Jump to content


  • Content Count

  • Joined

  • Last visited

Posts posted by mozzwald

  1. A new firmware update is available at https://fujinet.online/download

    FujiNet Version: 0.5.ed9d13bd
    Version Date/Time: 2021-03-15 00:55:47
    Build Date/Time: Mon Mar 15 03:32:59 UTC 2021

    Brilliant Bug Fixes:

    * Temporarily Limit bytes waiting to 512 to work around race condition of getting status during opening a file.


    While I was fixing the N: handler to properly handle CR/LF translation, I uncovered a race condition where as part of the initial interrupt after open, it was attempting to retrieve the entire file in order to do CR/LF translation if needed. Because this was happening BEFORE the SIO transaction had been acknowledged, it was causing time-outs when attempting to open a file. Until I can fully untangle it, I am limiting the total bytes waiting to 512 bytes or less, so that N: TNFS can work again for now. Sorry about any inconvenience caused.

    Please do not respond to this thread. If you have an issue, start a new post in the FujiNet subforum, open an issue on the github tracker or ask for assistance in Discord chat.

    • Like 3

  2. 33 minutes ago, x=usr(1536) said:

    Just curious if this is something that is being investigated, or if it would be worth filing a bug in github.  I'd really like to get a public-facing TNFS server up and running, but until it can be prototyped for home usage, I'm at an impasse for doing so.

    Try accessing both your TNFS servers (raspi & windows) using the python TNFS client. If it fails also, then you've got a windows problem.


  3. 1 hour ago, mytek said:

    Didn't realize there was a version with a PCB antenna only (found it: ESP32-WROVER-E). Thus far it's been working, but I guess it'd be much better signal strength with the internal antenna connected. Any pics showing the resistor mod?


    Edit: Found this reference in the datasheet that clarifies what you were suggesting (move resistor from R14 to R15). I think I'd have better luck with the solder blob method, now I just need to get out my magnifiers ;)

    If it helps, here's a picture of the resistor and which two pads to short for which antenna. It's a simple fix, but not easy for everyone ;) 


    • Like 1
    • Thanks 1

  4. On 3/7/2021 at 11:26 AM, mytek said:

    The SD problem is very baffling.

    Did you get anywhere with the SD card problem? I see you've got new boards incoming. What messages did you get about the SD card in debug output?


    I notice you are using the WROVER-IE module which is setup to use an external antenna. If you plan to use it without external antenna you should move the 0 ohm resistor next to the antenna connector which puts the wifi signal on the pcb antenna instead. Alternatively, you could remove the resistor and solder blob the correct set of pads (for those not able to handle the tiny parts).

  5. 14 minutes ago, Faicuai said:

    So how does Fujinet gets powered ensuring zero power-noise coming through SIO? 

    Schematics are on github if you'd like to see how it's powered by either SIO5V or external USB.

    14 minutes ago, Faicuai said:

    If the answer is external, then the next question is: If I run a POKE 710,2 and then 710,4, do I get perfectly clean, flat gray-fields (free of any structured-noise patterns), while Fujinet operates at full tilt?

    I just ran those while SIO and USB powered and did not notice any problems, but the esp32 is not running 'full tilt'. Need to think of a way to make the cpu stay busy to do a proper test. I also have UAV installed in this 800XL which could make a difference.

  6. A new firmware update is available at https://fujinet.online/download

    FujiNet Version: 0.5.043f69be
    Version Date/Time: 2021-03-11 02:29:28
    Build Date/Time: Thu Mar 11 02:58:53 UTC 2021

    * NEW 'mount and boot' feature: This makes FujiNet skip CONFIG and instead mounts all slots and cold starts for a faster boot up. You can press SELECT during the 'mount and boot' to re-enter CONFIG. In the web interface under BOOT SETTINGS you select/enable 'Mount Everything' to turn this on.


    Please do not respond to this thread. If you have an issue, start a new post in the FujiNet subforum, open an issue on the github tracker or ask for assistance in Discord chat.

    • Like 2

  7. 1 hour ago, _The Doctor__ said:

    Is there a reason for the long delay? Does it need time to make connections or something?

    Exactly this.


    As Thom explained, the network may not be ready so we need to wait for it if you have a TNFS disk selected in a slot. How are you supposed to boot a disk on the internet if there is no internet connection? Perhaps we can refine the process more so that if a SD disk image is chosen, then don't wait for wifi. For now this is a start in the right direction.

    • Like 3

  8. fujinet.atari8bit.net hosted by @Mr Robot appears to be back up now (https://atari8bit.net/fujinet-server/)


    2 hours ago, Gunstar said:

    there were several I deleted as nearly every time I try to use the there is an error and my Fujinet restarts

    The fujinet shouldn't 'restart' when it fails to connect to a host, after 3 failed attempts to connect (visible in the serial debug output) it should display an error on the bottom line then it verifies wifi connection before returning to the main slots screen. If something else is happening, your firmware may be out of date. (current version: https://atariage.com/forums/topic/313303-fujinet-firmware-update-information/?do=findComment&comment=4765383)


  9. 1 minute ago, mytek said:

    Yeah that's what I was seeing, but just wanted to verify. So if the 8GB board is used, what is the compromise, and does it affect what firmware gets programmed -or- the use of the FujiNet Flasher app?


