TheBackShed.com

Forum Index : Microcontroller and PC projects : PicoMite V6.01.00 release candidates

Page 10 of 18

Author

Message

dddns
Guru

Joined: 20/09/2024
Location: Germany
Posts: 703

Posted: 06:04pm 27 Oct 2025

Copy link to clipboard

Print this post

I want to use the new buffered driver for ILI9341.
I tried it before and still I get an error with Option lcd pins.
Do I overlook something?

PicoMite MMBasic USB RP2350B Edition V6.01.00RC9
OPTION SERIAL CONSOLE COM2,GP8,GP9
OPTION FLASH SIZE 16777216
OPTION COLOURCODE ON
OPTION KEYBOARD US
OPTION PICO OFF
OPTION CPUSPEED (KHz) 200000

> option lcd pins gp2,gp3,gp0
Error : Invalid Option
> option lcd pins gp14,gp15,gp12
Error : Invalid Option

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10653

Posted: 06:18pm 27 Oct 2025

Copy link to clipboard

Print this post

option lcd spi

dddns
Guru

Joined: 20/09/2024
Location: Germany
Posts: 703

Posted: 06:40pm 27 Oct 2025

Copy link to clipboard

Print this post

Thank you. I took the information from page 3..

Volhout
Guru

Joined: 05/03/2018
Location: Netherlands
Posts: 5485

Posted: 07:17pm 27 Oct 2025

Copy link to clipboard

Print this post

@Peter,

Closing MMedit makes no difference. Tried Minicom, Picocom, Putty. Only MMCC has this behaviour. This happens when you close MMCC, and also when you leave MMCC open but DISCONNECT. It has no relation to uploading a program via MMedit (what I thought originally).

Using MMCC as a terminal shows the behaviour, as does GFXterm64 (also both with closing, and using DISCONNECT).

It is something both have in common.

But it results in a change in the pico !! Becuase once MMCC or GFXterm64 cause this problem, re-opening the port in Putty/Picocom , and later closing it, does not fix it. It only gets repaired when the pico is CPU RESTARTed.

So it is caused by something MMCC / GFXterm64 configure into the pico that is restored with a CPU RESTART (and power cycle).

Volhout

EDIT: V60100RC7 also has this behaviour, but V60100b20 does not show this behaviour. I do not have in between versions anymore to pinpoint further in detail where the change occurred.
Edited 2025-10-28 05:32 by Volhout

PicomiteVGA PETSCII ROBOTS

Volhout
Guru

Joined: 05/03/2018
Location: Netherlands
Posts: 5485

Posted: 07:46pm 27 Oct 2025

Copy link to clipboard

Print this post

Peter, could it have Some relation with autosave N?
Where dat is zend to usb first, and only to screen when no N?
Where previous it was sent to screen first, and then to usb.

Blocking usb could have this effect when the order is swapped.

Volhout

PicomiteVGA PETSCII ROBOTS

Volhout
Guru

Joined: 05/03/2018
Location: Netherlands
Posts: 5485

Posted: 07:46pm 27 Oct 2025

Copy link to clipboard

Print this post

PicomiteVGA PETSCII ROBOTS

EDNEDN
Senior Member

Joined: 18/02/2023
Location: United States
Posts: 271

Posted: 07:51pm 27 Oct 2025

Copy link to clipboard

Print this post

Volhout said
EDIT: V60100RC7 also has this behaviour, but V60100b20 does not show this behaviour. I do not have in between versions anymore to pinpoint further in detail where the change occurred.

I saw some strange behavior yesterday. I was connecting and disconnecting MMCC and Tera Term back and forth. And all of a sudden I started seeing 'Stack Overflow' messages from MMBasic.

Power cycling the Pico didn't help. Doing a 'New' and Reloading the BASIC program did not help. Reflashing the firmware and doing other things would not get rid of the problem. I even FlashNuked it several times and that did not help.

What finally broke the cycle was loading B18 onto it. I only have high confidence in what I just stated. I do not have any confidence that I know what I did wrong.
Edited 2025-10-28 06:04 by EDNEDN

TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 6364

Posted: 08:36pm 27 Oct 2025

Copy link to clipboard

Print this post

@Volhout, Peter

MMCC does assert DTR. That was needed for any comms with the picomites.
I could try dropping DTR and wait 100mS when closing the port.
That should definitely stop the data flow.

In Linux I had to do a few register tweaks for baud rates over 115200. I know the baud rate does nothing with the picomite USB but If you have it set high, perhaps a lower setting is worth a try.

I don't have a VGA pico to test with and my Linux is in a VM so not a good test for USB/serial.

Jim

VK7JH
MMedit

Volhout
Guru

Joined: 05/03/2018
Location: Netherlands
Posts: 5485

Posted: 09:05pm 27 Oct 2025

Copy link to clipboard

Print this post

Jim,

I use 115200 baud.

Volhout

PicomiteVGA PETSCII ROBOTS

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10653

Posted: 10:14pm 27 Oct 2025

Copy link to clipboard

Print this post

I have massively improved USB CDC performance since b18 and don't intend to change back any further than the Linux specific fix incorporated yesterday.

