|
Forum Index : Microcontroller and PC projects : MMBasic V6.03.00 release candidates
| Author | Message | ||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 11352 |
V6.03.00RC7 PicoMiteRP2040V6.03.00RC7.zip PicoMiteRP2350V6.03.00RC7.zip Fixes bug introduced in rnd() Fixes bug introduced in POS/pos used in strings Improved error detection in WEB versions Additional parameter in OPTION WIFI If set correctly this will overcome issues with things like channel 12,13 usage by a router in countires where that is legal Edited 2026-05-12 19:22 by matherp |
||||
| BarryH Regular Member Joined: 05/01/2025 Location: AustraliaPosts: 44 |
Hi Peter, LIST PINS is still faulty on 2350B DIL (unless I'm using the wrong firmware) PicoMite MMBasic RP2350B V6.03.00RC7 OPTION FLASH SIZE 16777216 OPTION COLOURCODE ON OPTION PICO OFF OPTION CPUSPEED (KHz) 200000 GP0 1 GP1 2 GP19 25 GP20 26 GP21 27 GP22 29 GP23 41 GP24 42 GP25 43 GP26 31 GP27 32 GP28 34 GP29 44 GP30 45 I've done a bit of research, gone back through several previous versions, and found that even before LIST PINS was included in the V6.00.002 firmware, the use of the "FreePins" code snippet in V6.00.001 gives incorrect Pin / GPx associations. Each time I have been using Clear_Flash_RP2350.uf2 and FLASH ERASE ALL I could have sworn I've used the command successfully, previously. Edited 2026-05-12 22:13 by BarryH BarryH |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 11352 |
The firmware is correct. The pin mapping is for the Pico2 only. Pins above gp28 have "virtual" pin numbers that don't reflect any specific layout. The firmware can't know what variant of PCB a RP2350B is on and there isn't a Raspberry Pi standard for the RP2350B. Use GP designation and you can't go wrong and the code is transportable between different PCBs Edited 2026-05-12 23:09 by matherp |
||||
| BarryH Regular Member Joined: 05/01/2025 Location: AustraliaPosts: 44 |
thanks Peter - I fully understand the principal now. Edited 2026-05-12 23:38 by BarryH BarryH |
||||
| dddns Guru Joined: 20/09/2024 Location: GermanyPosts: 834 |
Sorry, my webmite still does not reconnect after CPU Restart connected to a Fritzbox 7520 via DHCP: > cpu restart /usr/bin/cu: Got hangup signal Disconnected. dietmar@Intel-Desktop:~/Projekte/VT-100$ /usr/bin/cu -l /dev/ttyACM0 -s 115200 Connected. de connecting to WiFi... failed to connect. > option list WebMite MMBasic RP2040 Edition V6.03.00RC7 OPTION COLOURCODE ON OPTION CPUSPEED (KHz) 252000 OPTION WIFI PicoLan, *********, de OPTION TELNET CONSOLE ON Not for me, but if length of the list exceeds local console/screen terminal rows then it won't wait for a key press. On local screen lines are gone.. Edited 2026-05-13 00:30 by dddns |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 11352 |
Nothing more I can do about the Fritzbox. I've done everything that the various AIs suggest. Re-purpose it as a boat anchor? Question: Does the Fritz!Box UI still show the device as “connected” for several minutes after the RP2350 has rebooted? Also, try this: Disable “PMF required” (802.11w) Set it to Optional, not “Required”. On Fritz!Box: Wi‑Fi → Security → “Protected logins” → set to Optional Edited 2026-05-13 00:58 by matherp |
||||
| dddns Guru Joined: 20/09/2024 Location: GermanyPosts: 834 |
Ok. My current setup is DHCP and FB assigns hostname and IP address based on MAC and support for legacy devices is enabled. I will look at all possible setups and report asap. |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 11352 |
The possible critical one is: Wi‑Fi → Security → “Protected logins” ✔️ The real fix that works with Fritz!Box and keeps a stable MAC There is exactly one robust solution that does not rely on deauth and does not require MAC changes: ➜ Force Fritz!Box to drop stale sessions immediately This is done by disabling or relaxing one Fritz!Box feature: Disable “PMF required” (802.11w) Set it to Optional, not “Required”. On Fritz!Box: Wi‑Fi → Security → “Protected logins” → set to Optional Why this works: When PMF is required, Fritz!Box must treat missing deauth frames as suspicious. When PMF is optional, Fritz!Box is allowed to clear the session when the client disappears. This single setting fixes reconnect issues for: ESP32 ESP8266 CYW43 (RP2040/RP2350) many IoT devices older Android phones some laptops with buggy drivers It’s the closest thing to a “universal fix”. ✔️ The second fix (if PMF is already optional) Disable “WPA3 transition mode” Fritz!Box sometimes mishandles WPA2/WPA3 mixed‑mode replay counters after a client reboot. Set Wi‑Fi security to: WPA2 only, or WPA3 only, but not mixed mode. ✔️ The third fix (if you want to keep PMF and WPA3) Shorten the Fritz!Box association timeout Hidden setting: Home Network → Network → Network Settings → Additional Settings → “Idle timeout for inactive devices” Set it to 1 minute. This forces Fritz!Box to drop stale sessions quickly. 🎯 Summary: Why Fritz!Box is “contrary” Because it: strictly enforces WPA2 replay counters strictly enforces PMF strictly enforces association state does not clear stale sessions quickly assumes reconnecting without deauth is suspicious treats embedded clients as potential attackers This is good security, but bad interoperability. |
||||
| ville56 Guru Joined: 08/06/2022 Location: AustriaPosts: 490 |
