I think the title is probably not the best choice. Otherwise the article is needlessly informative. while that may seem like a compliment, it really isn't. Basically, you don't really describe any rationale behind why a gain setting is any better then one that is higher or lower.
Here is my addition:
From a practical standpoint, users tend toward two seperate goals -- SPL vs non-SPL, and Objectiveness vs Subjectiveness.
I call a user who tries to get a high SPL score, something that can be objectively measured and is focused on SPL, a SPL competitor. For this user, the gain setting is all about what damages his equipment in a short period of time.
The user who loves loud music I call "Street". This is the target audience for car audio -- the person who love loud and somewhat clean music with a preferance to LOUD. This user may readily accept an excessive gain setting to benefit from the compression that comes along with clipping (eg, loud notes are only slightly louder while softer notes are much louder, and thus the music on average is LOUD!). This user will have some standard of what is excessive distortion. Typically this user will want to prevent clipping because it will prevent equipment damage.
The next user I call "Hi-Fi". This user wants objectively measureably perfect sound. The RTA is flat, there is no clipping, and everything is replayed exactly as recorded. This user doesn't care about SPL as much, as that would stress the equipment to the limit and introduce distortion. They also know how loud they want the music to be and don't increase it further.
The final user is "SQ". This guy wants the music to sound good. He isn't looking to play his system loud all the time. But he does want it to sound as good as possible to himself. This user might look at EQs, DSP, Tube-Amps, fancy cables and the lot to make things sound better then recorded.
Each of these guys may approach gains differently. The SPL competitor needs to know how much is too much, but can allow clipping (which does increase power in the core test tone as well) so long as it doesn't damage the equipment. His choice is really a risk v reward issue.
The HiFi guy will want to match the gains from channel to channel, ensure there is no clipping, AND most importantly, ensure that speakers blend correctly. This user NEEDS some way to set the gain on each channel. This means either setting the gains on the amps to something safe then using a high-end HU/Processor to adjust down anything that is too loud, or setting the gains lower on some channels to make sure things blend.
The street guy, which is the intended audience for car audio, is really the hardest guy to work with. He has to figure out what is acceptable and how much risk he wants to take. Objective method like the "DMM" method might give him lower gains then he really wants. But at the same time, without and objectivity, he might set the amp up to kill his speakers. He has the hardest choices -- how to get decent sound, high SPL, and not break things?
The SQ guy has it a bit easier. Because he's not looking to stress his equipment to the max, just looking for good sound, he can set gain pretty much by ear. (after all, that is how he judges success).
My point here is that there are fundamentally different goals, and fundamentally different ways to determine "success". Furthermore, many user will/can treat gains differently if they have some other way to adjust the volume of each channel.
And the street users tend to move towards whatever they want, even without a knowledge of how things work. How many time have you heard this: "My BRANDX amp is great, how else could I get so loud at ONLY volume 5 of 30?!" these are the people who have set the gain to max. This is fine though, so long as they keep the volume settings down.
