weact

Anything not related to STM32
Post Reply
ag123
Posts: 1785
Joined: Thu Dec 19, 2019 5:30 am
Answers: 28

weact

Post by ag123 »

I'm not too sure if this is after all weact's store
https://weactstudio.aliexpress.com/stor ... match_sort
but they seemed to have some new stm32 offerings
that stm32f103 'blue pill comes back'
stm32f411 'black pill' is there
stm32g431 the 'dac/opamp/comparator' board is there
high end boards stm32h723/h750 is there
stm32f405 is there
I particularly liked that stm32wb55cgu6 board, bluetooth LE yeah :P
etc
User avatar
Bakisha
Posts: 150
Joined: Fri Dec 20, 2019 6:50 pm
Answers: 6
Contact:

Re: weact

Post by Bakisha »

I think it is their official store. I've been ordering from them for the last few years.
Their bluepill is genuine STM32F103CB, the only difference is the LED on PB2 (and with USB-C)
STM32G431CB has DAC, with OK results.
I didn't have time to test STM32F411CE (25Mhz) with 8MB flash.
I did test STM32F412RET6 and SdCard with STM32SD. I couldn't get to work 100% (it works, but it can't read all subfolders).
Only bad thing that i can say about those boards is user button being on Vcc side with very small resistor, and buttons are very stiff and small, hard to press it...
ag123
Posts: 1785
Joined: Thu Dec 19, 2019 5:30 am
Answers: 28

Re: weact

Post by ag123 »

the stm32 offerings are quite interesting
the new boards I think are like the
stm32f405 and stm32wb55cgu6

wb55cgu6 is a good one as I've yet to see any other vendors offering that for now
the git repository is apparently here
https://github.com/WeActStudio/WeActStu ... 5CoreBoard

and stm32f405 here
https://github.com/WeActStudio/WeActStu ... _CoreBoard
GonzoG
Posts: 454
Joined: Wed Jan 15, 2020 11:30 am
Answers: 32
Location: Prudnik, Poland

Re: weact

Post by GonzoG »

What's most important, is that they put all necessary data on github.
Exact dimensions, diagrams, 3d model, example source code, even chip pdfs so you don't need to look anywhere else.

One thing I would like to see is a series of mini boards like seeed studio xiao but with STM's ULP MCUs.
STM32ardui
Posts: 131
Joined: Mon May 06, 2024 1:46 pm
Answers: 1
Location: Germany

Re: weact

Post by STM32ardui »

WeActStudio now have the STM32G474 also as "long board" .
And a new STM32H503CBU6: 128 kB Flash and 32 kB RAM, with 250 MHz may be a little bit faster than STM32G431.
Cortex-M33 ...
ag123
Posts: 1785
Joined: Thu Dec 19, 2019 5:30 am
Answers: 28

Re: weact

Post by ag123 »

for the 'blue pill' stm32f103c8/b fans
https://www.aliexpress.com/item/1005004918334754.html
weact seemed to stock their version of 'black/blue' pills :)
the improvements(differences) apparently are:
- the LED is apparently at PB2 active high
- buttons for boot0 (instead of jumpers) and a user button
Image
- sop8 pads for spi flash below
Image

they have a nice github repository for it
https://github.com/WeActStudio/BluePill-Plus
User avatar
Bakisha
Posts: 150
Joined: Fri Dec 20, 2019 6:50 pm
Answers: 6
Contact:

Re: weact

Post by Bakisha »

ag123 wrote: Wed Aug 28, 2024 3:18 pm
the improvements(differences) apparently are:
- the LED is apparently at PB2 active high
- buttons for boot0 (instead of jumpers) and a user button
Also, there is different layout of pins for power near USB jack. On classic bluepill, it's GND-GND-3V, while on a weact's boards it's 5V-GND-3.3V
GonzoG
Posts: 454
Joined: Wed Jan 15, 2020 11:30 am
Answers: 32
Location: Prudnik, Poland

Re: weact

Post by GonzoG »

There's new Black Pill with STM32H503CBU, 250 MHz, 128kB ROM, 32kB ROM. With 1 opamp and 1 comparator.
Pinout almost the same as F401/F411 black pills. Just one pin different, there's GND where PB9 was.

Image
https://github.com/WeActStudio/WeActStu ... 3CoreBoard
https://www.aliexpress.com/item/1005007452157414.html
ag123
Posts: 1785
Joined: Thu Dec 19, 2019 5:30 am
Answers: 28

Re: weact

Post by ag123 »

stm32h5 seemed to be a fairly new product line too
https://www.st.com/resource/en/datashee ... h503eb.pdf
cortex M33
quite similar to this I'd guess
viewtopic.php?t=2477

it is more like the F4 / G4 series with ART accelerator and all
I've got a H7 board
https://github.com/WeActStudio/WeActStu ... iSTM32H723
H7xx is apparently quite a different thing
https://www.st.com/resource/en/datashee ... h723zg.pdf
Cortex M7 with its own caches lines etc, the clock setup is also quite different from e.g. the F4xx, Cortex M4 series
H7 has system clocks that can literally be 'untangled' from the peripherals clocks
I'd guess the purpose is so that the cpu clocks can be changed on the fly without affecting IO which may require a fixed and/or lower clock speeds.
but that that setup is more 'complicated' vs F4 etc

for Cortex M33 accordingly it has an FPU
but that apparently it is same as F4
https://community.arm.com/support-forum ... -cortex-m4
while the simplified view of the FPU registers is like such
https://developer.arm.com/documentation ... -registers
Image
what is often less mentioned are that the floating point registers are also mapped as vector registors
https://developer.arm.com/documentation/ddi0553/latest/
^ it is 'briefly' covered in chapter B5 of the Arm-v8M manual

those probably account for the Vxxx instructions covered in pm0264-stm32-cortexm33-mcus-programming-manual
3.10 Floating-point instructions
https://www.st.com/resource/en/programm ... ronics.pdf

that probably accounts for the "surprisingly fast" whetstone benchmarks seen in the F4 family
viewtopic.php?t=27
that makes it likely that those 'Mflops' are likely not bluff
viewtopic.php?p=939#p939
those could very well be deemed the 'DSP' instructions
those numbers rival 'old Intel Pentium 4' like 200 MFlops
http://www.roylongbottom.org.uk/whetstone%20results.htm
just that on Intel P4 that is double precision, so that is still a lot of difference
in hardware at least
but the single precision FPU speeds makes it possible to do various DSP tasks in real time that would likely otherwise be not feasible.
Post Reply

Return to “Off topic”