Truncated match.
PICList
Thread
'Extending memory'
1998\11\10@070926
by
aya Baptista
Hi all.
I think that I know the answer to this (_stupid_) question as I read the
datasheet:
is there any way (and I really mean ANY way) to extend program and data
memory on the 16F84? A few kilobytes extra would be extremely useful...
Thanks,
AndrŽ Malafaya Baptista
1998\11\10@113614
by
Peter L. Peres
On Tue, 10 Nov 1998, [iso-8859-1] André Malafaya Baptista wrote:
> Hi all.
> I think that I know the answer to this (_stupid_) question as I read the
> datasheet:
> is there any way (and I really mean ANY way) to extend program and data
> memory on the 16F84? A few kilobytes extra would be extremely useful...
Sure ! Stretch the chip in a vise then fill the cracks with epoxy and use
;)
Seriously, if you want a large programming space investigate interpreters
of simple languages running on a PIC with program storage in an external
EEPROM or several.
Disadvantages: speed, code security, 3rd party programming tools or high
degree of experience required to make it work. Advantages: too many to
quote.
Peter
1998\11\10@164138
by
uter van ooijen / floortje hanneman
NO, NO
At least no way which is transparant to the programmer.
For data space, you could attach any kind of RAM (serial RAM is often used)
and
use your own put / get routines.
For code space idem (EEPROM instead of RAM), but you would have to add an
interpreter.
regards,
Wouter.
----------
{Quote hidden}> From: AndrŽ Malafaya Baptista <
spam_OUTABaptistaTakeThisOuT
SE.EFACEC.PT>
> To:
.....PICLISTKILLspam
@spam@MITVMA.MIT.EDU
> Subject: Extending memory
> Date: Tuesday, November 10, 1998 12:55
>
> Hi all.
> I think that I know the answer to this (_stupid_) question as I read the
> datasheet:
> is there any way (and I really mean ANY way) to extend program and data
> memory on the 16F84? A few kilobytes extra would be extremely useful...
>
> Thanks,
> AndrŽ Malafaya Baptista
1998\11\10@173611
by
paulb
Peter L. Peres wrote:
> Seriously, if you want a large programming space investigate
> interpreters of simple languages running on a PIC with program storage
> in an external EEPROM or several.
Actually, it's been done before. They called it the "BASIC STAMP".
> Disadvantages: speed, code security, 3rd party programming tools or
> high degree of experience required to make it work.
> Advantages: too many to quote.
I agree. It's a totally fascinating concept. Well worth re-inventing
the wheel for (I mean you *could* go buy a BS-II but ...) and "rolling
your own".
--
Cheers,
Paul B.
1998\11\10@180240
by
William Chops Westfield
At the embedded systems conference, Parallax had their (predictable) BS2
based on the scenix chip (faster, more memory!!) Amoung those features were
visible connector space for the "key" that allows in circuit programming and
debugging, and the parallax rep commented that that meant you COULD, in
theory, buy stamp-2 format boards (with Scenix chip) and program them with
your own code...
BillW
1998\11\10@185440
by
Russell McMahon
|
1. Data capacity can be easily extended with standard IIC ( I
squared C) memories.
Access is slower.
Standard program memory can not be extended by "normal" means, but
...
2. You could interpret code stored in an external memory (along
the lines of PIC STAMP BASIC)
3. Seeing you say you "really mean ANY" then I'll suggest the
following.
This obviously includes extra ICs
This obviously allows extra cost
So
You could consider using a second PIC for key functions and
communicate with it by passing parameters via a fastish serial
transfer technique. Something (very) like SPI would take 3 or 4 pins
and take a few 10's of cycles per byte transferred. Something like
IIC or serial would take 2 pins and be potentially slower (but a
pseudo synchronous version of standard async transfer could be quite
fast).
This may sound like a silly suggestion (it is :-)) but could be of
serious use. I presume you are wanting to extend the 16F84's capacity
due to its ease of reprogramming and internal EEROM. If so, you could
develop suitable routines in the 16F84 and then, once operational,
could put them in the "slave" processor and have them talk via the
serial link. It need not be a 16F84. For this to be worthwhile the
remote routines would have to be computationally complex enough to
justify the slower access time.
I'm actually doing something like this as an interim solution :-)
A 16C71? sends A2D data to a 16F84 in async serial on one pin. Later
on I'll change to a std serial A2D.
Many "real world" examples exist. In their very old HX20 portable
computer Epson had 2 x 6802 cpu's which communicated similarly - one
looked after the "real time" data tape recorder control tasks.
4 New microchip products are (allegedly) coming with more flash
rom
Changing to another manufacturers product is, unfortunately, a highly
tempting solution when you run out of 16F84 capacity.
From: AndrŽ Malafaya Baptista <ABaptista
KILLspamSE.EFACEC.PT>
is there any way (and I really mean ANY way) to extend program and da
ta
memory on the 16F84? A few kilobytes extra would be extremely
useful...
More... (looser matching)
- Last day of these posts
- In 1998
, 1999 only
- Today
- New search...