Actions
Engagement
  • Author Author: rscasny
  • Date Created: Date Created
  • Last Updated Last Updated:
  • Views 3176 views
  • Likes 2 likes
  • Comments 35 comments
Related
Recommended

Poll: Why Haven't You Applied for this Roadtest: STM32H7B3I-DK -  DISCOVERY KIT

Summertime is one of those busy times of year where people spend quality time with family, friends, and significant others. But this year you may be spending your summer time closer to home. And what better way to spend your quality time than with an element14 roadtest!

 

Inn particular, this roadtest: STM32H7B3I-DK -  DISCOVERY KIT

 

There are so many good things I could say about this dev board (with display). But the thing that stands out for me is what you can do with this board given all the peripherals you can experiment with:  the board help to evaluate the following peripherals: USB HS OTG, microSDTm card, 8-bit camera interface, audio DAC stereo with audio jack input and output, 128-Mbit SDRAM memory, 512-Mbit Octo-SPI Flash memory, Wi‑FiRegistered module (802.11 b/g/n compliant), I2C extension connector, FD-CAN, 20-pin microphone MEMS connector with DFSDM interface, 4.3-inch TFT-LCD (480*272) using an RGB interface with a capacitive touch panel.

 

You can even take advantage of three graphical stacks running on the same board with many featured applications:

– Menu launcher

– Audio player TGFX application (MP3 support)

– Clock and Weather with Alarm feature

– Video player TGFX application (enabled)

– TGFX Graphic demonstration

– STemWin Graphic demonstration

– EWZ Graphic demonstration

 

Poll Question: What's holding you back from applying to this roadtest?

  • in reply to BigG

    We're getting a bit sidetracked from the STM32 road test, but Go doesn't seem particularly well suited to embedded. In fact I didn't even know it was used (other than the fact that everything's been done once by somebody). A quick Google took me here where it seems that blinky takes up half the flash of an nRF52840! Garbage collection, etc. may be great for desktop but doesn't really fit with embedded programming in my opinion. Still better than MicroPython, of course. Go seems to be a great language for other things.

  • in reply to Fred27

    Thanks for the insight. It sounds interesting. Have you considered embedded Go by any chance or looked at it as comparison.

  • in reply to BigG

    Rust definitely has potential to be C++ without all the security issues and memory leaks. What really tipped me over the edge is that Microsoft had a session on it a Build recently and are now using it in place of C++ for some things. I'm not sure about embedded Rust as an alternative to C but it seemed like a good way to learn Rust if nothing else.

  • in reply to Fred27

    Embedded Rust, hey? Is it worth looking at? Will it catch on... anything but Python is my motto...

  • in reply to rscasny

      wrote:

     

    In hindsight, I figured it would be challenging, given all the features delineated in the datasheet. But good point: a complex board is likely to result in a complex roadtest. I don't want a roadtest to be overly complex such that it becomes a burden for the roadtester. I am thinking that I need to work with the supplier a little better on products that might be more appropriate for a 60-day roadtest (for someone who has a full time job).

    I tried to address that in my application text by listing the functionality I would cover in a project, and which options I wouldn't review in the 60 days.

  • I'd like to thank everyone who has replied for the time they took in voting and/or commenting about this roadtest.

     

    These are very useful comments. I realize that there is still about 2 weeks left in the enrollment, but I kind of felt the applications were lower than they should be. You all gave good reasons why that could be the case. In hindsight, I figured it would be challenging, given all the features delineated in the datasheet. But good point: a complex board is likely to result in a complex roadtest. I don't want a roadtest to be overly complex such that it becomes a burden for the roadtester. I am thinking that I need to work with the supplier a little better on products that might be more appropriate for a 60-day roadtest (for someone who has a full time job). Let me think more about it more. Thank you for your input.

     

    Randall Scasny

    RoadTest Program Manager

  • The board shares a problem with many other dev boards - it's stuffed with goodies and that means it's only suitable for fun and not much use for work.

    I've looked at the chip and might use it, I have current projects using an STM32H7xxx as well as lots of other lower spec STM32xxx parts, so I know the basic

    processor well enough to choose a new part from a data sheet.

    If I wanted a board to try something specific with this chip I'd buy a Nucleo which comes much closer to the ideal of "processor and the least possible support stuff"

    with lots of break out connections.

     

    This board has a nice display, but my experience of nice displays on dev boards is that they are impossible to source for your own production - so if I want to

    do a design that needs a display I would rather choose my own.

     

    This kind of board comes and goes, so you wouldn't want to design the whole thing into a product.

     

    Finally - it only costs £62, so if I wanted one I could buy it, and blog about if I felt like it.

    To do a decent write up on it would take up time many, many times the value of the board - so you are only going to get any one to do it if they really fancy this particular

    combination of features  and don't mind road testing it for altruistic reasons.

     

    But you've got 14 takers already and more will pop up near the end - that doesn't seem too bad (to me) for the kind of part it is.

     

    MK

  • As others have said, it seems like this eval kit, which is rich in features, will require a roadtest plan with some degree of complexity. Some of us could be starting with ST, so we will require extra resources (time and effort) just to set up the toolchain, environment, etc. in order to get used to the platform and propose a roadtest plan that could show the value of the kit.

  • I realised a while ago that I had too many dev boards from different manufacturers. I don't have the time to be familiar with the toolchains, peripherals and quirks of them all.

     

    In particular I had a box of ST boards that I haven't touched for years. The toolchain isn't even installed and I believe has changed completely since I last used it. There's absolutely nothing wrong with ST, I just can't spread myself too thin. Having said that I recently decided to look at embedded Rust and one of those boards came in useful as all the examples are for the STM32F3DISCOVERY. I had to grab one back from the person I'd donated them to.