I started looking at this. I get the identical memory issue when the program runs automatically as code.py. If I import code.py in the REPL it runs fine.

otter_data.py has a hideously large set of nested lists of small integers and could be made a lot more efficient.

Here is an instrumented view of memory consumption:

import board
import gc
print("before most imports", gc.mem_free())
import displayio
print("a", gc.mem_free())
from time import sleep
import otter_data
print("b", gc.mem_free())
display = board.DISPLAY
group = displayio.Group()
print("c", gc.mem_free())
otter_bitmap = displayio.Bitmap(384,96,18)   # <---- fails here
print("d", gc.mem_free())
background_file = open("/images/background.bmp", "rb")
background_bitmap = displayio.OnDiskBitmap(background_file)
background_tile = displayio.TileGrid(background_bitmap, pixel_shader=displayio.ColorConverter())
#...

Automatic run:

Auto-reload is on. Simply save files over USB to run them or enter REPL to disable.
code.py output:
before most imports 220080
a 220064
b 32000
c 31904
Traceback (most recent call last):
  File "code.py", line 13, in <module>
MemoryError: memory allocation failed, allocating 36864 bytes

Manual import of code.py:

Adafruit CircuitPython 5.0.0-beta.5 on 2020-02-05; Adafruit PyPortal with samd51j20
>>> import code
before most imports 219744
a 219728
b 70336
c 70240
d 33328

Description

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions