I see that TwoPynts gave us an additional way to look at,

RAW -vs- JPEG

I looked at that page and have some questions (observations) ...

1. they say ...

at least 8 bits per color - red, green, and blue (12-bits per X,Y location),
though most DSLRs record 12-bit color (36-bits per location).

This must be an example of 'new math'.  I thought that 8 bits per color
... and 3 colors (RGB) would be 24 bits ... 3 X 8 = 24 ... right? 

2. they say ...

In comparison a JPEG is…

exactly 8-bits per color (12-bits per location).

Once again, I'm having a little trouble with the new math. 

3. They say ...

The major actor in this case is the Discrete Cosine Transforamtion
(or DCT) which divides the image into blocks (usually 8×8 pixels) and
determines what can be “safely” thrown away because it is less
perceivable (the higher the compression ration/lower quality JPEG,
the more is thrown away during this step).

This statement is partially correct ...

Actually, the image is divided into a serries of 8X8 blocks and the DCT
is applied to each block.  IOW, the authors have confused effect and
cause ... errrrrrrr ... Oh, I meant cause and effect ... well, you get the
idea -- even if they didn't get it exactly right. 

The Wikipedia JPEG article is much more authorative ... and is what we
should use as a baseline for discussions about JPEGs or JPEG v. Raw. 

BTW, if you don't like the complexity of the Wikopedia JPEG article,
I suggest that you consider this John von Neumann quote ...

**If people do not believe that mathematics is simple, it is
only because they do not realize how complicated life is.
--John von Neumann
**
Quotation source ...
"John von Neumann and von Neumann Architecture for Computers (1945)"

--
Martin