices with
http://www.findchips.com, which gives ~ $2 for both @ 100.

> The Cortex M3 core in
> the Luminary part is supposed to outperform ARM7 "under many
> circumstances";

But I guess not when running at 20 MHz and the ARM at 60.

> it's "real" enough, it's just the explanations
> and documentation that I find badly organized if this is your
> "intro to ARMs."

The LPC documentation is not that easy either.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\18@213750 by William Chops Westfield

face picon face

On Feb 18, 2007, at 4:38 PM, Wouter van Ooijen wrote:

>> It's not even DIP; just SOIC (though that's still more hobbyist
>> friendly than most.)
>
> I keep forgetteing that, but indeed the SOIC pitch is much
>  more friendly than the TQFP 48.
>
Yes; I have at "theoretical" PCB layout for the Luminary parts
that is single-sided and hobbyist-friendly.  My attempts to do
similar with the LPC are not going very well :-(

BillW

2007\02\20@113713 by Darrell Bellerive

flavicon
face
I too am new to the world of microcontrollers. Well, I have used one in a
protocol converter 20 years ago, but the world has changed so much that I
indeed am new to it again.

I've been lurking here on the list for a while and reading quite a lot on the
web, but I must admit that I am somewhat overwhelmed by the sheer number of
different microcontrollers.

Perhaps some kind souls here would be so nice as to point me in the right
direction. I am wondering if there are microcontrollers capable of ethernet
connectivity, TCP/IP, and functioning as a software audio codec for an audio
stream at a maximum of 64 kbps?

Darrell


--
Darrell Bellerive
Amateur Radio Stations VA7TO and VE7CLA
Grand Forks, British Columbia, Canada

2007\02\20@122737 by Wouter van Ooijen

face picon face
> Perhaps some kind souls here would be so nice as to point me
> in the right
> direction. I am wondering if there are microcontrollers
> capable of ethernet
> connectivity, TCP/IP, and functioning as a software audio
> codec for an audio
> stream at a maximum of 64 kbps?

you mean
- capable == it will do so without effort from your side, or
- capable == some hardware+software guru might get this to work,
eventually?

For TCP/IP I would go for a separate chip, like Microchip's ENC28. For
the software audio a 30F might do, but I would prefer an ARM.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\20@125555 by Darrell Bellerive

flavicon
face
On February 20, 2007 09:27 am, Wouter van Ooijen wrote:
> > Perhaps some kind souls here would be so nice as to point me
> > in the right
> > direction. I am wondering if there are microcontrollers
> > capable of ethernet
> > connectivity, TCP/IP, and functioning as a software audio
> > codec for an audio
> > stream at a maximum of 64 kbps?
>
> you mean
> - capable == it will do so without effort from your side, or
> - capable == some hardware+software guru might get this to work,
> eventually?

Somewhere in the middle. The hardware should be fast enough to handle the
tasks, and I would hope that freely available software libaries would also
exist to speed software development.

I know that I could accomplish these tasks with a Mini-ITX motherboard and
microprocessor running a Linux based system. The question is: Are
microcontrollers getting fast and complex enough to consider them as
alternatives?

So far from what I have read, the Mini-ITX/Linux system seems to be the better
choice, but I as I said, I am new to the microcontroller game, and don't want
to rule out microcontrollers based on my limited knowledge of them.

Darrell

--
Darrell Bellerive
Amateur Radio Stations VA7TO and VE7CLA
Grand Forks, British Columbia, Canada

2007\02\20@131321 by peter green

flavicon
face

> Perhaps some kind souls here would be so nice as to point me in the right
> direction. I am wondering if there are microcontrollers capable
> of ethernet
> connectivity, TCP/IP, and functioning as a software audio codec
> for an audio
> stream at a maximum of 64 kbps?
PIC18s are barely capable of TCP/IP and aren't going to stange a snowballs chance in hell of handling complex audio codecs.

Some arms could certainly do it but you may be more into the range of microprocessors than microcontrollers by the time you get to one that is fast enough.


2007\02\20@133604 by Maarten Hofman

face picon face
>
> PIC18s are barely capable of TCP/IP and aren't going to stange a snowballs
> chance in hell of handling complex audio codecs.
>
> Some arms could certainly do it but you may be more into the range of
> microprocessors than microcontrollers by the time you get to one that is
> fast enough.


During a Microsoft competition I believe someone managed to do MP3 on a
dsPIC. However, this would still be very tricky (I did some calculations,
and it is very tight). It is much easier to get the chip for the specific
codec and have the PIC control such a decoder (which is the solution I would
opt for... Although I'm not entirely certain what the problem is, yet).

2007\02\20@133751 by Herbert Graf

flavicon
face
On Tue, 2007-02-20 at 09:55 -0800, Darrell Bellerive wrote:
> Somewhere in the middle. The hardware should be fast enough to handle the
> tasks, and I would hope that freely available software libaries would also
> exist to speed software development.

The hardware is fast enough (maybe not a PIC necessarily, although some
of the higher end dsPICs might be able to do it, depending on how
complicated your decode is), the software libraries are iffy depending
on what you want to do.

> I know that I could accomplish these tasks with a Mini-ITX motherboard and
> microprocessor running a Linux based system. The question is: Are
> microcontrollers getting fast and complex enough to consider them as
> alternatives?

Depends. What reasons do you have for dropping a Mini-ITX system?
Weight? Price? Power? The sort of thing you describe CAN be done with
MCUs, but there will be a significant amount of effort required on the
software side, compared to using a PC. Off hand, a bigger ARM could
certainly handle what you describe, but the software to do it isn't
trivial.

> So far from what I have read, the Mini-ITX/Linux system seems to be the better
> choice, but I as I said, I am new to the microcontroller game, and don't want
> to rule out microcontrollers based on my limited knowledge of them.

You have to weigh your options based on your needs. The Linux system
will probably get you quickest to something working. The MCU solution
will surely be smaller, cheaper and lower power. The tradeoff is the
extra time it'll take you to get it working.

TTYL

2007\02\20@135608 by Wouter van Ooijen

face picon face
> It is much easier to get the chip for
> the specific
> codec and have the PIC control such a decoder (which is the
> solution I would
> opt for... Although I'm not entirely certain what the problem
> is, yet).

Codec libraries form most formats exists, so if sufficients MIPS ara
available a separate chip is not necessary. But that of course depends
on the format used, available library, compiler, etc.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\02\20@162121 by William Chops Westfield

face picon face

> I am wondering if there are microcontrollers capable of
> ethernet connectivity, TCP/IP, and functioning as a software
> audio codec for an audio stream at a maximum of 64 kbps?

Sure.  There are MIPS, ARM, PPC, and Coldfire chips that are
capable of doing all that.  Open up most any consumer wireless
router or IP phone and you'll see one there.  You may need
extra (external) RAM to do it all, though.  You might consider
100pin BGAs as beyond the edge of "microcontroller."

BillW

2007\02\20@164117 by peter green

flavicon
face

> Sure.  There are MIPS, ARM, PPC, and Coldfire chips that are
> capable of doing all that.  Open up most any consumer wireless
> router or IP phone and you'll see one there.  You may need
> extra (external) RAM to do it all, though.
imho if it needs external ram its no longer a microcontroller.


2007\02\20@171951 by Bob Axtell

face picon face
peter green wrote:
>> Sure.  There are MIPS, ARM, PPC, and Coldfire chips that are
>> capable of doing all that.  Open up most any consumer wireless
>> router or IP phone and you'll see one there.  You may need
>> extra (external) RAM to do it all, though.
>>    
> imho if it needs external ram its no longer a microcontroller.
>
>
>  
What? how about Serial EEPROM? or an SPI RAMTRON or EEPROM? or an SD card?
if you interface to these you no longer use a uP?

Am I missing the lions' share of this thread somehow?

