Matrix user forums

Thank´s Leigh. I perform a test later this afternoon. First I save the old one.

You hear from me

Jorgen.

Hi Leigh!

I follow your link, but still no luck, same error.

Br Jorgen

Very odd.
I will add some diagnostics into the file on Monday.
If you can then retry
Thanks
Leigh

Thank you. Yes I will do all test. It would be nice to get it work Have a nice weekend...………

Jorgen

Hi Jorgen,

Sorry, looks like something has slipped through the net here, the path in our batch file does match the installation location for the ST-LINK programmer.
Could you please try the attached updated batch file?
This is to go into …
C:\Program Files (x86)\Flowcode\Common\Compilers\starm\batch

This new batch file assumes that the ST-LINK_CLI.exe is located in this directory …
C:\Program Files (x86)\Flowcode v8\tools\ST-LINK Utility
If you could just check that is the case for your installation.

Thanks!
Leigh

Hi Leigh!

Doesn´t work.

But I found the file ST-LINK_CLI.exe in: C:\Program Files\Flowcode v8\tools\ST-LINK Utility

Not in: (x86)\Flowcode v8\tools\ST-LINK Utility

This here is on an "little" older pc, but still Win10. Is it maybe possible to without danger to copy the diretory TOOL in the (x86) directory?

I will test it on my newer PC later this evening.

Later :

Yes Leigh, finally I am able to program the STM32F0308 Discoveery board on my other PC On this pc the ST-LINK_CLI.exe is located ac you mentioned.

So, please what do I do in the other PC, is it ok to copy the TOOL lib. into the (x86) lib without destroy anything for other device???

Later again: Unfortunately I am not able to program the STM32F410 Nucleo-64

I am grateful for your help.

Br Jorgen.

Hi Jorgen,
Yes, you can copy the tools directory.
Or edit the path in stm32prog.bat

The Nucleo boards are programmed differently, they do integrate the ST-LINK programmer,
but the default setting is to appear as an external drive when plugged into the PC USB port.
And that is used by Flowcode in the case of a Nucleo board selected as the target.
Leigh

Hi Jorgen,
Well this is curious, it turns out that the issues regarding ST-LINK and stm32prog.bat are due to the website STARM Toolchain download file not being an up to date version.
We are getting this fixed asap
Leigh

Hi Leigh!

Does it mean that there also is a chance to be able to program the Nucleo also? And these correction will they also be pushed to update via help?

Thank´s in advance.

Br Jorgen

Hi Jorgen,

For the Discovery board (that uses ST-Link) I would suggest that you copy the attached bat file into the batch directory normally at:
C:\Program Files (x86)\Flowcode\Common\Compilers\starm\batch
(If the batch directory is in a different place on your other PC then that is not a problem)

Then copy the whole of the “ST-LINK Utility” subdirectory (from tools) into the starm directory:
C:\Program Files (x86)\Flowcode\Common\Compilers\starm
Which should then look like this ...
The Discovery board programming should now work well on both PCs

When I have confirmation that the correct STARM toolchain pack is on the website, downloading and installing this will achieve the same as the above changes.


Next ...
The Nucleo boards (with default settings) work differently, what happens when you plug it into the USB?

Thanks
Leigh

Hi Leigh!

You rote:
"When I have confirmation that the correct STARM toolchain pack is on the website, downloading and installing this will achieve the same as the above changes."
So this mean that a new toolchain for STARM will soon come and this will make the correction as you mentioned automatic? In that case I wait until new update.Please let us know when this will be on. At least now I can program the Discoboard from one PC

Nucleo:

When I attach the USB I says: NODE410 (E:) decide what to do with this drive. And I have att. my testprogram for the Nucleo.
hope i'm not in too much trouble

Br jorgen.

Hi Jorgen,
The updated STARM Toolchain is now on the website.
Looks like there is a problem with the Nucleo 410, I'm currently looking into it
Leigh

Hi Jorgen,
Please can you update to the attached FCD file for the F410 Nucleo64, to go into:
C:\ProgramData\MatrixTSL\FlowcodeV8\FCD\ARM
I will get this into the update system tomorrow.
Thanks
Leigh

Hi again Leigh!

I can´t find th link: C:\ProgramData\MatrixTSL\FlowcodeV8\FCD\ARM. So not able to test today, I wait until tomorrow when I get it via update.
Thank´s in advance.

Br Jorgen

C:\ProgramData\ is usually a hidden directory,
you might need to change folder options

or wait until tomorrow

Hi Leigh!

I have just installed the new toolchain on two PC. The Discovery board is still progammable, perect

Look forward to test the Nuclea tomorrow after update.

When it work, I have some question regarding the config setup, hope it´s ok

Br Jorgen

Hi Leigh!

Yes, after update, testing the Nucleo also, programming perfect, working. You MM guy´s and Valued Contributor are amazing…...

Now I have to figure out to set the settings in config, how I use the PLL ect. I want it to be fast as possible. I´ll see if I can speed the Ili9341 up a bit.

Again thank´s and have a nice weekend.

Br Jorgen

Hi Leigh.

I made some test. I have test the Nucleo with an ILI9341. And compared with different Arduino Devices, I think the Ili9341 updating with Nucleo is twice as fast than the other devices waov And output is 3,3V so it is easy to connect. So, really powerfull.

I made also a test with Arduino Mega Eico by turning an output on and off ith a delay of 1 MicroSec. see att. I measure with my scope:

Nucleo : Exactly 1 uSec.

Arduino Mega : approx. 2,2 uSec.

Eico: approx. 3 uSec.

I am not sure that it is the rigth way to test the speed versus correct timing. But look like

I see in the config there are lot of setting, with PLL and ect. I have no experience with these. Could you please have a look at the att. program, maybe it is possible to speed it further up

Thank´s in advance

Jorgen

Hi Jorgen,
Yes, that looks fine, and you are running at max
ST have a good tool that helps with clock configuration....
https://www.st.com/en/development-tools ... ubemx.html

Hi Leigh!

Thank you.

Br Jorgen

Hi to All

Following the instructions I managed to correctly program the stm32f411e discovery

I tried playing a little with the clocks and setting it to maximum (100MHz) I managed to get 8MHz of toogle on the pins.

I am comparing this new mcu with the current pic32mx470 we are using for our projects.

From the first tests I noticed this:
although the speed of toggle pin is equal to twice that obtained with pics, the writing speed of the display is slightly lower. (Ili9341)
Especially the clear that the pic executes it instantaneously, on the stm32 it takes us almost 0.5s.

The display is connected in parallel mode 8 bits in both cases.

This is probably due to our use of the ST HAL for I/O

LeighM wrote:This is probably due to our use of the ST HAL for I/O

Do you have suggestions to improve the speed with the display?

Can I try to do something ??

with pic32mx I had to insert some c code to set the clocks etc correctly in order to get the required performance

Hi
just as a note your test with the Pin toggel on the PIC32 is horribly wrong if you search on the Microchip web site/Forum you will find some C code which they ran and people use to test the clock speeds this code yields 70 Mhz on a pin toggle so don't be so happy that the STM is showing faster speeds(joke)the PIC eat a STM for breakfast if used as intended
what peripheral clock speed setting you use on the PIC32? that matter a big deal!

QMESAR wrote: what peripheral clock speed setting you use on the PIC32? that matter a big deal!

The pbclk is set to 20 MHz while the SYSlClock is set to 40 MHz

In my application I tried to set the clock to maximum but I didn't get a improvement.