Upload & Sell: Off
Brian makes an excellent point. In most lighting scenarios (flat ring lighting being an exception) a dominant off camera axis Key source is used to create the 3D highlight clues over a foundation of Fill which allows the sensor range to record normal looking detail in the shadows.
While not necessarily the workflow normally used by most this scenario explains the cause and effect of ratios and expose:
1) Dark Room (no light), Camera set to f/8 (to keep entire head in focus)
2) Turn on first light as FILL. Move it around the camera until minimal shadows are seen. If fill source is creating shadows will there be Fill there? No. There will be dark distracting shadows in placs like smile lines, inside mouth and teeth, base of nostrils which are unflattering. Since a goal of the exercise of portraiture is flattering the subject the location for the Fill which best meets that goal is just above the lens of the camera where no shadows are noticed on the face.
3) Now that where to put the fill has been decided based on the goal how big should it be? The size to be concerned with here isn't the size of the modifier but rather the size of the modifers light footprint For example I use a Buff 22" dish most of the time over the camera lens because from 8-9' where the camera is the footprint is so wide it spills off the ceiling and walls.
It is instructive to think of light as vectors... straight line arrows indicating direction and strength (varying length or width). By using the dish for my Fill and spilling it off the walls and ceiling I get strong vectors along the camera axis directly, and weaker vectors bouncing off the room surfaces hitting the subject from many angles "wrapping" around and making the shadows there lighter than they would be in a larger space of if the walls were black. Understanding that cause and effect I have black King sheets I hang like drapes on the walls when I don't wan't that lighter shadow, wrapped fill effect. Or alternately I switch fill modifier and use one that has a footprint big enough to meet the goal of the shot. When shooting portraits on dark backgrounds I usually want the edges of the frame darker (a vignette) so I'll use a fill modifier that creates a "spotlight" effect with it's small footprint. Mind you I'm still discussing the flat fill foundation, not creating the 3D modeling.
4) The next decision is Fill intensity. That's answer by the camera playback and a black/white target on a stand where the face will be (the actual subject isn't there yet. I take a shot and look at the results. Is there detail in the black target? Is it too gray looking? I just adjust the fill intensity until the shadow target looks just "right" how it looks by eye. Then since I have an incident meter I put it in front of the target and take a reading ... f/5.6 a stop under the aperture setting. I note that for future reference. Next time I'll first meter the fill to 5.6 but still check how it works on my target
5) Fill set for shadow detail per the target it's now time to set the Key light postion. Don't think in terms of where it is on the floor, think about where it needs to be relative to the nose and eyes of the subject . Again remember the goal is flattering the subject. I do that by putting my key light where it reaches past the brow into the eyes (about 45 degrees to the front plane of the face) and where the shadow it casts on the nose falls down at a 45 degree angle (because the light is above the nose) and neatly over 1/2 of it without hanging out an distracting. That happens when the Key light is about 45 degrees to the side of the nose. For full face poses I'll often put the key light directly over the camera at a 45 degree angle because then there are no sideways shadows to provide clues about the 3D shape of the nose (a good way to hide the fact its humongous). So for most portraits I only use two key light positions: 45V/45H to the nose for oblique, profile and full face views, and 45V/0H (centered) for full face views of naturally symmetrical faces. Using centered strategy with a lopsided face the natural asymmetry obvious so I'd use the 45/45 strategy for those in a full face pose with the Key light on the smaller side of the face to make the face seem more even in the photo (it's an optical illusion but it works). Or I'll put the key light at 45V/45V to the nose and move the camera so it sees the face obliquely or from profile. Other patterns iike Rembrandt and Loop you may have seen are just variations on those to basic configurations.
For Rembrandt the Key is move further to the side of the nose (50-60 degree) which puts the far side more in shadow creating a small triangle of light in the far eye. The vertical height is also sometimes adjusted to intentionally partially shade the eyes with the brow to create a wise / thoughtful or sad/angry impression. If trying to create those impressions with the lighting less Fill creating darker shadows is needed so you'd start with less fill in the step above and wind up with a meter reading between f/5.6 and f/4 when it looks "right"; which for Rembrandt is like a grumpy old painter (matching his famous self-portrait). Loop is created from the baseline of the 45V/0V pattern by moving the key lightly to the side and often higher if the subject doesn't have a prominent brow: 60V/5H relative to the nose. The overall modeling of the face is similar to 45V/0V it just creates that hanging loop shaded shadow clue. I find it distracting so I never use a Loop pattern. Others find it attractive. Learn to do both and you can decide for yourself. As you see it's not rocket science just simple geometry of matching key light angle to 3D shape of the face (which varies) to find the most flattering (or grumpy looking) key light vector for that face.
6) Key light modifer size: What Key light modifier size controls are the vectors. A small modifier sends all the light on the same vector which is why it will create distinct shadows. As the source gets larger the vectors from the edges go in different directions than those is the center and wind up being different strengths by the time they hit the face. What happens with a speedlight when bounced vs. used straight is that more different vectors are created in more directions and strength .
The "best" modifier for the key light varies with the goal for the look of the shot and with the fill strategy being used The reason I started with fill here is so you'll realize that last part. When I use my dish for fill and bounce it off the walls that helps to make the gradients the key light vectors create more gradual, which is perceived as rounder vs. angular cheek bones. Because I create "wrap" with my Fill source I don't need or want a huge Key light. The Max Wrap Fill [tm] / Smaller Key is necessary because I don't have ceilings high enough to use my Large Photoflex Multidome as Key. If I chose to shoot to make money I'd want a larger studio with 20' ceilings so I could use big modifiers. I'd need them too because I wouldn't be getting that Max Wrap Fill [tm] effect off the walls.
The take away for you here is there are a lot of variables affecting results you now don't have a clue about. The best way to get a clue is start with small footprint modifiers which aren't bouncing all over the room. You can create bounced "Max Wrap Fill" just by moving them back (what I do with my 22" dish for fill) or bouncing your fill off the wall behind the camera (which flood the room with similar strength fill vectors from many directions. Then gradually buy bigger ones as you find you need them, like shooting on location in a bigger room. Outdoors you have skylight for fill if you learn to use it effectively and really don't need huge modifiers either.
6) The last part of the lighting scenario is key light exposure. That's not rocket science either and doesn't require a meter. After selecting Key position (relative to face) and it's modifier footprint, just adjust power until the white target I just under clipping in the playback at the f/8 camera setting selected in step 1. Depending on the footprint / spill of the Key modifier it may make the shadows lighter than seen when Fill was set visually. With experience with you lights you'll be able to anticipate that just like learning to tie your shoes or ride a bike.
The "magic" of setting fill and key so the shadows and highlights "look right" visually is that all the tones in the middle will "look right" also automatically because the camera and file format encoding / decoding are designed to match the response of human vision with the goal of capturing an image that looks "normal".
What you will be doing by starting with more fill or less fill from that fits sensor range perfectly baseline outlined above is changing lighting ratio and mood to create some other than "average" look: lighter/upbeat/happy or darker/downbeat/sad. It's not necessary to follow the steps in that order (that was for edification) you just need to wind up with details in the darkest shadows and in the highlights at the same time by controlling fill for the shadows and then adjusting key light so it's not clipping the highlights.
Meters? Back in the days of film a photographer would know a 2:1 created lighter than average, 3:1 average, 4:1 darker than average and have that goal in mind before metering the lighting ratio to create the desired mood in the lighting. Those ratios are about the same with a digital camera sensor today so if using a meter you would want to start with the same goal about desired mood, then set the ratio numerically. But thanks to the playback of digital its possible to skip the metering step, or if metering confirm that the numerical ratio does in fact meet the goals.
Circling back to topic of modifiers, if your goal is light happy lighting on a white background then light spilling off the walls is desirable. But if trying to make grandpa look like Rembrandt with dark moody shadows you'll need to eliminate all that "Max Wrap Fill" coming from your Fill and Key lights by using modifiers with smaller, more controlled footprints and/or drapes on the walls to increase (white draping) or decrease (black draping) the spill.
1) Start with the goal - happy / average / moody
2) Pick the strategy which meets that goal - facial angle - key light vectors (parallel or diffuse) - fill intensity/vectors (wrap)
3) Select the tools and technique (modifier footprint and placement) which best accomplishes the strategy and goal.
The goals are not always the same which is why you need a variety of modifiers to meet them.