dddns,

Thank you for the reply.
Maybe I missed the change? Due to the sheer number of HDMI options, I've lost track.
Unfortunately, I don't have a complete board with PSRAM and HDMI yet. I might add a separate HDMI port to my Olimex-XXL later.
I'll add a note to my help.txt.

Regards,

Jan.

In MMCC, using AUTOSAVE N on a 950 line file reduced the upload time from 23 seconds to 5 seconds. That includes time to check if AUTOSAVE N is a valid option and the final file saving time.

A huge improvement!

I need some more testing before updating the MMEdit download.

Jim

  dddns said  

  JanVolk said  

What am I doing wrong here, or is this a mistake?

It has changed in b11. You need PSRAM to use files command in Resolution 800

Maybe Peter should prevent resolution 800x600 when no PSRAM. You end up with a crippled system without PSRAM, and now dddns remembered why. But 3months from now?

Honestly, I love the better pictures and higher resolutions, but when you cannot do anything with the pico in that mode… you cannot even make a Photo carrousel since that needs the sed card directory..

Volhout

Maybe an extra warning message after the option command is sufficient.

Gerald

  matherp said  

Sounds like something going on in H/W relating to GP27 and GP28 both being used. In both your examples which error that looks like the common factor. There is nothing I can think of in the firmware that could create that as a problem.
I haven't won't have access to any H/W that can mimic your configuration for another 9 days

After a long search I found a workaround for both of my problems.

Device(s) connected below gp30 have influence to device(s) connected above gp29
I found VGA pins, SDcard and Audio
On 2350USB, a combination that gives me read errors on SDcard can also scramble the LCD

But I also found that on 2350USB: SDcard gp36-gp39 and Audio gp30,gp31 fails, changing Audio to gp26,gp27 works. So maybe not only pins above gp29

Edit:
Another device(bus) causing problems in combination is system I2C
Edited 2025-10-25 17:14 by dddns

  Quote  

Maybe Peter should prevent resolution 800x600 when no PSRAM. You end up with a crippled system without PSRAM, and now dddns remembered why. But 3months from now?

I'll fix the code so the maximum number of files scales to the memory available

  Quote  

Device(s) connected below gp30 have influence to device(s) connected above gp29
I found VGA pins, SDcard and Audio
On 2350USB, a combination that gives me read errors on SDcard can also scramble the LCD

Thanks for all your work on this. All GPIO access in the firmware should be using the 48-pin sdk variants. Assuming I can replicate your findings there are three possibilities.
1: I've missed one or more conversions from the original sdk calls to the new ones when I added support for the RP2350B
2: There is a bug somewhere in the sdk in the handling of 48-pins
3: There is some sort of H/W issue. This is why I asked specifically about your H/W platform. For example the Pico Plus 2 from Pimoroni ties GP40-GP43 to GP26-GP29 to allow ADC access on the standard 40-pin header format (yuk!!!)

Many thanks for investigating!
I use the PGA2350 because it's all pinout and there is nothing in between.
I'll double check my carrier boards too.

Hi Peter,

I found another bug, I think:

Play mp3 doen't work anymore on the following (tested!) versions:

- PicoMiteHDMIUSBV6.01.00RC7.uf2
- PicoMiteHDMIV6.01.00RC7.uf2

- PicoMiteHDMIUSBV6.01.00RC5.uf2
- PicoMiteHDMIV6.01.00RC5.uf2

so I tested also the beta version 22 doesn't work either. Please note, I only tested these versions, maybe others have problems, too.

Symptoms: total crash and sometimes even a full reset...


The last version play mp3 works is:

https://www.thebackshed.com/forum/uploads/matherp/2025-10-03_211515_PicoMiteV6.01.00b18.zip

My minimal system config is:

Option Audio Gp6,Gp7
Option SDCARD Gp5,Gp2,Gp3,Gp4

everything else is default!


Greetings
Daniel
Edited 2025-10-25 18:53 by Amnesie

Please try to move audio to gp10,gp11
If it fails to gp18,gp19

  dddns said  

Please try to move audio to gp10,gp11
If it fails to gp18,gp19

This isn't possible, I can't re-design all my PCBs:




..and in beta 18 everything works, so there must be some change in later versions.


Greetings
Daniel

I thought so..
But for a test you could just reconfigure audio and see if still crashes

edit:just saw the outside GPs. Can you use these?
Edited 2025-10-25 20:12 by dddns

  dddns said  

I thought so..
But for a test you could just reconfigure audio and see if still crashes

edit:just saw the outside GPs. Can you use these?

Good Idea, I just tried this with:

Option Audio Gp20,Gp21

as well as:

Option Audio Gp0,Gp1

Same crash, if I do this two or three times the pico does a hard reset and is locked...

As said: Play Mp3 works well until version b18 - I'm just using this right now...

Greetings
Edited 2025-10-25 21:18 by Amnesie

Last thought of mine is to try playing mp3 from A: drive as test.
I can't get Audio and SD run together so I'll leave out audio for the moment.
Play tone works, reading mp3 file causes mp3 init error, guess because disk read errors. If I leave audio output connected and Option audio unconfigured it works, as soon as I enable Option audio it will fail.

I have isolated wires for SDcard and audio, no ribbon where wires are side be side
It fails while neither Tone nor mp3 is playing and there is no audio output
Edited 2025-10-25 23:05 by dddns

Don't waste time it is a simple memory issue that I will fix in the next rc

Thanks :)

  matherp said  

Don't waste time it is a simple memory issue that I will fix in the next rc

   

How about this? I think I've finally cracked it

https://youtube.com/shorts/ErnYxtuYhac
Edited 2025-10-26 00:39 by matherp

  matherp said  

V6.01.00RC7

Changes USB CDC polling interval from 1000uS to 400uS. This definitely improves performance of the RP2040 on EDNEDN's test program although this will be at the expense of a slight performance hit. I could go lower but it ran the test for me until I got bored (I get bored easily)

I don't see any improvement.   The test program still locks up quickly (in under a minute).    And while the test program is set up to let you move the token back and forth it isn't necessary to do that.   You can just press and hold any key on the keyboard that auto-repeats.   You will know from the line counter at the start of the line that it has frozen.

I verified that the problem still exists on both RP2040 and RP2350 PicoMite's and WebMites.  

So, it may make sense to put the USB CDC polling interval back to 1000us.   If it helps, it isn't helping very much.


.
Edited 2025-10-26 04:51 by EDNEDN

V6.01.00RC8

PicoMiteV6.01.00RC8.zip


Yet more tweaking of the USB CDC environment. This is now about as far as I can take it to try and improve performance and reliability so if there are still issues you will need to code round them with pauses etc.

Hi Peter,

I assume there is no way you can implement dithering for jpg in the 2040 VGA code ?
I you cannot find a "streaming" library that does it, you would probably not been able to find the RAM to do it in 2 known steps, and flash will also not be available to hold the 24bit RGB image needed between jpg decode and BMP dithering. It would seem the MOD buffer can be used, but that means that you cannot use the feature while playing music, and flash slots are to small.

Volhout