Microcontroller Family Selection Guide

Jim Granville wrote:
> 
> Winbonds' W89E series also have OCD

oops - typo, for anyone searching it is

Winbond's W79E series also have OCD
[W79E201, W79E5xx, W79E8xx LPC series ]
Digikey have these as 'bulk order' items.

and I should add the new NXP P89LPC652/P89LPC954 also have OCD

-jg

On Tue, 06 Nov 2007 10:33:59 -0800, I said, "Pick a card, any card"
and Eric <englere_geo@yahoo.com> instead replied:

>Thanks to all for the helpful comments. I agree with several posters
>that a Wiki would be a good way to let various people contribute. I
>don't have any experience setting them up and I'd appreciate it if
>someone could help set one up.

Go here:
http://en.wikipedia.org/wiki/Main_Page

Create an account and set up your page. From the Babel-like comments
in the responses here, your best option is a matrix with comments
allowed.
--
Ray

On Nov 5, 12:56 am, Eric <englere_...@yahoo.com> wrote:
> I wrote up some of my thoughts about various microcontroller families.
> My intent is to mostly to help newbies and students decide what type
> of microcontrollers to start with.
>
> http://www.ericengler.com/downloads/microcontroller%20family%20select...
>
> Feel free to correct any possible errors, but I have a right to keep
> my opinions unless you can change them by presenting new information.

At risk of piling on, I think you have an error in your description of
the Arm Cortex M3.  You say "all instructions are 16 bit
instructions", which I don't think is correct.  I believe they are all
Thumb instructions including both the original Thumb and the Thumb2
instructions, but some of the Thumb2 instructions are 32 bit. I
believe this is organized with a 16 bit "prefix" and a 16 bit
instruction similar to some of the x86 instructions.

Please correct me if *I* am wrong... :^)

rickman wrote:
> At risk of piling on, I think you have an error in your description of
> the Arm Cortex M3.  You say "all instructions are 16 bit
> instructions", which I don't think is correct.  I believe they are all
> Thumb instructions including both the original Thumb and the Thumb2
> instructions, but some of the Thumb2 instructions are 32 bit. I
> believe this is organized with a 16 bit "prefix" and a 16 bit
> instruction similar to some of the x86 instructions.
> 
> Please correct me if *I* am wrong... :^)

That's correct (you for instance in Section 1.2 of DDI0338G).


			Laurent

On Mon, 05 Nov 2007 09:28:06 -0800, steve <bungalow_steve@yahoo.com>
wrote:

>On Nov 5, 12:56 am, Eric <englere_...@yahoo.com> wrote:
>> I wrote up some of my thoughts about various microcontroller families.
>> My intent is to mostly to help newbies and students decide what type
>> of microcontrollers to start with.
>>
>> http://www.ericengler.com/downloads/microcontroller%20family%20select...
>>
>> Feel free to correct any possible errors, but I have a right to keep
>> my opinions unless you can change them by presenting new information.
>>
>> There are many families I have never tried so I did not try to comment
>> on them.
>>
>> I want to follow this up with another document that points out some
>> good developer boards, ISP programmers, and JTAG debuggers for the
>> various families.
>>
>> Eric
>
>Nice job, your always going to get complains when you hit a nerve of
>someone's "love child"
>
>anyway, newbies typical uses small processors to do floating point, so
>here is a modified list I posted a while back to give newbies a
>ballpark estimate what they can expect
>
>
>MPS430, 32 bit floats, imagecraft complier, typical cycles
>add 158
>sub 184
>mul 332
>div  620

><snip>

I remember your post.  Thanks!  It was around the time I decided to
revisit some 32-bit fp-div code I'd written in 2004 and around the
time that Wilco pressed me on some issues I'd not handled well enough.

The final version I have has been posted up to the yahoo msp430 forum
in the files section and achieves 255 cycles (with call/return
included) for an FP 32-bit division on the MSP430.  (+/-1 cycle.)
About 2.4X.

