diff --git a/legendary/core.py b/legendary/core.py
index 85b0108..1d525bc 100644
--- a/legendary/core.py
+++ b/legendary/core.py
@@ -1094,7 +1094,6 @@ class LegendaryCore:
             delta = max(0, analysis.install_size - current_size)
             min_disk_space = delta + analysis.biggest_file_size
 
-        # todo when resuming, only check remaining files
         _, _, free = shutil.disk_usage(os.path.split(install.install_path)[0])
         if free < min_disk_space:
             free_mib = free / 1024 / 1024
diff --git a/legendary/lfs/lgndry.py b/legendary/lfs/lgndry.py
index b307117..e20fc5b 100644
--- a/legendary/lfs/lgndry.py
+++ b/legendary/lfs/lgndry.py
@@ -103,7 +103,8 @@ class LGDLFS:
 
         try:
             self._installed = json.load(open(os.path.join(self.path, 'installed.json')))
-        except Exception as e:  # todo do not do this
+        except Exception as e:
+            self.log.debug(f'Loading installed games failed: {e!r}')
             self._installed = None
 
         # load existing app metadata
diff --git a/legendary/models/chunk.py b/legendary/models/chunk.py
index 510083d..c32a2b5 100644
--- a/legendary/models/chunk.py
+++ b/legendary/models/chunk.py
@@ -10,7 +10,6 @@ from uuid import uuid4
 from legendary.utils.rolling_hash import get_hash
 
 
-# ToDo do some reworking to make this more memory efficient
 class Chunk:
     header_magic = 0xB1FE3AA2
 
diff --git a/legendary/models/exceptions.py b/legendary/models/exceptions.py
index 9f26931..7602b49 100644
--- a/legendary/models/exceptions.py
+++ b/legendary/models/exceptions.py
@@ -1,12 +1,4 @@
 # coding: utf-8
 
-# ToDo more custom exceptions where it makes sense
-
-
-class CaptchaError(Exception):
-    """Raised by core if direct login fails"""
-    pass
-
-
 class InvalidCredentialsError(Exception):
     pass
diff --git a/legendary/models/manifest.py b/legendary/models/manifest.py
index c66b9cb..ad3d54a 100644
--- a/legendary/models/manifest.py
+++ b/legendary/models/manifest.py
@@ -16,8 +16,8 @@ def read_fstring(bio):
 
     # if the length is negative the string is UTF-16 encoded, this was a pain to figure out.
     if length < 0:
-        # utf-16 chars are 2 bytes wide but the length is # of characters, not bytes
-        # todo actually make sure utf-16 characters can't be longer than 2 bytes
+        # utf-16 chars are (generally) 2 bytes wide, but the length is # of characters, not bytes.
+        # 4-byte wide chars exist, but best I can tell Epic's (de)serializer doesn't support those.
         length *= -2
         s = bio.read(length - 2).decode('utf-16')
         bio.seek(2, 1)  # utf-16 strings have two byte null terminators