Do you see any future to the 8-bit MCU's?

Hi Vladimir,

On 7/22/2011 10:13 AM, Vladimir Vassilevsky wrote:
> Don Y wrote:
>> Do you need a 32b core in a mouse? Keyboard?
>
> Optical and/or wireless mouse/keyboard.

I was going to pull apart one of my cordless keyboards
just to see what they used inside.  Then, realized I would,
no doubt, end up with a lap full of little plastic keycaps,
etc.  :<

So, instead, I tore open a wireless "laser" mouse (Logitech
M/N: M-RAG97).  This is "powered" by an ST72F324LJ4T6 -- an
8 bit MCU with 16KB of flash in a LQFP44.

Granted, a keyboard has more "buttons" but I suspect the
mouse and keyboard have similar numbers of I/O's (since
the keys are undoubtedly organized in a matrix).

As for complexity of scanning that matrix vs. the handful of
buttons on the mouse (7 in this case, plus the wheel and
"laser"), I doubt there is anything that would require the
power of a 32b processor for the keyboard while the mouse
gets by with just an 8b controller.

>> Microwave oven?
>
> Unicode support for multi language user interface.

*If* your microwave has "language prompts", it's doubtful
that they would be stored in Unicode instead of just as
"segment/pixel maps".  It's a fixed vocabulary device
so you would only store what the device *needed* to interact
with the user.

(I suspect you could control most microwave ovens with
a *4* bit MCU.)

>> Controlling the power windows in your car?
>
> J1839 protocol stack.

J1939 isn't a heavy stack.  And, for a "door controller",
that device would only have to interact with the few controls
*in* that door (window up, window down, door lock, door unlock,
seat up/down/fwd/bwd) as well as the actuators (lock door,
unlock door, window motor up, window motor down, seat motor
up/down/fwd/bwd).

The "door controller" need not perform all the duties of
an ECU -- it just has to be able to interpret *commands*
to drive its actuators and report the status of the various
controls (assuming those actions aren't handled locally)

[IMO, this is how many SCADA applications are headed -- small,
*dumb* motes that act as multiplexors/demultiplexors to cut
down on wiring costs]

> Running (i.e., *in*)
>> your furnace or ACbrrr? As your "intelligent thermostat"?
>
> Multi language user interface, connectivity and compatibility with the
> different hardwares.

Again, the user interface is canned.  Your thermostat will
never be called upon to display "O Romeo, Romeo, wherefore
art thou Romeo?"

As for "hardware compatibility", a volt-free contact closure
(or two) suffices in most of the ~100M households in this
country.

>> Scanning credit cards for sales authorizations?
>
> Certified WAN-class TCP/IP stack and https.

You don't need a 32b MCU to get a TCP/IP stack.  And, many credit
card terminals use dialup *or* talk (serially) to a smart(er)
host that actually provides the transport protocol (e.g., a
PC acting as a cash register)

>> We see all these "big" applications (iPhones, etc.) with *huge*
>> volumes...
>
> What you mentioned are all big applications for 32-bitters.

<grin>  I guess you and I approach problems with considerably
different bits of iron!  :>

>> and forget that their actual volumes are *tiny* when
>> you think of all the other "non-glorious" things that are out there
>> "just doing their jobs"...
>
> Toys, smartcards, timers, RKE, sensors, battery maintainers,
> miscellaneous small controllers and other large volume standalone stuff
> is the area of 8- and 4-bit MCUs.

Don Y wrote:

> (I suspect you could control most microwave ovens with
> a *4* bit MCU.)

Although I agree with most that 8 bit isn't dead a microwave
oven is one case where a 32 bit cortex m3 might be an
appropriate choice. A typical microwave implementation
that I am familiar with uses 40-60K of 8 bit processor code
for oven control and separate micro for display and keyboard.

Over the last few years the oven control code has reached
the limits of 8 bit processors for a typical oven and requires
considerably more processing power  for a full featured
microwave oven.