Volhout
Guru

Joined: 05/03/2018
Location: Netherlands
Posts: 5485

Posted: 07:00am 28 Oct 2025

Copy link to clipboard

Print this post

Peter,

Jim fixed it in MMCC, pulling DTR low 100ms before closing the CDC channel.
A similar fix may be needed in GFXterm64.

Volhout

PicomiteVGA PETSCII ROBOTS

Bleep
Guru

Joined: 09/01/2022
Location: United Kingdom
Posts: 694

Posted: 05:00pm 28 Oct 2025

Copy link to clipboard

Print this post

Hi Peter,
Not sure what you have managed to do when you say "some performance improvement for all version" but some of my screen intensive programs, on both a HDMI and a ST7796 (not Buffered) have hugely sped up, from 6.00.02 to 6.01.00RC9 either processor, though it looks like the 2350 is by far the biggest improvment, these are all USB if it makes a difference.

HDMI, 2350, old firmware 329mS per screen update, now 285mS !!!
ST7796, 2350, old firmware 320mS per screen update, now 275mS !!!
VGA, 2040, old firmware 616mS per screen update, now 599mS

Thanks for the latest, all round, improvements. :-)
Regards, Kevin.

Footnote added 2025-10-31 02:34 by Bleep
VGA, 2350, old firmware 328mS per screen update, now 285mS, not a USB version.

EDNEDN
Senior Member

Joined: 18/02/2023
Location: United States
Posts: 271

Posted: 08:08pm 28 Oct 2025

Copy link to clipboard

Print this post

matherp said I have massively improved USB CDC performance since b18 and don't intend to change back any further than the Linux specific fix incorporated yesterday.

Is the source for RC9 on GitHub now? There are a lot of files that have been recently updated. But the README.md and Version.h both claim to be RC8.

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10653

Posted: 10:02pm 28 Oct 2025

Copy link to clipboard

Print this post

Quote Is the source for RC9 on GitHub now?

No: will be a day or three

disco4now

Guru

Joined: 18/12/2014
Location: Australia
Posts: 1060

Posted: 11:41am 30 Oct 2025

Copy link to clipboard

Print this post

Peter,
Since the new parameter type // has been added to the error() message format the errors for trying to use //000 and /&00 to represent a Null character are being formatted as integers.
e.g.

"Null character 0&00 in escape sequence - use CHR"
"Null character 0000 in escape sequence - use CHR"
"Null character 2007AEA7&00 in escape sequence - use CHR(0)"
"Null character 80000000000 in escape sequence - use CHR(0)"

Suggested this new error message is used in place of the current error message at the locations below. i.e. don't included the //000 and /&00 which are being formatted.

if(i==0)error("Illegal escape sequence, use CHR$(0) for the Null character","$");

F4 H7 FotS F4xGT

Bleep
Guru

Joined: 09/01/2022
Location: United Kingdom
Posts: 694

Posted: 09:22pm 30 Oct 2025

Copy link to clipboard

Print this post

Hi Peter,
I was playing with the new jpg load command on a 480x320 lcd 2350 rc9
LOAD JPG file$ [, x] [, y] [,mode] [,ximage] [,yimage]
I don't think the yimage is working at all, ximage is fine it shifts the image left, but I was expecting yimage to shift up, but as I said it does nothing.
Regards Kevin.

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10653

Posted: 08:57am 31 Oct 2025

Copy link to clipboard

Print this post

Kevin
Thanks for the two reports - mwill fix

Pluto
Guru

Joined: 09/06/2017
Location: Finland
Posts: 398

Posted: 03:11pm 02 Nov 2025

Copy link to clipboard

Print this post

Peter,
I just noticed that the transparent option for the TEXT command is not anymore OK. Instead of transparent the background is white.
This is WebMite MMBasic RP2040 Edition V6.01.00RC9 (as far as I know the latest edition).

> option list
WebMite MMBasic RP2040 Edition V6.01.00RC9
OPTION SYSTEM SPI GP18,GP19,GP16
OPTION AUTORUN ON
OPTION COLOURCODE ON
OPTION CPUSPEED (KHz) 200000
OPTION LCDPANEL ILI9488, LANDSCAPE,GP17,GP14,GP13
OPTION WIFI HuaHuaNisse, ***********, ElprisWebMite
OPTION TCP SERVER PORT 80
OPTION TOUCH GP12,GP11
GUI CALIBRATE 0, 3946, 3793, -1271, -894
> cls rgb(blue)
> text 10,300,"WebMite",LTU,5,1,RGB(BLACK),rgb(yellow)
> text 60,300,"V6.01.00RC9",LTU,5,1,RGB(BLACK),-1
>

The last line used to make a transparent background, but not anymore.

matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 10653

Posted: 03:20pm 02 Nov 2025

Copy link to clipboard

Print this post

Depends on the display. If reading isn't enabled (ILI9488 modified) then it can't work

Pluto
Guru

Joined: 09/06/2017
Location: Finland
Posts: 398

Posted: 07:40pm 02 Nov 2025

Copy link to clipboard

Print this post

Thanks Peter,
I had made the ILI9488 modification (diode-resistor connection) wrong on my second unit.
Now everything is OK with WebMite MMBasic RP2040 Edition V6.01.00RC9.
Funny how difficult it can be to critically inspect own connections.

Pluto

Page 10 of 18

Print this page