@dddns: are you on the latest firmware level? I'm running 8.25 on a FB 7590 and fighting with similar probs. Tried to use Peters latest mod and supplied the country code, also had channel 12/13 enabled in the FB (disabled now). Before had the problem that the Pico lost connection sometimes (every 2 to 3 days), now I have the situation that the Pico does not always connect on CPU restart, mostly each second try works. It is too early to say if the connection is now stable, but if not I will try using WPA2 only (current setting WPA2 + WPA3) with the setting "Unterstützung für geschützte Anmeldungen von WLAN-Geräten (PMF) aktivieren" deactivated which is currently ticked and greyed out because of WPA3 enabled. 73 de OE1HGA, Gerald |
||||
| JanVolk Guru Joined: 28/01/2023 Location: NetherlandsPosts: 352 |
toml_12953 and ville56, Thanks for the reply and the translation. The help.txt in English has been updated to V6.03.00 Copy it to MMEdit and edit it, then copy it to the PicoMite board to flash drive with the filename help.txt help.txt-V6.03.00-eng.txt.zip Jan. |
||||
| toml_12953 Guru Joined: 13/02/2015 Location: United StatesPosts: 637 |
Thanks! |
||||
| toml_12953 Guru Joined: 13/02/2015 Location: United StatesPosts: 637 |
Is there any way to change the width of listings on a PicoCalc (40-column LCD display)? I have an 80-column serial printer and it's a waste to have 40-column listings come out. I'd like to leave PRINT as is but be able to change the output of the LIST command only. Maybe even have LLIST and LPRINT commands that print to a serial port rather than the console. Edited 2026-05-13 13:51 by toml_12953 |
||||
| ville56 Guru Joined: 08/06/2022 Location: AustriaPosts: 490 |
Jan, many thanks for the new help file. 73 de OE1HGA, Gerald |
||||
| ville56 Guru Joined: 08/06/2022 Location: AustriaPosts: 490 |
Peter, the report of the line causing an error seems to be 1 line off. running Do mx=MOUSE(x,2) my=MOUSE(y,2) If mx <> tx Or my <> ty Then Print mx, my tx = mx : ty = my Loop gives RUN [1] mx=MOUSE(x,2) Error : MOUSE is not declared > the correct error line should be [2] Version: PicoMiteHDMI MMBasic USB RP2350B Edition V6.03.00RC7 BTW, the example is taken from the manual "Advanced Graphics Functions for the PicoMite Firmware" which IMHO should be revised. Or is the manual already invalidated? Gerald Edited 2026-05-13 16:57 by ville56 73 de OE1HGA, Gerald |
||||
| Frank N. Furter Guru Joined: 28/05/2012 Location: GermanyPosts: 1098 |
How did you connect the printer to the PicoCalc??? Frank |
||||
| toml_12953 Guru Joined: 13/02/2015 Location: United StatesPosts: 637 |
It's an Oki Microline 184 Turbo serial printer connected to serial port 1 on pins GP0 and GP1 at 19200 baud. I use a TTL to RS-232 converter (about $3.00 USD). I had an Epson receipt printer on there at 115200 baud (you should see the paper fly out of that thing!) but wanted a standard paper printer rather than thermal paper. Edited 2026-05-13 17:49 by toml_12953 |
||||
| bfwolf Senior Member Joined: 03/01/2025 Location: GermanyPosts: 246 |
If you could live with only being able to "list" programs on the printer that are already saved as files: The programs are, after all, stored in the files as "ASCII streams" (using CR as the line separator). So, write a program that reads a file line by line and then outputs it line by line via the serial interface. Check the manual under "PRINT #nbr, expression [[,; ]expression] … etc." The "LLIST" program could also be stored as a "resident" subroutine within the library. It could accept the filename as a string parameter. Regards |
||||
| dddns Guru Joined: 20/09/2024 Location: GermanyPosts: 834 |
Wi‑Fi → Security → “Protected logins” I tried all I could. With disabled PMF, static and DHCP, WPA2 and WPA2+3 without success but: Doing a CPU Restart for the first time and the connection fails, doing it a second time and it connects. I found that it connects very robust every second attempt..I tested it 10 times in a row EDIT: I tested it with Android mobile as hotspot and it shows the same. Doing CPU Restart in a row let every second try be successfully connected..it toggles Edited 2026-05-13 20:27 by dddns |
||||
| ville56 Guru Joined: 08/06/2022 Location: AustriaPosts: 490 |
@dddns, I think what you've achieved, is the best you can get with MMBasic and FB. Same here. MMBasic and Fritz!Box are no friends, even though ESP32 with Annex Basic connects without a hitch as well as other ESP8266 boards. It's probably an issue with the SDK or wherever the IP stack comes from. What I do is test at the very beginning of the program if the Pico has an IP address other than 0.0.0.0. If not--> CPU RESTART. This works for me since long time and is reliable. As long as you don't have unexpected disconnects as I do, which are not easy to discover, you should consider to live with it .... btw, from my experiance, this behaviour has already changed to normal and back in the last 3 to 4 years serveral times, I think with the updates of the FB firmware but I have no way to prove that theory. Currenty the actual RC version of MMBasic looks good as far as disconnects are concerned. First time connects are rare though. Gerald 73 de OE1HGA, Gerald |
||||
| dddns Guru Joined: 20/09/2024 Location: GermanyPosts: 834 |
Thanks. As it really toggles _every_ time between failure and connection after CPU Restart I think there is hope |
||||
| The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2026 |