Regards,


walter..
--
Walter Banks
Byte Craft Limited
http://www.bytecraft.com

Don Y <nowhere@here.com> writes:

> Hi Mark,
>
> On 7/25/2011 3:36 PM, Mark Borgerson wrote:
>> In article<j0ca7a$6ku$1@dont-email.me>, hamilton@nothere.com says...
>
>>>> We see all these "big" applications (iPhones, etc.) with *huge*
>>>> volumes... and forget that their actual volumes are *tiny* when
>>>> you think of all the other "non-glorious" things that are out there
>>>> "just doing their jobs"...
>>>
>>> LOL, remember that there is an two 8049s shipped with every computer
>>> ever sold !!!
>>>
>>> One is in the keyboard.
>>> The other is in the keyboard controller on the motherboard.
>>>
>>> So for every one million computers shipped, there are two million 8049s
>>> shipped.
>> A few questions:
>>
>> 1.  Are those real 8049s or are they  ASICS that emulate the 8049
>> behavior at the external interface?
>
> Nowadays, I think they are just interface emulation.  I.e.,
> there isn't even an 8049 *core* involved.  E.g., I haven't
> *ever* used an 804x in a keyboard interface.

I did a project once with them, involved arbitrary precision integer
math. Had to write my own assembler in basic on a BBC micro. Gah!

>> 2.  Does the 8049 handle USB and/or  Bluetooth?

Haha

[...]

>
> I shake my head every time I come across my old GI catalog and
> wonder how this stuff ever came to *be* -- let alone how (relatively)
> "successful"!

-- 

John Devereux

On 07/23/2011 04:56 PM, Don Y wrote:
> Hi Dimiter (and Hans-Bernhard),
>
> On 7/23/2011 1:51 PM, dp wrote:
>> On Jul 23, 11:42 pm, Hans-Bernhard Br&#4294967295;ker<HBBroe...@t-online.de>
>> wrote:
>>> On 23.07.2011 04:40, Don Y wrote:
>>>
>>>> On 7/22/2011 4:30 PM, Hans-Bernhard Br&#4294967295;ker wrote:
>>>>> And I'm pretty sure GG broke that on purpose. Mere negligence just
>>>>> doesn't explain the level of blatant incompetence they've repeatedly
>>>>> reached whenever make any change to their service.
>>>> Assuming Google does nothing without a *reason*, this begs the
>>>> question:
>>>> what do they gain by doing this? (since gain they must!)
>>>
>>> I rather suspect their long-term master plan is to poison Newsgroups as
>>> a medium, to a point where people will flock to their silly imitation
>>> web page, where they get to sell advertising.
>
> <frown> Possibly...
>
>> Sounds plausible to me, probably that's the plan. The old google
>> groups which I use do a pretty decent job of threading ("like a tree"
>> etc), but they are pushing the "new" google groups which are by far
>> not as usenet friendly (I think; I only remember I looked at it and
>> discarded it without hesitation, but I don't really remember the reason).
>
> But why *use* google groups instead of a "real" news service?
>
> (sorry, I don't use anything but google search -- and that, DECREASINGLY
> so...)

I'm using Clear at the moment. Previously TW, previously ATT. None offer 
NNTP anymore (some time now).

So, it's just one more expense to have real NNTP, but I found out, after 
some effort, that low byte-count NNTP is real cheap. Just another 
account and password, like you don't have too many already.

Hi Walter,

On 7/27/2011 2:01 AM, Walter Banks wrote:
> Don Y wrote:
>
>> (I suspect you could control most microwave ovens with

--------------------------------^^^^^^^^^^^^^^^^^^^^

>> a *4* bit MCU.)
>
> Although I agree with most that 8 bit isn't dead a microwave
> oven is one case where a 32 bit cortex m3 might be an
> appropriate choice. A typical microwave implementation
> that I am familiar with uses 40-60K of 8 bit processor code
> for oven control and separate micro for display and keyboard.

Is this a "typical microwave oven"?  Or, just one that you are
familiar with?

