New P2 boards - what would you like?
VonSzarvas
Posts: 3,448
@SaucySoliton mentioned a dream board in another thread. https://forums.parallax.com/discussion/comment/1561842/#Comment_1561842
Which begs the question... what would we all like for general use?
I wonder if consolidating popular add-on boards into one or two variants of a more complete "ready-to-go" P2 carrier (similar size to the Propeller Activity Board) would be appropriate now; such boards should be useful for system-builders, developers and lean into education. And probably be powered/programmed by USB-C, and have some QWIC/STEMMA/etc.. headers, rather than too many bulky P2 accessory headers?
Ideally a consolidated carrier should cost not much more than $30, to allow a bundle price point of ~$100 for the Edge + Board.
But what does everyone think / want / would buy ?
Comments
BTW... I mentioned the P2 Explorer in the other thread. That was designed with a right-angle P2 Edge socket, with USB-C, HDMI, couple QWIC ports, space for screw terminals, USB slaves, couple buttons/LEDs, and a prototyping area that would fit an optional mini breadboard, or just solder direct. It had two 8-port headers arranged so they could be stacked if someone wanted to grow a stack of edge modules and have an 8 or 16-bit bus between modules. No additional add-ons required to get started, but you could plug in 3rd party or Parallax accessories/sensors easily enough as required.
Similar to how the low-cost FLiP protoboard turned out- though not quite
But yeah, probably some influence!
Anyhow- seems like the P2 development concept could be moved onward a little now that addon board functions have matured, and most code/drivers are available. So is now the time to merge things, bring some fresh life, and if so.... how should they look ?
IMHO, if you want a widespread adoption from hobbysts, take the Arduino approach: a small board with the P2 and flash, nothing else except power and an USB connector for programming (maybe a led or two for the blink example), expose all pins on headers and have stackable "shields" for specialized features. Keep the cost down to the bare minimum.
I second that. I'm currently playing a bit with Arduino Mega 256 and the sensor shield for it. Not because it's so good or it's so cheap. It just came together with the controls I bought for my flight simulator and it's already supported by the software. So as long as I can live with it's limitations compared to the Propeller it saves me the work of writing new PC software.
We already have the KISS board that looks quite similar. Maybe we have to add a higher power voltage regulator and an oscillator so that it performs better with overclocking. But it has to stay small, simple and cheap. Not as cheap as the Arduino, but I think something below $50 should be possible which is some kind of "unconvenience threshold".
Something like FLiP2 ?
Or some other form factor?
Has a Flip2 ever been produced? I can't find anything with Google or the forum search, except this mention.
I think a FLIP with a P2 and all 64 IO pins brought out to single line 1/10" pinheaders would be too big. There were several experiments with 1/20" headers (see P2D2) but it never really prevailed because it can't be used in classic breadboards. That's why I provided two options for the KISS board:
1. occupy only one row of pins and limit the IOs to 32, or
2. occupy two rows and have P2EVAL compatible headers for accessory style addon boards
Explanation see this post
I did a rough FLiP2 back in the day. I think it spanned all but 2 rows of the breadboard on each side, and brought out 32io’s, with onboard stuff using the other ios.. Then I tried a vertical version that straddled just the middle of the breadboard. Then moved to the Edge module format.
I quite liked the F2, and got it more compact than other efforts I’d seen here. But pros and cons… like having signals forced over a breadboard… in many cases a decent socket/carrier arrangement would be needed anyway, which means the Edge concept should be more robust and lower cost to mfg.
About KISS board- I really like that too, and the pinout flexibility. With a 20mhz clock and maybe some more time for layout and power supply work it would be a nice plug and play package for prototyping.
I'd like a RP2350B or RP2354B (the ones with the most pins) + P2 + some other bits.
This would satisfy both the Python and MMBasic fans .
Happy to pay for the design
Craig
I third the Arduino approach, but at a glance it doesn't look like anyone's actually specifically said Arduino (and/or other popular form factors) compatible. There are zillions of shields already out there that people (including "prospective" P1/P2 users) that may have several of. It seems like it'd be better and cheaper to leverage that so people aren't forced to buy yet another footprint of something they might already have. Would an Edge and its connector fit in that space?
Personally, the P2 Accessory Header is the way I'm going for work. "All-in-one" projects generate about 5 minutes of excitement and then get sold off for dimes-on-the-dollar when the inventory sits on the shelf for too long. The P2AH approach allows for improvements and iterations of circuits that would be left alone if on an all-in-one PCB.
Example: We have a small, 1W mono amp that we use in our P1 laser tag products. One of the field employees asked if I could add a jack that would turn off the internal amp and route the single to an external amp. In a few hours I had a new design. I reviewed it the next day and sent the files to PCBWay. In under a week I had a new board for product prototyping. If all goes well, the new amp with x-jack will go into P2 products.
Think Lego. Build more bricks that customers can combine to create prototype circuits. And don't release any "bricks" without a working library and real-world example programs. IMO, that's the only way to attract programmers currently working with other devices.
A big issue for me is packaging. I'd like to build products that look like they came from a real manufacturer.
One of the best that I have come across is the Tibbo system. I have a bunch of their stuff but I don't like their main CPU. Their programming languages don't even have a ms timer.
BUT an identical motherboard, featuring the P2 would be something else.
They have beautiful enclosures and even cardboard boxes for the finished product. Lots of nice plugin modules "Tibbits".
Tibbit modules
Enclosures
Motherboard that needs a P2 alternative
Craig
Wonder if HDMI out along with power in and USB programming would be possible with USB-C connector...
My laptop only has USB-C, so using an accessory to provide HDMI output, USB input, and power input.
Think it's possible that that is all passive, but not sure...
Think there needs to be a second uSD, or at least the uSD moved to different pins.
Would be nice to use uSD and flash concurrently...
How about modules functionally equivalent to the edge in it's current variants with 4 sides castellated? It would probably fit in a little bigger than 2" X 2" with 100 mil spacing. easy to solder down on a perfboard for a one-off, and not terrible for use on a custom board. (Put it on the bottom and P2 for virtually zero footprint!)
I keep a dozen or so FLIPS in the cabinet at work.. turn 'em over in a couple of months or so. This would get at least that much play...
I'd like to see a clear reason behind this new board idea because:
For me, the ideal board that fits the general use purpose is, or rather was because it is no longer available, the KISS board. Also for me, the audio/video/retro gaming is by no means general use.
If we define the "general use" and the target group then there will be the right time for discussing the the features and the price.
Why not design the "has it all" gizmo and come up with the reasonable maximum price tag, announce it (the board and the resulting price) and then see what people say (likes, dislikes) ? It should reveal how much people are willing to pay for what features and how big the respective groups are for each type of board.
I'm liking this idea... If you take the edge and drop the dip switch, uSD, and leds, perhaps it could be small square with castellated sides. Maybe just two castellated sides? Or, maybe 4 is better for trace length.
I could go for that. But, pricewise, that would be the same as the Edge, I think...
Didn’t someone make a castellated format?
@Cluso99 did. I have one....very nice except that he didn't get all the pins to the perimeter.
Craig
Oops, no he didn't
There's the P2-STAMP: https://forums.parallax.com/discussion/174317/new-p2-module https://github.com/knivd/P2-STAMP
I have a sample of that, solid little contraption
Hi
Anyone remember the P2D2
https://forums.parallax.com/discussion/168645/p2d2-an-open-hardware-reference-design-for-the-p2-cpu/p1
Whatever happened to Peter?
Dave
Something like @Cluso99 board would be good, but with castellated edges, flash instead of uSD. One USB with FTDI chip onboard would be nice. Switching power supplies...
Not sure about FTDI though... Maybe PropPlug header and just power over USB...
Wonder if the relatively high price of current options is a problem…
How much could be saved by dropping down to 4 layers?
Wonder how low price could be if made as bare as possible…
He's flogging his Forth on the RP2040.
Think people expect to be able to connect and power via usb and do things…. Guess I’m thinking low cost might be another way to go…
Personally, contemplating board with either Trion T20 or Titanium Ti60 along with P2…
Yep. If you want to engage the average Joe or student market to grow the P2 demand moving forward, this would be needed. P2 is otherwise difficult to get into unless you're a somewhat experienced PCB designer/EE type and build your own custom boards. The main option now is to go with the P2-Edge which in itself is a pretty good board, but doesn't have a huge number of available devices to directly use with. You need to go down the somewhat pricey accessory header route and get extra PropPlugs and buy/design your own boards for that which ONLY work with that 12/24 pin interface "standard". The way I see it is that Arduino and Pi got the benefit of a large number of add-on boards becoming available with simple USB connections into PC's for programming it (or direct development on the Pi itself) and people could just plug and play and focus on software instead of hardware complexities. They didn't need to worry about HW design so much.
A more tolerant power input would be nice
A 5Volt power is hard to find except a usb plug.
More are 12Volt ++
I use on my stuffs an OKI-78SR-5/1.5-W36-C
Those are 78XX pining and go to 36Volt input
Or keep a place to insert one.
The Edge modules have- Voltage input requirements: 5 to 16 VDC
Would keeping that range suffice ?
Caveat: Not withstanding that with a larger "all in" PCB, it would be nice (or important?) to have an on-board 5V regulator for 5V accessories, regardless of VIN.
I still think a low cost universal board with the following features would be good:
Teensy 4.1 board is quiet ideal, although a P2 board will have to be a little bit broader.
It is more important to have the board available than to say, that it is not possible to have all 64ports at the edges.
Teensy 4.1 brings out 42 ports
https://www.pjrc.com/store/teensy41.html
ESP32 T8 have somewhat similar setup.
Teensy4.1 has similar power like P2, so it sets the price target. If necessary, add 10$ for low volume P2. =42$.
If you want new customers, the board has to be competitive.
#
I was a fan of the P1 project boards, and I did not understand, why Parallax dumped them. I have used some boards, which consumed them, so I bought additional ones for new projects, while I think, that the expensive edge board is reused for new projects....
A P2 project board would have similar features as mentioned above but has a prototype area and places to solder VGA, perhaps HDMI, USB, PS2
#
Of course Ada's emulators are most impressing, but I don't think that most applications need that top RAM amount and top speed.
Cheers Christof