UIFlow doesn't start on new Core2

puzznic

hello,
my new Core2 has just been delivered and I believe my Win10 install is working fine (I can successfully burn and run other downloadable apps such as draw_by_touch or Core2FactoryTest) but burning UIFlow_Core2 v1.7.1-core2 (or any other previous version actually) doesn't work :

the device turns on, screen turns on too but only displays an empty black screen (not the expected UIFlow logo or anything)

(please find below the M5Burner log that reports it was successfully burnt -- please see log below)

could you think of something critical I am missing here?
many thanks,

=================
Start Burn
=================
--chip esp32 --port COM5 --baud 1500000 --before default_reset write_flash -z --flash_mode dio --flash_freq 80m --flash_size detect 0x7ff000 C:\Users\vpennel\Downloads\M5Burner\packages\fw\other\wifi.bin 0x1000 C:\Users\vpennel\Downloads\M5Burner\packages\fw\core\UIFlow_Core2-v1.7.1-core2.bin
=================
esptool.py v2.5.0
Serial port COM5
Connecting...
.
Chip is ESP32D0WDQ6 (revision 1)
Features: WiFi, BT, Dual Core, 240MHz, VRef calibration in efuse
MAC: b8:f0:09:c5:b1:34
Uploading stub...
Running stub...
Stub running...
Changing baud rate to 1500000
Changed.
Configuring flash size...
Auto-detected Flash size: 16MB
Compressed 100 bytes to 24...
Writing at 0x007ff000... (100 %)
Wrote 100 bytes (24 compressed) at 0x007ff000 in 0.0 seconds (effective 230.2 kbit/s)...
Hash of data verified.
Compressed 8376320 bytes to 3457107...
Writing at 0x00001000... (0 %)
[...]
Writing at 0x00349000... (99 %)
Writing at 0x0034d000... (100 %)
Wrote 8376320 bytes (3457107 compressed) at 0x00001000 in 55.3 seconds (effective 1210.9 kbit/s)...
Hash of data verified.
Leaving...
Hard resetting via RTS pin...
=================
Burn Successfully
=================

felmue

Hello @puzznic

hmm, that is strange. I've just verified with my M5Core2 and after burning UIFlow_Core2 v1.7.1-core2 it restarted with UIFlow just fine.

Have you tried to re-download the UIFlow firmware? (Although I don't think that is the issue as my burning shows exactly the same amount of data written as yours, and the hash is verified as well.)

Maybe try to erase your M5Core2 from within M5Burner first the burn again?

Thanks
Felix

puzznic

thank you for your reply Felix.
I indeed erased and burnt UIFlow mutlitpl times from different Windows laptops with the same (bad) results :(
could it be a faulty unit then? even if the FactoryTest reports total success?

=================
Erase now
=================
esptool.py v2.5.0
Serial port COM5
Connecting....
Detecting chip type... ESP32
Chip is ESP32D0WDQ6 (revision 1)
Features: WiFi, BT, Dual Core, 240MHz, VRef calibration in efuse
MAC: b8:f0:09:c5:b1:34
Uploading stub...
Running stub...
Stub running...
Erasing flash (this may take a while)...
Chip erase completed successfully in 35.7s
Hard resetting via RTS pin...
=================
Erase Successfully
=================~~~~

felmue

Hello @puzznic

just to be clear, when you run the Factory Test the screen in your M5Core2 works, correct?

After burning UIFlow, have you tried to reset your M5Stack with the reset button?

Also have you tried to fully shutdown your M5Stack by pressing and holding the power button, then power it on again?

Thanks
Felix

puzznic

hello Felix,

just to be clear, when you run the Factory Test the screen in your M5Core2 works, correct? correct. here's a screenshot of the test running:

![alt text]( 0_1609747947765_IMG-2857 (Small).jpg image url)

After burning UIFlow, have you tried to reset your M5Stack with the reset button?
Also have you tried to fully shutdown your M5Stack by pressing and holding the power button, then power it on again?

the device has been reset and turned off and on multiple times indeed.
so weir, I reckon...

guillaumelf26

Hello @puzznic,

I just received my first Core2 yesterday, and I had the same problem trying to run the uiFlow 1.7.1 firmware.