It is free for anyone's use and if interested, you can siphon it also
at:

http://www.infinitefactors.org/jonk/zips+src/FP32DIVF.S43

There is a somewhat slower, but also smaller version at:

http://www.infinitefactors.org/jonk/zips+src/FP32DIVS.S43

I'm just reminding folks to keep in mind that not every library
routine used by every compiler vendor across families of cpus are
comparably optimal -- some library routines are much worse than others
and at least a part of whether or not they are better will depend a
little on just how valuable and competitive the marketplace is.

So keep it in mind when examining cycle counts.  The MSP430 may seem
like a 620 cycle fp divide situation or it might look like 255 cycles,
depending.  But it is the same cpu, either way, despite the 2.4X
apparent difference.

Jon

On Nov 6, 3:44 pm, Jim Granville <no.s...@designtools.maps.co.nz>
wrote:

> The PIC32 may yet be similar, as Microchip are talking about screen
> PICtails plug in cards.

Are they going to have any DIP PIC32's? Microchip is normally very
friendly to DIP packages so I hope so. Of course we'd have to clock it
down to use it in a proto board.

> Cost of tooling ?! - is ~$40 hard to mitigate :)
>
> http://www.freescale.com/files/microcontrollers/doc/fact_sheet/USBSPY...

I didn't know about the spider thing. I only looked at the RS08 in the
early days after it was announced.

> For this you get a Free C Compiler, USB debug, and RS08 and HC08 support

Are you saying CodeWarrior's C compiler for RS08 and HC08 is free and
has no size restrictions? I didn't know that and it would be very
important if this is the case.

Now why don't they make a reasonably strong hc08 in a DIP? They lost a
lot of students and hobbiests in the early days of migrating from hc05
to hc08 because of this lack

> There are thousands of RS08 on the shelf in catalogue companies, in DIP8.

certainly a good thing, especially if the C compiler is free without
strings.

> To me that all adds up to VERY student accessible, and a simple core is
> not a bad place to start learning how these things actually work.

But it needs a good cross-section of on-chip peripherals in order to
be good for students. If it's too weak there won't be many good
lessons there. That's why most Dev boards have strong processors in a
given family.

> At the Student end of the scale, the focus has moved partly from Chips,
> to Modules & Debug Access - see the TI and Silabs latest low-cost
> solutions of a 'dock-able' USB debug, and Freescale's, Infineon's and
> ST's  USB sticks (etc).

I understand the convenice and low price of the USB dongles but I'm a
little concerned about this trend. I think it's important for students
to mess with chips, wires, and proto boards. But the USB dongles that
have docking boards can fill that gap.

> The new PIC32 goes to 64K, and most cores have some form of open source
> compiler, that may not be the tightest code, but does work.

A little early for me to jump on these, but I really loved the PIC24,
aside from it's somewhat high current demand.

> Zilog have free, and unlmited tools (they bought a compiler company)
> Zilog also have a opto isolated USB debug, good for power development
> work. Zilog also have good simulators.

They just scared me about being in and out of bankruptcy over the
years.

> Then there is the Propellor chip -http://www.parallax.com/propeller/media.asp

This is a difficult chip to teach with because of the learning curve.
Maybe it's the wave of the future? I'll wait longer before deciding on
that.

Eric

Eric wrote:
> On Nov 6, 3:44 pm, Jim Granville <no.s...@designtools.maps.co.nz>
> wrote:
> 
> 
>>The PIC32 may yet be similar, as Microchip are talking about screen
>>PICtails plug in cards.
> 
> 
> Are they going to have any DIP PIC32's? Microchip is normally very
> friendly to DIP packages so I hope so. Of course we'd have to clock it
> down to use it in a proto board.

