Basic Stamp 1 and Microchip PIC16c56-XT/P
microcontrolleruser
Posts: 1,194
in BASIC Stamp
If you look on the Stamp here you can see Microchip chip number.
https://www.parallax.com/product/pbasic1_p
Here's the datasheet for PIC.
http://www.mouser.com/ds/2/268/30453E-278021.pdf
Full of interesting stuff.
Anybody know the process of how Parallax made one of these into a Stamp 1?
Seems like it's a giant bootloader
Comments
The PIC16c56-XT/P is now very much a trailing edge device, with a trailing edge price to boot.
If you pop over to Digikey and enter microcontroller and sort by price 1000+(active), you can find 41,622 items
Cheapest ones come sub 40c, including some quite capable ones with Serial peripherals & ADC, like EFM8BB10F2G, and STM8S003F3P6
I don't think Parallax is going to share their Intellectual Property on the Stamp line of chips.
If someone does want Source Basic, there are many flavours out there...
https://en.wikipedia.org/wiki/List_of_BASIC_dialects
https://en.wikipedia.org/wiki/Tiny_BASIC
http://efton.sk/basic/index.htm
etc...
Also for the Prop1, if anyone wants a Basic Compiler, is Bean's nice PropBASIC.
https://github.com/parallaxinc/PropBASIC
Compiles using FreePascal, and then compiles .pbas files to P1.
The PIC can be clocked at 40mhz.
Probably what makes that undoable with stamp is the software.
Looked at I/O register scheme in PIC '56 and it is unwieldy.
Don't know how much Parallax software affects hardware.
If you compare it to PIC10F200 the 200 is simpler.
I think they overhauled the 'PIC way' of I/O handling somewhere along the line.
What we were thinking is a Stamp/PIC using Parallax front end
programmable in Assembler and C.
I guess the achievable is a PIC with a bootloader
and just use mplabs.
Parallax would be the only one with such a product.
The could up it to a PIC 32bit.
Well, yes, the PIC '56 is the great grand-daddy of small MCUs.
Not quite sure what you mean here ?
Plenty of people use PIC32 in development boards.
Many MCU development boards support in-system Debug, and can be programmed in ASM & C.
Digilent has a 'single solution' and so does CSS Compilers.
Then there's Mighty mite and Picaxe.
Unfortunately Digilent dumbed there's down to Arduino.
Basically you just hook up board write your program and download.
Need a path to Assembler and C.
Looks like a stable bootloader like Microchip Ansi 851 is frontrunner right now.
If you want Debug, SiLabs, Microchip, ST, Nuvoton et al have compact, low cost boards, with USB-Debug bridges and IDEs to drive them included.
Cheapest USB bootloader-shipped I think is the EFM8UB1, comfortably sub $1
Just to make it clear.
I think Parallax/Chip Gracy made a wrong turn with the Propellor.
You'll paint yourself into a corner learning Multi core when
nobody else is adopting it. PIC and AVR are the biggies and I don't see any multi cores.
If the Arms are multi core then I'm mistaken. I don't pay attention to Arm's because
learning Stamp 1 and 2, pic's and avr's is enough to do.
It's not too late for Parallax to come out with another
'family' of micro's unless we've heard the last from Mr. Gracey.
I think a bootloading NON INTERPRETED Basic C and Assembler
would be a good one.
One click programming like Stamp.
I say keep it simple and just move up to 16bit PIC or 24F not all the way to 32bit.
32 bit would probably keep up with demands in developments better though.
.. Not at the 8-bit end, no.
A number of ARMS are multicore.
Many new WiFi/BLE parts use a dual core approach, and RaspPi is up to 4 cores.
This depends on the target price points, but certainly new parts like these, could support some solid features -
Silabs:
EFR32BG12 Blue Gecko SoCs feature a 2 Mbps Bluetooth PHY, providing ample throughput for applications running a Bluetooth 5-compliant stack. The Bluetooth 5 standard enables four times the range, twice the speed, 800 percent greater broadcasting capacity and improved co-existence with other wireless IoT protocols.
... EFR32xG12 SoCs offer four times more flash memory (up to 1024 kB with a dual-bank architecture) and eight times more RAM (up to 256 kB) than previous-generation Wireless Gecko devices.
Cypress ::
PSoC 6 dual-core ARM M4 and M0+
NXP::
K28, 150MHz M4, 2MB Flash 1MB (!) RAM
Nuvoton ::
M480 M4 core, 200MHz, with high speed comms peripherals.
256/512 KB zero-wait state Flash memory 128/160 KB RAM (with 32 KB hardware parity check)
SPI Flash controller with 32 KB cache memory, 100 MHz code execution
4 KB SPROM for secure boot 2 KB OTP for general purpose
https://www.embedded-world.de/en/ausstellerprodukte/embwld17/product-9863795/numicro-m480-series-microcontroller
Or, you could decide to complement the RaspPi etc, rather more than go head on, and instead target the small MCUs (Still well above PIC 56 in abilities)
Your original question had to do with how Chip transformed a PIC into a BASIC Stamp 1. I was surprised to still find Chuck McManis' 1994 deconstruction on his web site,
http://www.mcmanis.com/chuck/Robotics/stamp-decode.html
It is an exercise in analysis and deduction, to show how the Stamp 1 works its magic as an interpreter of tightly packed tokens. It can lead you to a greater appreciation of the Stamp 1 and the ingenuity behind it.
Which they kind of did with Stamp2P etc.
I like part in deconstruction article about Parallax I/O handling.
I knew something was amiss there.
Parallax seriously needs to run off some more Stamp2P's and above DIP's.
Either that or drop price on Modules to @ $29.95 or even better $19.95.
I think new dip's is healthier idea.
Stamp's are still very very useful for quick prototyping.
Parallax left the 'Serious Hobbyist' high and dry running out of
Stamp 2P's and above dip's.
$80 each for modules. What?!!!
Those things are going to sit in the warehouse until
California becomes a conservative state again.