My *toaster* has all sorts of silly blinking lights.  But, it
really only needs a bimetallic strip to open the flow of power
when the toast reaches a certain (approximate) temperature.
(for the life of me, I can't understand why it needs to *beep*
to signal the toast is ready... isn't the sound of it being
ejected enough of a signal??)

I recall seeing a microwave oven with a web browser/email client
built into the front door (seems like an ideal place for a
display in a kitchen!).  But, we must live in a poorer part of
town as none of *our* neighbors have such a capability.  :-/

A few minutes with google suggests worldwide production of
microwave ovens is at least 50M/annum.  How many of those,
do you think, need/have that level of control sophistication?

I'll have to start making mental notes of the features present
in the microwave ovens that I encounter (neighbors, businesses,
etc.).  A neighbor owns one of the larger appliance stores -- maybe
I'll start with him...

> Over the last few years the oven control code has reached
> the limits of 8 bit processors for a typical oven and requires
> considerably more processing power  for a full featured
> microwave oven.

On 7/27/2011 9:51 PM, bbhack wrote:
> On 07/23/2011 04:56 PM, Don Y wrote:

>> But why *use* google groups instead of a "real" news service?
>>
>> (sorry, I don't use anything but google search -- and that, DECREASINGLY
>> so...)
>
> I'm using Clear at the moment. Previously TW, previously ATT. None offer
> NNTP anymore (some time now).

There are some "free" NNTP services (with limitations).

Locally, smaller providers (typically those with UN*X services)
will often give you access to an NNTP -- if not a complete *feed*
(as well as shell access, etc.)

> So, it's just one more expense to have real NNTP, but I found out, after
> some effort, that low byte-count NNTP is real cheap. Just another
> account and password, like you don't have too many already.

<frown>  Yeah, something needs to be done about this:

"Never use the same password for different accounts"
"Never write down your password"
"Never use an easy to guess (remember!) password"
"Never use less than 14 characters containing a mix of
    alpha, numeric, punctuation and differing cases."

OK, yet every financial institution, internet service,
retailer, etc. wants me to set up a *separate* account
with them... just how much of this am I supposed to
commit to memory?  :>

[I am actually amused at just how much of this cruft I
*can* keep in my head!]

Don Y wrote:

> Hi Walter,
>
> On 7/27/2011 2:01 AM, Walter Banks wrote:
> >  A typical microwave implementation
> > that I am familiar with uses 40-60K of 8 bit processor code
> > for oven control and separate micro for display and keyboard.
>
> Is this a "typical microwave oven"?  Or, just one that you are
> familiar with?
>

I am familiar with two companies microwave oven product lines. Both are
large companies with a range of models and features. Both ship products
world wide.  My typical comment was a statistical mean of the products
they ship. I am familiar with both their low end product and high end
products. A the low end software has few user features but still
requires oven control for reasonable product life and safety. The high
end models implement indirect instrumentation to measure contents and
data bases of cooking profiles.

Regards,


w..
--
Walter Banks
Byte Craft Limited
http://www.bytecraft.com

On Thu, 28 Jul 2011 01:59:13 -0700, Don Y <nowhere@here.com> wrote:

>On 7/27/2011 9:51 PM, bbhack wrote:
>> So, it's just one more expense to have real NNTP, but I found out, after
>> some effort, that low byte-count NNTP is real cheap. Just another
>> account and password, like you don't have too many already.
>
><frown>  Yeah, something needs to be done about this:
>
>"Never use the same password for different accounts"
>"Never write down your password"
>"Never use an easy to guess (remember!) password"
>"Never use less than 14 characters containing a mix of
>    alpha, numeric, punctuation and differing cases."
>
>OK, yet every financial institution, internet service,
>retailer, etc. wants me to set up a *separate* account
>with them... just how much of this am I supposed to
>commit to memory?  :>
>
>[I am actually amused at just how much of this cruft I
>*can* keep in my head!]