Unlikely, with the pin counts of the first releases. (maybe later)

If you want a Plug-able high end core, the best is probably the Z16F,
as that does have a PLCC package option. [has 32.32->64 opcodes]


>>Cost of tooling ?! - is ~$40 hard to mitigate :)
>>
>>http://www.freescale.com/files/microcontrollers/doc/fact_sheet/USBSPY...
> 
> 
> I didn't know about the spider thing. I only looked at the RS08 in the
> early days after it was announced.
> 
> 
>>For this you get a Free C Compiler, USB debug, and RS08 and HC08 support
> 
> 
> Are you saying CodeWarrior's C compiler for RS08 and HC08 is free and
> has no size restrictions? I didn't know that and it would be very
> important if this is the case.

My understanding is the RS08 compiler is free for all RS08 (since it's a 
16K
Max core, with only 1K shipping), the HC08 compiler IIRC is
free to 16K? and the Coldfire V1 is free to 128K?

> 
> Now why don't they make a reasonably strong hc08 in a DIP? They lost a
> lot of students and hobbiests in the early days of migrating from hc05
> to hc08 because of this lack

DIP is trailing edge technology. Silabs have one DIP20 offering, and I
see Philips and Atmel have PLCC32 packages as their 'plugable' offering
of what are otherwise all TSOP or QFP 'volume' devices.

Then there are always 'carrier modules'...

> 
>>There are thousands of RS08 on the shelf in catalogue companies, in DIP8.
> 
> 
> certainly a good thing, especially if the C compiler is free without
> strings.
> 
> 
>>To me that all adds up to VERY student accessible, and a simple core is
>>not a bad place to start learning how these things actually work.
> 
> 
> But it needs a good cross-section of on-chip peripherals in order to
> be good for students. If it's too weak there won't be many good
> lessons there. That's why most Dev boards have strong processors in a
> given family.

Depends on the depth of the course. The RS08 could reach down into
high schools, and introductory courses.
If the tools are easy, it can also be used to show students what's
involved in using more than one core - since it is quite likely students
will work on cores NOT the same as they are first taught, showing them
the jump is a good thing :)

Freescale do seem to be slow releasing the 'next' RS08 members.
The Vreg and Osc details are impressive, so more variants would
make users happier.

> 
>>At the Student end of the scale, the focus has moved partly from Chips,
>>to Modules & Debug Access - see the TI and Silabs latest low-cost
>>solutions of a 'dock-able' USB debug, and Freescale's, Infineon's and
>>ST's  USB sticks (etc).
> 
> 
> I understand the convenice and low price of the USB dongles but I'm a
> little concerned about this trend. I think it's important for students
> to mess with chips, wires, and proto boards.

Yes, building something that works, is cornerstone to embedded design

> But the USB dongles that
> have docking boards can fill that gap.

Exactly. 0.1" headers can serve that task.
Also the Silabs USB splits on a PCB edge connector, so add one
of those and you have the same USB debug.


> 
>>The new PIC32 goes to 64K, and most cores have some form of open source
>>compiler, that may not be the tightest code, but does work.
> 
> 
> A little early for me to jump on these, but I really loved the PIC24,
> aside from it's somewhat high current demand.
> 
> 
>>Zilog have free, and unlmited tools (they bought a compiler company)
>>Zilog also have a opto isolated USB debug, good for power development
>>work. Zilog also have good simulators.
> 
> 
> They just scared me about being in and out of bankruptcy over the
> years.

True, but their documentation is quite good, as are the tools.
Their cores are also designed from the ground-up as Microcontrollers,
so avoid some of the knee/bottlenecks in other cores.


> 
>>Then there is the Propellor chip -http://www.parallax.com/propeller/media.asp
> 
> 
> This is a difficult chip to teach with because of the learning curve.
> Maybe it's the wave of the future? I'll wait longer before deciding on
> that.

It is certainly different, but that makes it interesting :)

-jg

On Nov 5, 12:56 am, Eric <englere_...@yahoo.com> wrote:
> I wrote up some of my thoughts about various microcontroller families.
> My intent is to mostly to help newbies and students decide what type
> of microcontrollers to start with.