--Bob

2007\02\20@232430 by Herbert Graf

flavicon
face
On Tue, 2007-02-20 at 15:20 -0700, Bob Axtell wrote:
> peter green wrote:
> >> Sure.  There are MIPS, ARM, PPC, and Coldfire chips that are
> >> capable of doing all that.  Open up most any consumer wireless
> >> router or IP phone and you'll see one there.  You may need
> >> extra (external) RAM to do it all, though.
> >>    
> > imho if it needs external ram its no longer a microcontroller.
> >
> >
> >  
> What? how about Serial EEPROM? or an SPI RAMTRON or EEPROM? or an SD card?
> if you interface to these you no longer use a uP?
>
> Am I missing the lions' share of this thread somehow?

I believe so. Peter was referring to external RAM used by the MCU/MPU as
program RAM. This is distinct from the memory types you mention.

It's a difficult line to draw, where an MCU becomes an MPU. Certainly
some MCUs CAN use external RAM to expand available RAM to their programs
(PICs even), but does that then make them MPUs?

Personally I'd have to agree with Peter, if a processor REQUIRES
external memory as it's RAM, then I'd consider that to be an MPU. A
processor that has SOME internal RAM can probably be considered to be an
MCU.

The distinction is mostly just words anyways these days, performance
wise the waters are alot muddier. TTYL

2007\02\21@012536 by William Chops Westfield

face picon face

On Feb 20, 2007, at 1:40 PM, peter green wrote:

>>  You may need extra (external) RAM to do it all, though.
> imho if it needs external ram its no longer a microcontroller.
>
Many of the chips I mentioned DO have internal RAM.  Just not
enough to do a credible job of doing a stream decode of a 64kbps
TCP stream to audio...  It MIGHT be doable; I just got a sample
of an NXP LPC23xx series ARM7, and they have up to 512k flash
(which should be plenty) and 58k of RAM (which is big by micro
controller standards, but tiny by the standards of most (ie
based on the original 4.3BSD) TCP implementations.)

BillW

2007\02\21@093913 by William Couture

face picon face
On 2/21/07, William Chops Westfield <EraseMEwestfwspam_OUTspamTakeThisOuTmac.com> wrote:
>
> On Feb 20, 2007, at 1:40 PM, peter green wrote:
>
> >>  You may need extra (external) RAM to do it all, though.
> > imho if it needs external ram its no longer a microcontroller.
> >
> Many of the chips I mentioned DO have internal RAM.  Just not
> enough to do a credible job of doing a stream decode of a 64kbps
> TCP stream to audio...  It MIGHT be doable; I just got a sample
> of an NXP LPC23xx series ARM7, and they have up to 512k flash
> (which should be plenty) and 58k of RAM (which is big by micro
> controller standards, but tiny by the standards of most (ie
> based on the original 4.3BSD) TCP implementations.)

If you're doing TCP/IP based on a "big-iron" implementation,
you will definately need more ram.  What you should be looking
at is a TCP/IP written from scrach for microprocessors.

http://www.iosoft.co.uk/tcplean.php

Bill

--
Psst...  Hey, you... Buddy...  Want a kitten?  straycatblues.petfinder.org

2007\02\21@100156 by David VanHorn

picon face
It used to be that the microprocessor had no internal ram or rom, and
usually no peripherals either.
(Z-80 as classic example)   Microcontrollers had onboard peripherals, and
usually had ram and/or program rom.

In the end, it's all marketing words. Like "RISC" whose limiting case is
a machine that executes NOPS really fast.

2007\02\21@141737 by William Chops Westfield

face picon face

On Feb 21, 2007, at 6:39 AM, William Couture wrote:

> What you should be looking at is a TCP/IP written from
>  scrach for microprocessors.
> http://www.iosoft.co.uk/tcplean.php
>
Is it any good?  What is the footprint and performance of a small TCP
implementation capable of streaming data at a reasonable rate (ie not
just one that provides a web-page now and then.)

It did occur to me that I wasn't well acquainted with the state of
the art in "small" TCP implementations, but I thought most of them
had sort of given up as "small" microcontrollers got "bigger", and
small devices got bigger computers.  (a cell-phone or PDA might be
a good example for the application suggested, but these days those
are 100+MHz CPUs with multiple megabytes of ram and flash.)  The
original cisco router code did a reasonable job in 1M of ram with
both code and data, and that was expected to handle several tens of
TCP connections, so I don't have any problem with believing it can
all be done

BillW

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...