After many attempts, I finally tried the 1.7.0 version, and it worked ! Maybe try this version instead of the latest one.

felmue

Hello @puzznic

after burning UIFlow, what do you see in the COM Monitor log of M5Burner when you reset M5Core2? Any errors?

Thanks
Felix

NetworkUke

I'm running into the same issue. Right after

"Hard resetting via RTS pin...

Burn Successfully"

The m5stack display freezes and in the console output is:

rst:0x10 (RTCWDT_RTC_RESET),boot:0x17 (SPI_FAST_FLASH_BOOT)
invalid header: 0x4f44213c
invalid header: 0x4f44213c
invalid header: 0x4f44213c
invalid header: 0x4f44213c
invalid header: 0x4f44213c
invalid header: 0x4f44213c
invalid header: 0x4f44213c
invalid header: 0x4f44213c
ets Jul 29 2019 12:21:46

on an infinite loop.

Even after re-burning it with the easy loader when I attempt to burn UIFlow_Core2 the same issue occurs. Any advice is greatly appreciated.

felmue

Hi guys

@NetworkUke : please see your other post. In your case the firmware file size is way to small.

@puzznic : have you tried to reduce the baudrate? Not sure that will help but worth a try.

Thanks
Felix

inasawa

It seems to be the same situation as the Core 2 I bought yesterday.
Core2FactoryTest and draw_by_touch work fine, but UIFlow_Core2 v1.7.1.1-core2 (I also tried v1.7.0 / v1.7.1) doesn't work.

However, it seems running with black screen. I tried connect to Core2 from UIFlow Cloud and run the program. print() can print text to the serial console, but M5Label will show nothing on the Core2 screen. just black screen.

I tried burn with low baudrate (115200), but nothing changed

inasawa

Since UIFlow display nothing on the screen, I tried the test with the ILI9341 library for MicroPython I found.
https://github.com/tuupola/micropython-ili934x/blob/master/ili934x.py

I copied the above library to Core2 and ran the following code from the serial console.

import ili934x
from machine import Pin, SPI
spi = SPI(miso=Pin(38), mosi=Pin(23, Pin.OUT), sck=Pin(18, Pin.OUT))
display = ili934x.ILI9341(spi, cs=Pin(5), dc=Pin(15), rst=Pin(4))
display.fill(ili934x.color565(0xff, 0x11, 0x22))
display.pixel(120, 160, 0)
display.print('1234567890 ABCDEFGHIJKLMNOPQRSTUVWXYZ')
     :     :     :
display.print('1234567890 ABCDEFGHIJKLMNOPQRSTUVWXYZ')

As a result, the characters are displayed as shown in the image below.
I'm not sure what this means, but can it help analyze the problem?

OffLineR

I had the same problem.
Core2 did not turn on after 1.7.1.1
Currently I am using 1.7.0 instead.

inasawa

I don't know what happened. Core2 that has not been displayed with UIFlow for 2 days since I bought it, suddenly, yes suddenly, when I reset Core2, something was displayed, and every time I reset it, the display was distorted or displayed upside down. After several resets, the display is now showing normally.

I feel that there is an unstable part in the display-related HW. Firmware is UIFlow_Core2-v1.7.1.1-core2.bin.

inasawa

It seems that it was a short-lived joy. Revert to a black screen, distorted screen, or an upside-down screen.
I tried v1.7.1.1, v1.7.1, v1.7.0, v1.6.6 and they are all the same and have a black screen.
It's HW problem?

Also, even in this situation with UIFlow is running, if I burn factory test firmware, it will work normally.

inasawa

After struggling for a while, I had this Core2 replaced at the shop where I bought it as an initial defect.
I'm not sure, but it seems that the display mechanism in Arduino and the display mechanism in UIFlow are different, and even if Factory Test works normally, it may not work properly in UIFlow.