I updated my paper with some of the corrections you folks pointed out.
I appreciate your help. I did not info on new families aside from my
"first thoughts" on the PIC32.

As time allows I want to do another paper on development tools for
newbies. The main problems newbies have is to determine what type of
board(s) to buy, what type of software to install, and how to program
and debug the devices. The answers to these questions vary a lot by
family so that paper could line up with this one to some extent.

I'm not going to start a Wiki myself but anyone can use my material if
they want to make one, and I'll try to help by adding new content
also.

Eric
Microcontroller Family Selection Guide:
http://www.EricEngler.com/downloads/Microcontroller%20Family%20Selection%20Guide.pdf

"Eric" <englere_geo@yahoo.com> skrev i meddelandet 
news:1194905516.739123.253960@v65g2000hsc.googlegroups.com...
> On Nov 5, 12:56 am, Eric <englere_...@yahoo.com> wrote:
>> I wrote up some of my thoughts about various microcontroller families.
>> My intent is to mostly to help newbies and students decide what type
>> of microcontrollers to start with.
>
> I updated my paper with some of the corrections you folks pointed out.
> I appreciate your help. I did not info on new families aside from my
> "first thoughts" on the PIC32.
>
You may want to add that the AVR can, if running from the internal R/C 
oscillator
start from power down is less than 1 us.
Significantly better than the 7 us, quoted for the MSP430.

Also, the AVR can normally be run at a lower voltage than the MSP430
which affects the power consumption.
The AVR can be programmed with Vcc downto 1,8V, some downto 1.62
even if that is not reflected in the official datasheet.

Think you need to dig into the Brown-Out stuff as well.
Some of the MSP parts are quite weak here.

> As time allows I want to do another paper on development tools for
> newbies. The main problems newbies have is to determine what type of
> board(s) to buy, what type of software to install, and how to program
> and debug the devices. The answers to these questions vary a lot by
> family so that paper could line up with this one to some extent.
>
> I'm not going to start a Wiki myself but anyone can use my material if
> they want to make one, and I'll try to help by adding new content
> also.
>
> Eric
> Microcontroller Family Selection Guide:
> http://www.EricEngler.com/downloads/Microcontroller%20Family%20Selection%20Guide.pdf
>


-- 
Best Regards,
Ulf Samuelsson
This is intended to be my personal opinion which may,
or may not be shared by my employer Atmel Nordic AB 

On Nov 12, 6:43 pm, "Ulf Samuelsson" <u...@a-t-m-e-l.com> wrote:
> "Eric" <englere_...@yahoo.com> skrev i meddelandetnews:1194905516.739123.253960@v65g2000hsc.googlegroups.com...> On Nov 5, 12:56 am, Eric <englere_...@yahoo.com> wrote:
> >> I wrote up some of my thoughts about various microcontroller families.
> >> My intent is to mostly to help newbies and students decide what type
> >> of microcontrollers to start with.
>
> > I updated my paper with some of the corrections you folks pointed out.
> > I appreciate your help. I did not info on new families aside from my
> > "first thoughts" on the PIC32.
>
> You may want to add that the AVR can, if running from the internal R/C
> oscillator
> start from power down is less than 1 us.
> Significantly better than the 7 us, quoted for the MSP430.
>
> Also, the AVR can normally be run at a lower voltage than the MSP430
> which affects the power consumption.
> The AVR can be programmed with Vcc downto 1,8V, some downto 1.62
> even if that is not reflected in the official datasheet.
>
> Think you need to dig into the Brown-Out stuff as well.
> Some of the MSP parts are quite weak here.

The newest msp430's are better in a number of categories over their
predecessors, but I'm guessing the Picopower devices may indeed be
better in certain situations. I went to the Picopower seminar and I
want to start working with these soon.

I'll be working on the tools paper next. Does Atmel or any other
company mentioned in my paper have special discounts for colleges or
students? Educators usually have tight budgets. I work with professors
sometimes so I know that price often drives their board and tool
choices.

Eric