Raspberry Pi Pico 2 W Launches For $7

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Grawp
    replied
    Errata, errata, errata. E9 bug. To anybody not especially interested and following embedded world but having basic electronics knowledge:

    When a pin is configured as input (regardless whether it is used as general purpose input or is part of some bus) it is generally expected to have *very* high impedance and not to take any current. Low voltage read as 0, high as 1, there are some margins and an undefined zone. Standard stuff.

    But on RP2350 when a voltage is somewhere in the middle (e.g. when input is changing) it starts leaking current *heavily* in or out until 2.2V is reached or you drive it hard low or high away from the leakage zone.
    This has massive consequences on HW design and also power consumption.
    Last edited by Grawp; 27 November 2024, 06:34 AM. Reason: typo

    Leave a comment:


  • gary7
    replied
    ari55 Yes that was a very interesting video. Thanks for the link. I have to say that I have never been a big fan of just about any Microsoft software (massive understatement). However it still surprises me that VS Code pulls in half baked tools for the RISC-V and even the ARM processors in the 2350. I have thought about using VS Code since it became available for Linux and I had read some very positive comments (normally a command line guy when it comes to programming tools), however unless there is some way to make it pull in better tool chains I think I'll give it a pass.

    justinkb, as far as I know there is no "fixed" version of the 2350, or even if they are working on a new chip stepping yet. They are going to have to fix it eventually. I'm fairly certain that this affects the 2350B too.

    Leave a comment:


  • ari55
    replied
    Originally posted by gary7 View Post
    The Pico 2 and 2W ARM processors have floating point. It looks like the RISC-V processors do not.
    This video is showing some interesting comparisons.
    Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.

    Leave a comment:


  • justinkb
    replied
    Does this reuse the bad rp2350s with the wrong part that leads to all kinds of weird unexpected behavior? Or is it a newer, fixed revision?

    Leave a comment:


  • gary7
    replied
    The Pico 2 and 2W ARM processors have floating point. It looks like the RISC-V processors do not. ESP32's have excellent user support and a nice product line so they might be the best choice depending on your application. Just did a quick check on the 2350 errata and it appears to be worse than what I originally read about. Definitely waiting for a new chip revision before I buy one. Weird problem to be having with IO pins. Very not good that it was not caught in chip testing. I assume that would be because they have been doing programmable function IO pins for years now but something was clearly changed and needed to be tested.

    Leave a comment:


  • willmore
    replied
    Originally posted by Developer12 View Post

    Every chip has errata, but few have errata this bad. No thank you, I think I'll sit this one out.
    Agreed, this is a big one. Maybe wait for a C stepping.

    Leave a comment:


  • Developer12
    replied
    Originally posted by colejohnson66 View Post

    Every Ax stepping is going to have errata. That's just the nature of silicon. The only difference is that Raspberry Pi tells you about it instead of quietly working around it in the SDK.
    Every chip has errata, but few have errata this bad. No thank you, I think I'll sit this one out.

    Leave a comment:


  • Spacefish
    replied
    I love the C6, it´s cheaper has a RISC-V CPU, Bluetooth as well as a Thread RF module and the WiFi module support 802.11ax (WiFi 6) which saves a lot of power with Target Wake Time.

    Via a Thread Boarder Router you can just du ipv6 networking via Thread with very low power (battery driven over multiple weeks / months).

    The ESP32-C5 once released will add 5GHz WiFi.

    ESP32 devboards have USB-C since years + you get an integrated high speed JTAG debugger via USB with multiple hardware breakpoints and such. Quite helpful when developing something while testing it on real hardware.
    Last edited by Spacefish; 25 November 2024, 04:24 PM.

    Leave a comment:


  • rclark
    replied
    Just ordered up 4 of 'em. Just like potato chips... I really don't see what people gripe about. These little boards are not very expensive, do the job, and support western technology (I've never ordered from ali) . I have a lot of fun with programming them for this or that. One of my side goals is to see about using the RISC-V cores with C/assembly. Of course if using python (circuit or mico) it doesn't matter what you are running on.

    For those that complain about no USB-C, just use the Pimoroni Pico 2 boards like the Tiny or Plus boards (have both). I've been using a Tiny for a few weeks now for fun. Works great.
    Last edited by rclark; 25 November 2024, 03:56 PM.

    Leave a comment:


  • ari55
    replied
    Originally posted by scottishduck View Post
    You seem to have missed that this new microcontroller also has 2 RISC-V cores on board and can be booted in “RISC-V mode” which is really cool.
    It's also possible to have one core running arm + one core running riscv.

    Leave a comment:

Working...
X