Is anyone here willing to try my port of PicoMite6.00.02RC17 to run on the PicoCalc? If so, try this link. I’d appreciate your feedback! Please remember that this is alpha software, never tried on real hardware. If you’re not comfortable recovering from hard lockups or don’t know how to totally erase flash, please don’t use this! Wait for after the tests!
I forgot to say that this is for the Pico 2W. My bad!
Hi i think it would be better to post it here so we can try and give feedback.
How do I do that? I can’t figure out how to post a file here.
Just upload the files on some cloud platform like google drive megaupload etc and post a link here.
OK, I posted the file in my original message. Good luck!
Tried it and it works like a charm. I was able to telnet to the device, open a few connections, and it all seems to work fine.
TY for the first feedback! I rely on such testing because my PicoCalc hasn’t shipped yet.
It worked! I knew if I said my unit hadn’t shipped yet, it would. While I was typing my previous message, I got the email notifying me that Order # 347xx has shipped. Works every time!
Any chance you can share the code on github or somewhere? Currently there’s not really a place to get code for a 100% working WebMite build. I never did work out what was causing that keyboard disconnect in my fork. And I still need to update it to the latest version. I’m also curious what you mentioned in that other thread about limitations on overclocking, because I never saw what you were describing in the PicoMite/WebMite sources. Limits were there, but not matching up with the numbers you mentioned.
It would be even better if you made this available to @guu so it could be rolled into the official PicoCalc repo. Though it would also be nice to have a repo that is cleanly forked off the official PicoMite one and could not only diff nicely against it but also easily incorporate updates from it.
The PicoMite source for RC17 is at PicoMite-6.00.02RC17 I compared the source to
the source at your RC9 site. The changes in the two release candidates are minimal. Since your source changes fit into the new source so perfectly, it seems to work fine. Even the keyboard issue was fixed with the new source, according to a post on The Back Shed, although I haven’t compiled it for any version other than WebMite for PICO 2.
I don’t get any keyboard input if I compile the code you are pointing. Anything else needs to be defined besides :
set(COMPILE PICORP2350)
set(COMPILE VGARP2350)
set(COMPILE PICOUSBRP2350)
set(COMPILE VGAUSBRP2350)
set(COMPILE HDMI)
set(COMPILE HDMIUSB)
set(COMPILE WEBRP2350)
I started getting the I2C Keyboard not responding message after about a half hour using this uf2 file. Looking at this thread, it seems like I am the only one. I did set two options, I set CPUSPEED to 250000 and I connected to my wifi network. I also ran a couple of programs that always ran fine on guu’s 6.000.02RC7 build.
That’s the main PicoMite repo, so it doesn’t have support for any of the PicoCalc hardware. The offical PicoCalc repo is out of date and unlikely to be updated (kind of how it usually works with Clockwork software).
I’ve been updating my fork of it, and successfully compiled and tested (as much as I could with my broken screen which is getting even harder to read) the newest WebMite on my PicoCalc. The regular PicoMite build for Pico2 completed, but I haven’t tested it. The build for the Pico failed though. Not sure if it’s broken in the main repo or if it’s something only broken on the PicoCalc.
I’m going to go ahead and release this so others can test the WebMite and PicoMite builds for Pico2. It should be the same as the build Toml_12953 posted a binary of, but since they didn’t share the source, there’s no way of knowing for sure.
And as @cjstoddard indicated, it sounds like it still has the keyboard disconnect error. @Toml_12953 claimed it was somehow fixed, but since they didn’t share source, there’s nothing to compare against. And if you were using their binary and it has the disconnect error, then I guess it hasn’t been fixed. 
When I updated my repo just now, I accidentally rebased it. (I’m still getting used to using git.) So it lost the old history, but the new merge is basically aplying all the old changes to the latest RC17 official PicoMite repo. I’ll post the release binaries in a few mintes along with a link here…
Here’s a link to the latest releases for Pico2.
Just got the Pico build done. Untested, but I think it should work. Looks like work was being done on MP3 support and most of it was wrapped in #ifdefs as only supporting the Pico2. But there were a few blocks of related code that weren’t #ifdef’d out, so the Pico build failed with that unsupported code.
I’ve got the bootloader version building now, and when it’s done I’ll add them to the releases with the Pico2 stuff that’s already there.
EDIT:
Ok, all the releases are up. I only tested the WebMite for Pico2, but the others should work in theory. 
Only set ONE of those. If you set more than one at a time. who knows what may happen!
Select the one that matches your system.
The changes I used are YOUR changes so the source is available but in pieces. The RC17 unchanged source is available and your changes are available by reading your changed source.
Right. Thanks for trying the latest changes from the new RCs. Anyhow, code is up there now. (PRs are always welcome.)
Unfortunate about the keyboard thing. I suppose I could compare the older versions again to see if I missed anything from the guu sources (based on V6.00.02RC7) to the next PicoMite release(s), but I already double checked that back when it started happening. I’m not familiar enough with the platform to know where things might have gone wrong. The PicoCalc modifications are relatively isolated, but I’d guess it’s some kind of weird timing issue. Or maybe there’s some kind of conflict with other hardware options that PicoMite supports. The strange thing is that it usually works for a while, and then eventually disconnects. Sometimes it can go for quite some time, and sometimes it happens more quickly. And it doesn’t even seem to depend on using the keyboard, with it happening whether the device is idle or actively being used. It also occurs on both Pico and Pico2 platforms.
Thanks I understand better the source situation.
As for the keyboard not responding bug, could it be picomite changing the i2c configuration or something else interfering with the pins used for i2c?
FYI: RC18 of the original code is now available. Go to the Back Shed to get the link