    FujiNet flasher only supports the 16GB WROVER module. You can use PlatformIO/VSCode to build and flash the 8MB board by using the 'fujinet-v1-8mb' board config option that was recently added. The wiki on github has instructions for setting up pio/vscode to build fujinet firmware.


    Currently, we aren't using the entire 16MB and that's why the 8MB variant still works. In the future as we add features the 8MB may become 'obsolete'. There is effectively no price difference b/w the 8MB and 16MB module so we went all the way ;)


    • Thanks 1

  10. On 3/7/2021 at 10:50 PM, mytek said:

    Thanks for the idea, and I will try it tomorrow

    I just read your latest blog post. Too bad it didn't work.



    I really didn't like where the ADC VREF voltage divider was getting its positive voltage from (SIO), so I moved it over to the regulated 3.3V bus instead

    The voltage divider is not the ADC VREF, it's measuring the Atari 5V and should remain on SIO5V. It's currently only used for show in the web admin. It was previously used to determine if the Atari was on or not.

  11. 9 minutes ago, Lumpy01 said:

    [20:27:08]Setup complete @ 461 (457ms)
    [20:27:20]FujiNet 0.5.d4cb81e3 2021-02-24 05:19:32 Started @ 4

    Similar results. Sounds like a hardware problem to me. Contact your vendor about a replacement

  12. 48 minutes ago, Lumpy01 said:

    [19:11:22]BUTTON_B: LONG PRESS
    [19:11:22]ACTION: Reboot

    It appears that you get one command frame from the Atari and then the FujiNet resets and there's one more command frame at the end. 


    Try getting the logs with the following process:


    1. Plug Fujinet into Turned off Atari

    2. Plug microusb cable from PC into Fujinet

    3. Load FujiNet Flasher and press the serial debug button

    4. Short press the RESET button (far right button) on the Fujinet. You should see the startup messages from fujinet on the PC

    5. Turn on Atari

    6. When Atari is finished attempting to boot, copy and paste the debug output here



  13. 2 minutes ago, ijor said:

    I understand that the ESP32 is not 5V tolerant.

    Actually, the esp32 IS 5V tolerant for IO but must be powered by 3v3 (per the Espressif CEO). I tried to go the 'cheap' route with fujinet 1.0 and skip using a level shifter or buffers until we had problems with other devices on the bus.


    10 minutes ago, ijor said:

    I understand that ESP32 has a full GPIO matrix, so that you can drive any pin with the PWM compare timers, right?

    Not all the pins can be used for every function, but most are muxable. We already use PWM for CLOCKIN to set the MIDI baud rate for networked MIDIMaze. I checked and the pin we use for CLOCKOUT can be setup for PWM also.



  14. 18 hours ago, ijor said:

    Sure. I didn't know the FujiNet connects the SIO clock signals already. Nice!


    Make sure the SIO clock signals are connected to the MCU hardware compare/capture timer. You also need to be able to drive the SIO actively, not with open drain. I don't see any voltage level shifter in the schematics? Anyway, feel free to contact me by PM.


    18 hours ago, ijor said:

    I was looking at an old schematics. I see now that you are using 74LS07 buffers. That won't work. You need a bidirectional buffer for the clock signal, and you need buffers that are not permanently open drain.

    FujiNet 1.0 has no buffer or level shifter and only series resistors connecting SIO to the esp32. Both clock lines are connected to IO pins that can be used with the esp32 ADC. Is that sufficient to get this working on 1.0 hardware?

  15. 7 hours ago, mytek said:

    The SD problem is very baffling. I have thus far traced the connections between the socket and the ESP32, and then compared that to the schematic (mine, Mr Robot's, and mozzwalds) and all seems correct.

    Try adding pull up resistors (10k) on IO19/SPI_MISO, IO23/SPI_MOSI and IO5/SPI_CS. I have added these to the next FujiNet revision after reading in some espressif docs to 'not rely on the internal pull ups for SD card'. 

    • Thanks 1

  16. 12 minutes ago, Lumpy01 said:

    So, I picked up a FujiNet a few months ago and finally got the time to play around with it. I didn't get very far. It just doesn't boot.


    Things I've tried so far:


    Every conceivable button combination.

    The serial logs when booting while connected to the latest flash utility tell me nothing.

    I successfully flashed it to the latest firmware with no impact on booting.

    I tried it with a 5v USB power supply attached even though I read it's not necessary on the 130XE's.

    I get the same results on two different 130XE's.

    I tried it on a 1200XL (with the 5v SIO fix) with the same results.


    I know it's probably something simple. What the heck am I doing wrong? Video attached. Pay no attention to the portly shiny-headed reflection in the monitor.🙂

    The SIO LED is constantly lit which means something isn't right. Can you paste the serial debug log output of a boot attempt? It may tell you nothing, but it may tell us something ;) 

  • Create New...