While a global authentication scheme has its attractions (and its
drawbacks), in the interim something like Password Safe can go a long
way to making the multitude of accounts and passwords problem
tractable:

http://passwordsafe.sourceforge.net/

On Thu, 28 Jul 2011 06:44:46 -0500, Robert Wessel
<robertwessel2@yahoo.com> wrote:

>On Thu, 28 Jul 2011 01:59:13 -0700, Don Y <nowhere@here.com> wrote:
>
>>On 7/27/2011 9:51 PM, bbhack wrote:
>>> So, it's just one more expense to have real NNTP, but I found out, after
>>> some effort, that low byte-count NNTP is real cheap. Just another
>>> account and password, like you don't have too many already.
>>
>><frown>  Yeah, something needs to be done about this:
>>
>>"Never use the same password for different accounts"
>>"Never write down your password"
>>"Never use an easy to guess (remember!) password"
>>"Never use less than 14 characters containing a mix of
>>    alpha, numeric, punctuation and differing cases."
>>
>>OK, yet every financial institution, internet service,
>>retailer, etc. wants me to set up a *separate* account
>>with them... just how much of this am I supposed to
>>commit to memory?  :>
>>
>>[I am actually amused at just how much of this cruft I
>>*can* keep in my head!]
>
>
>While a global authentication scheme has its attractions (and its
>drawbacks), in the interim something like Password Safe can go a long
>way to making the multitude of accounts and passwords problem
>tractable:
>
>http://passwordsafe.sourceforge.net/

There's also http://keepass.info/, another FOSS password keeper. A nice
feature (that may be in Password Safe but I've never used that one) is
that it can be set to clear the clipboard after X seconds, so that a
copy-n-paste operation to use an account's password isn't vulnerable to
a clipboard sniffer (or accidentally pasting the password somewhere else
in the clear fifteen minutes later).

-- 
Rich Webb     Norfolk, VA

On Thu, 28 Jul 2011 08:04:11 -0400, Rich Webb
<bbew.ar@mapson.nozirev.ten> wrote:

>On Thu, 28 Jul 2011 06:44:46 -0500, Robert Wessel
><robertwessel2@yahoo.com> wrote:
>
>>On Thu, 28 Jul 2011 01:59:13 -0700, Don Y <nowhere@here.com> wrote:
>>
>>>On 7/27/2011 9:51 PM, bbhack wrote:
>>>> So, it's just one more expense to have real NNTP, but I found out, after
>>>> some effort, that low byte-count NNTP is real cheap. Just another
>>>> account and password, like you don't have too many already.
>>>
>>><frown>  Yeah, something needs to be done about this:
>>>
>>>"Never use the same password for different accounts"
>>>"Never write down your password"
>>>"Never use an easy to guess (remember!) password"
>>>"Never use less than 14 characters containing a mix of
>>>    alpha, numeric, punctuation and differing cases."
>>>
>>>OK, yet every financial institution, internet service,
>>>retailer, etc. wants me to set up a *separate* account
>>>with them... just how much of this am I supposed to
>>>commit to memory?  :>
>>>
>>>[I am actually amused at just how much of this cruft I
>>>*can* keep in my head!]
>>
>>
>>While a global authentication scheme has its attractions (and its
>>drawbacks), in the interim something like Password Safe can go a long
>>way to making the multitude of accounts and passwords problem
>>tractable:
>>
>>http://passwordsafe.sourceforge.net/
>
>There's also http://keepass.info/, another FOSS password keeper. A nice
>feature (that may be in Password Safe but I've never used that one) is
>that it can be set to clear the clipboard after X seconds, so that a
>copy-n-paste operation to use an account's password isn't vulnerable to
>a clipboard sniffer (or accidentally pasting the password somewhere else
>in the clear fifteen minutes later).


Password Safe is similar.  It locks the database (so you have to enter
the database password again) and clears the clipboard after the
configured interval, and usually clears the clipboard after you
minimize the app too.

I've not used Keepass, but there are a lot of nice features in PWS,
and I've used it for years.  I don't think I could live without
something like that - I've got nearly 300 entries in mine.