Comments by gboost (Top 4 by date)

I very much appreciate the effort and material/info you've provided. Now I both have a working solution, and a good understanding of why it didn't work as expected; so I've learned a lot.

If I can accept both your very detailed/exhaustive answer and #4 (since it's a working example), I will do that.
Thank you :)

Well, this is a much more interesting reply than I imagined I'd get.
(Sorry if I don't reply to your satisfaction!)

I'd have to agree with you about it not making (much) sense.

I thought the problem would be something someone would spot just with those snippets as an 'oops, you did something wrong (languange-wise)', and didn't think more details were needed at the time. But I suppose it wouldn't have hurt to include them anyway -- and I probably should have.
Also hadn't considered the difference between the object types, which may be what the issue is altogether. (?)

To be a little more specific:
The strange character calculations really weren't my idea to start with. Another application is doing this to a configuration/settings file, I suppose to make it unreadable (e.g. in a text editor) to avoid issues with a user going in and changing things without being certain of what they're doing.
I'm actually just making a tool to tweak the settings (in a safe way), but need to be able to "decode"/"encode" the way said-app does -- which I was able to mimic in C, but obviously not C# (which I am trying to learn / adapt to)
Otherwise I probably wouldn't even want to do such a thing! :)

I posted an example of the text/characters to another answer.
I'm honestly not 100% sure if the encoding is 1252, but the results came out even worse unless I specified 1252 (which is what a text editor told me the encoding was, IF it is.)
Don't you love how certain .. I mean, confused, I am?

Anyway (again), an example is:

"normal"/decoded text (characters):
testing_one = "something"
testing_two = "100"

"encoded" text (by other app, or my C code):
‹šŒ‹–‘˜ ‘šßÂßÝŒ’š‹—–‘˜Ýß
‹šŒ‹–‘˜ ‹ˆßÂßÝÎÏÏÝ

attempt to "decode" the "encoded" text with C# code above
Æ?­Æìç#_oç? = "­oæ?Æëìç#"
Æ?­Æìç#_Æ9o = "100"
(only some characters are "decoded" correctly -- it should show the 'normal'/decoded text as above)

I'm not sure if that helps any or not.

I am going to try to analyze what you wrote better and check out the links though, I am interested in learning more and understanding what's actually going on here. I appreciate it.

Hmm, not sure how to take that last statement about prose. In a way it could be insulting, I suppose. But maybe not! Because? (Kidding...)

This seems to work fine, except it's still not converting it as I expected. I wonder if it does have something to do with what lewax said (8 bit vs 16 bit)

For example, this "encoded" text:
‹šŒ‹–‘˜ ‘šßÂßÝŒ’š‹—–‘˜Ýß
‹šŒ‹–‘˜ ‹ˆßÂßÝÎÏÏÝ

Comes out as this (in C# -- looks like some of it converted ok):
Æ?­Æìç#_oç? = "­oæ?Æëìç#"
Æ?­Æìç#_Æ9o = "100"

But it's supposed to be (and comes out as this with C):
testing_one = "something"
testing_two = "100"