Shop OBEX P1 Docs P2 Docs Learn Events
Smartpin Diagram (now with %P..P bit mode table) - Page 8 — Parallax Forums

Smartpin Diagram (now with %P..P bit mode table)

1234568»

Comments

  • Might I suggest that you use an O (letter O) and 0 (zero) that are more distinct from eachother in your chart Rayman?
  • evanhevanh Posts: 15,906
    edited 2020-05-29 11:01
    Ha, now you've highlighted that issue, I noticed Rayman has used "0"s in the ADC mode bits where "O"s are in Chip's original chart.
  • RaymanRayman Posts: 14,626
    edited 2020-05-29 13:12
    Ok, so you want it to look good and be correct at the same time :)
    How's this (Note: Fixed sub table fonts after posting and reposted):
    1277 x 921 - 131K
  • RaymanRayman Posts: 14,626
    edited 2020-05-29 15:43
    Maybe we should call the first group "Regular (Non-ADC, Non-DAC) Modes" ?
    Also, I'm not sure how I got two different Ohm symbols in there...
  • RaymanRayman Posts: 14,626
    I'm not sure I like the way the zeros look with this Jetbrains font... Should I change fonts?
  • RaymanRayman Posts: 14,626
    I think I like this font (Roboto) better...
    1270 x 924 - 164K
  • Maybe use the Parallax TTF font. The one that installs with Propeller Tool. All the tables and code snippets in the P1 manual use it.

    Slightly offtopic:
    The best monospace font I've used is probably Consolas. You know, the default one that comes with Windows. Depressingly boring, yet somehow looks better than everything else I've tried, especially at low resolution.
  • RaymanRayman Posts: 14,626
    Ok, here's with the Parallax font:
    1413 x 1041 - 199K
  • Cluso99Cluso99 Posts: 18,069
    edited 2020-05-29 21:35
    I prefer the robo fonts (ie 3 posts back)
    It's a document so it doesn't need to be the parallax font, which BTW, a number of people seem to have difficulty installing. Shame because I love to include diagrams using it with the source code :(
  • evanhevanh Posts: 15,906
    edited 2020-05-30 05:53
    Cluso99 wrote: »
    ... a number of people seem to have difficulty installing.
    Wasn't that for the old broken font only? It got fixed a long time ago I thought.

    PS: Here's the one I have installed
  • Cluso99Cluso99 Posts: 18,069
    I’ve never had a problem but others keep reporting problems every now and then. Not sure if it’s gone away.
    Then there’s issues with different compilers not recognising the extended characters in the source. I had this problem just a few days ago - may have been with pnut but can’t be sure.
  • Colored Smart-Pin block diagrams in this discussion include left-side labels:
    "Physical Even# Pin" (Even Pin, pins 0, 2, 4,...62), and
    "Physical Odd# Pin" (Odd Pin, pins 1, 3, 5,..63).

    This seems to indicate that the odd-numbered I/O pins, such as P1, P3, and so on, that I connect devices to have only those functions in the lower section of the diagram. Is that correct? Likewise for the even-numbered pins.

    I thought each Smart Pin had the complete functionality shown in the entire block diagram. The diagram, however, indicates differences in the circuits that connect to even pins and to odd pins.

    Would someone explain, please? --Jon
  • I think the only difference is for USB support with odd and even pins. Everything else is the same.

    Enjoy!

    Mike
  • evanhevanh Posts: 15,906
    edited 2020-09-18 04:38
    Mainly it's about the interconnect for supporting differential output using Other and input with PinA and PinB.

    And, yes, the USB smarts happens to be managed from the even Smartpin also. USB needs a differential pair so it was easiest to do it that way I presume.

  • @Rayman & @evanh Do you mind if I take your diagrams as inspiration for another experiment? If it works out, I'll give attribution, of course.
    I'm considering an attempt to make the smartpin routing and configuration something much more dynamic that someone can play with on a web page.

    What do you think?
  • RaymanRayman Posts: 14,626
    I don't mind...
  • evanhevanh Posts: 15,906
    edited 2020-11-10 04:01
    Sure thing, go for it. Sounds a little like one of those logic simulators.

    EDIT: Hehe, the schematic like diagrams that I did were done with Logisim. They were done as documentation but it also allowed me to verify the function by twiddling the inputs.

  • I'm thinking of taking the block diagram and pin descriptions as the base layer and adding a configuration selector for each set of pins.
    The idea is to visually illustrate the active routings and component level configuration to make it visually obvious.
    I have a concept in mind for how to do it, but the tools to implement cleanly and without extra complication it are still being sought. Early experiments show promise but I'm still skirmishing with tooling.
    Ideally it should be easy to throw onto a static hosting site like gh pages where we can all use it. First things first, though.
  • RaymanRayman Posts: 14,626

    @evanh Is this how you wanted it changed?

    1998 x 1800 - 201K
    1998 x 1800 - 268K
  • evanhevanh Posts: 15,906

    Excellent. Just remembered one more thing - Put an arrow head on the dashed line through the smartpin block to IN.

  • RaymanRayman Posts: 14,626

    Ok, how's this?

    1998 x 1800 - 268K
  • Evanh
    This sure helps with select A\B. If you don' mind I would like to include drawing in my notes document.
    Regards
    Bob (WRD)

  • evanhevanh Posts: 15,906
    edited 2021-11-01 03:51

    Bob, sure thing.
    Rayman, maybe give also change DAC label on the green line to DACbus or DACBUS or DAC_Bus. Whichever you like most. Just to make it clearer the line is the 8-bit data lines and not the analogue.

  • RaymanRayman Posts: 14,626

    Maybe like this?

    1998 x 1800 - 277K
  • evanhevanh Posts: 15,906
    edited 2021-11-01 16:35

    Yes, nice touch with the bit widths.

  • evanhevanh Posts: 15,906

    I've just updated the T-block functional schematic too. Corrected the logic for OUT override with smartpin configured. And also now uses %SSSSS labelling for smartpin mode field instead of the older %MMMMM.

    968 x 822 - 22K
  • evanhevanh Posts: 15,906
    edited 2022-04-20 09:44

    And matching M-block (was called P-block), just new labelling, no logic change:

    868 x 906 - 22K
  • evanhevanh Posts: 15,906
    edited 2022-10-01 03:03

    [doh! Spoke too soon. Didn't double check on the scope first]

Sign In or Register to comment.